{
  "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": 0.04857712,
      "auxiliary_loss_mlp": 0.0199187,
      "balance_loss_clip": 2.39669204,
      "balance_loss_mlp": 1.67505896,
      "epoch": 0.00012024289063909097,
      "flos": 24932483919360.0,
      "grad_norm": 40.11934686879225,
      "language_loss": 2.58061337,
      "learning_rate": 0.0,
      "loss": 1.90124202,
      "num_input_tokens_seen": 20375,
      "step": 1,
      "time_per_iteration": 12.615964412689209
    },
    {
      "auxiliary_loss_clip": 0.03217209,
      "auxiliary_loss_mlp": 0.01379494,
      "balance_loss_clip": 1.59208441,
      "balance_loss_mlp": 1.16548967,
      "epoch": 0.00024048578127818193,
      "flos": 30664624377600.0,
      "grad_norm": 54.660974841893434,
      "language_loss": 1.89303315,
      "learning_rate": 5.021476677069823e-07,
      "loss": 1.93900037,
      "num_input_tokens_seen": 39035,
      "step": 2,
      "time_per_iteration": 2.5258588790893555
    },
    {
      "auxiliary_loss_clip": 0.03213671,
      "auxiliary_loss_mlp": 0.01349028,
      "balance_loss_clip": 1.60020781,
      "balance_loss_mlp": 1.14379764,
      "epoch": 0.0003607286719172729,
      "flos": 19026227969280.0,
      "grad_norm": 40.482147465856336,
      "language_loss": 1.61610591,
      "learning_rate": 7.958852231401551e-07,
      "loss": 1.66173291,
      "num_input_tokens_seen": 57600,
      "step": 3,
      "time_per_iteration": 2.3406474590301514
    },
    {
      "auxiliary_loss_clip": 0.03232699,
      "auxiliary_loss_mlp": 0.01318564,
      "balance_loss_clip": 1.59900165,
      "balance_loss_mlp": 1.10885084,
      "epoch": 0.00048097156255636386,
      "flos": 19316314206720.0,
      "grad_norm": 36.579217578925636,
      "language_loss": 1.64638352,
      "learning_rate": 1.0042953354139647e-06,
      "loss": 1.6918962,
      "num_input_tokens_seen": 76465,
      "step": 4,
      "time_per_iteration": 2.456038475036621
    },
    {
      "auxiliary_loss_clip": 0.03190038,
      "auxiliary_loss_mlp": 0.01345037,
      "balance_loss_clip": 1.58716583,
      "balance_loss_mlp": 1.13751745,
      "epoch": 0.0006012144531954548,
      "flos": 13991264893440.0,
      "grad_norm": 54.9508034791524,
      "language_loss": 1.93431103,
      "learning_rate": 1.1659507774310057e-06,
      "loss": 1.97966182,
      "num_input_tokens_seen": 94350,
      "step": 5,
      "time_per_iteration": 2.6019227504730225
    },
    {
      "auxiliary_loss_clip": 0.03238536,
      "auxiliary_loss_mlp": 0.01345963,
      "balance_loss_clip": 1.60101831,
      "balance_loss_mlp": 1.14244902,
      "epoch": 0.0007214573438345458,
      "flos": 23148988225920.0,
      "grad_norm": 45.126698464283194,
      "language_loss": 1.60975599,
      "learning_rate": 1.2980328908471373e-06,
      "loss": 1.65560091,
      "num_input_tokens_seen": 114595,
      "step": 6,
      "time_per_iteration": 2.7047126293182373
    },
    {
      "auxiliary_loss_clip": 0.03341887,
      "auxiliary_loss_mlp": 0.01596753,
      "balance_loss_clip": 1.70233011,
      "balance_loss_mlp": 1.43958735,
      "epoch": 0.0008417002344736367,
      "flos": 67663246170240.0,
      "grad_norm": 4.610361848558148,
      "language_loss": 0.81500804,
      "learning_rate": 1.4097067265369432e-06,
      "loss": 0.86439443,
      "num_input_tokens_seen": 179590,
      "step": 7,
      "time_per_iteration": 3.134592056274414
    },
    {
      "auxiliary_loss_clip": 0.03219778,
      "auxiliary_loss_mlp": 0.01327512,
      "balance_loss_clip": 1.59819603,
      "balance_loss_mlp": 1.11570144,
      "epoch": 0.0009619431251127277,
      "flos": 21281381504640.0,
      "grad_norm": 41.439896748159214,
      "language_loss": 1.58792031,
      "learning_rate": 1.506443003120947e-06,
      "loss": 1.63339305,
      "num_input_tokens_seen": 195090,
      "step": 8,
      "time_per_iteration": 2.6757330894470215
    },
    {
      "auxiliary_loss_clip": 0.03175941,
      "auxiliary_loss_mlp": 0.01310091,
      "balance_loss_clip": 1.58784723,
      "balance_loss_mlp": 1.10686302,
      "epoch": 0.0010821860157518186,
      "flos": 23331342597120.0,
      "grad_norm": 18.15604455892974,
      "language_loss": 1.47693348,
      "learning_rate": 1.5917704462803102e-06,
      "loss": 1.52179372,
      "num_input_tokens_seen": 211635,
      "step": 9,
      "time_per_iteration": 2.6493964195251465
    },
    {
      "auxiliary_loss_clip": 0.03206275,
      "auxiliary_loss_mlp": 0.01350741,
      "balance_loss_clip": 1.59173989,
      "balance_loss_mlp": 1.14407992,
      "epoch": 0.0012024289063909096,
      "flos": 17010166337280.0,
      "grad_norm": 13.062396004046407,
      "language_loss": 1.52989841,
      "learning_rate": 1.6680984451379884e-06,
      "loss": 1.57546866,
      "num_input_tokens_seen": 224705,
      "step": 10,
      "time_per_iteration": 2.645115613937378
    },
    {
      "auxiliary_loss_clip": 0.03193913,
      "auxiliary_loss_mlp": 0.01362597,
      "balance_loss_clip": 1.59084797,
      "balance_loss_mlp": 1.14592206,
      "epoch": 0.0013226717970300007,
      "flos": 21288133261440.0,
      "grad_norm": 13.95555698780662,
      "language_loss": 1.32547235,
      "learning_rate": 1.7371455188905097e-06,
      "loss": 1.37103748,
      "num_input_tokens_seen": 244635,
      "step": 11,
      "time_per_iteration": 2.798292398452759
    },
    {
      "auxiliary_loss_clip": 0.03209931,
      "auxiliary_loss_mlp": 0.01358659,
      "balance_loss_clip": 1.59634089,
      "balance_loss_mlp": 1.14618027,
      "epoch": 0.0014429146876690916,
      "flos": 27237884935680.0,
      "grad_norm": 10.668008037061085,
      "language_loss": 1.25325942,
      "learning_rate": 1.8001805585541196e-06,
      "loss": 1.29894519,
      "num_input_tokens_seen": 265765,
      "step": 12,
      "time_per_iteration": 2.798614501953125
    },
    {
      "auxiliary_loss_clip": 0.03196177,
      "auxiliary_loss_mlp": 0.01363921,
      "balance_loss_clip": 1.59832883,
      "balance_loss_mlp": 1.15449381,
      "epoch": 0.0015631575783081825,
      "flos": 19062174504960.0,
      "grad_norm": 6.740215697457617,
      "language_loss": 1.29070807,
      "learning_rate": 1.8581671739548328e-06,
      "loss": 1.33630896,
      "num_input_tokens_seen": 283500,
      "step": 13,
      "time_per_iteration": 2.7212860584259033
    },
    {
      "auxiliary_loss_clip": 0.03197996,
      "auxiliary_loss_mlp": 0.01335406,
      "balance_loss_clip": 1.59409261,
      "balance_loss_mlp": 1.1274097,
      "epoch": 0.0016834004689472734,
      "flos": 48139473985920.0,
      "grad_norm": 6.121092676105596,
      "language_loss": 1.13598537,
      "learning_rate": 1.9118543942439254e-06,
      "loss": 1.18131948,
      "num_input_tokens_seen": 305685,
      "step": 14,
      "time_per_iteration": 4.335437774658203
    },
    {
      "auxiliary_loss_clip": 0.03127148,
      "auxiliary_loss_mlp": 0.0134348,
      "balance_loss_clip": 1.58910632,
      "balance_loss_mlp": 1.13309956,
      "epoch": 0.0018036433595863645,
      "flos": 34970026314240.0,
      "grad_norm": 5.904381974006851,
      "language_loss": 1.12619162,
      "learning_rate": 1.961836000571161e-06,
      "loss": 1.17089796,
      "num_input_tokens_seen": 327340,
      "step": 15,
      "time_per_iteration": 3.5974414348602295
    },
    {
      "auxiliary_loss_clip": 0.03198165,
      "auxiliary_loss_mlp": 0.0146863,
      "balance_loss_clip": 1.68115842,
      "balance_loss_mlp": 1.32062018,
      "epoch": 0.0019238862502254555,
      "flos": 59768284440960.0,
      "grad_norm": 3.8014902663919754,
      "language_loss": 0.64689642,
      "learning_rate": 2.0085906708279293e-06,
      "loss": 0.69356436,
      "num_input_tokens_seen": 382710,
      "step": 16,
      "time_per_iteration": 3.1316792964935303
    },
    {
      "auxiliary_loss_clip": 0.03131644,
      "auxiliary_loss_mlp": 0.01352452,
      "balance_loss_clip": 1.58373582,
      "balance_loss_mlp": 1.14788878,
      "epoch": 0.0020441291408645466,
      "flos": 20814543417600.0,
      "grad_norm": 4.8916894515364735,
      "language_loss": 1.15955639,
      "learning_rate": 2.0525099325728135e-06,
      "loss": 1.20439732,
      "num_input_tokens_seen": 400890,
      "step": 17,
      "time_per_iteration": 2.6377735137939453
    },
    {
      "auxiliary_loss_clip": 0.03155775,
      "auxiliary_loss_mlp": 0.01429185,
      "balance_loss_clip": 1.67447209,
      "balance_loss_mlp": 1.28422666,
      "epoch": 0.0021643720315036373,
      "flos": 63857001582720.0,
      "grad_norm": 3.575649053011835,
      "language_loss": 0.72171211,
      "learning_rate": 2.0939181139872922e-06,
      "loss": 0.76756167,
      "num_input_tokens_seen": 462605,
      "step": 18,
      "time_per_iteration": 3.095980167388916
    },
    {
      "auxiliary_loss_clip": 0.03136447,
      "auxiliary_loss_mlp": 0.01294968,
      "balance_loss_clip": 1.58663404,
      "balance_loss_mlp": 1.0953635,
      "epoch": 0.0022846149221427284,
      "flos": 31284981192960.0,
      "grad_norm": 4.756459959382579,
      "language_loss": 1.01624095,
      "learning_rate": 2.1330868934640175e-06,
      "loss": 1.0605551,
      "num_input_tokens_seen": 483280,
      "step": 19,
      "time_per_iteration": 2.7125473022460938
    },
    {
      "auxiliary_loss_clip": 0.03092702,
      "auxiliary_loss_mlp": 0.01394372,
      "balance_loss_clip": 1.66258645,
      "balance_loss_mlp": 1.25399137,
      "epoch": 0.002404857812781819,
      "flos": 51083648161920.0,
      "grad_norm": 3.5580436290505593,
      "language_loss": 0.76393998,
      "learning_rate": 2.170246112844971e-06,
      "loss": 0.80881071,
      "num_input_tokens_seen": 537620,
      "step": 20,
      "time_per_iteration": 2.8881995677948
    },
    {
      "auxiliary_loss_clip": 0.03062899,
      "auxiliary_loss_mlp": 0.0132019,
      "balance_loss_clip": 1.56706142,
      "balance_loss_mlp": 1.11658025,
      "epoch": 0.0025251007034209102,
      "flos": 15815347309440.0,
      "grad_norm": 4.37472530942615,
      "language_loss": 1.01542091,
      "learning_rate": 2.2055919496770983e-06,
      "loss": 1.05925179,
      "num_input_tokens_seen": 555760,
      "step": 21,
      "time_per_iteration": 2.6456518173217773
    },
    {
      "auxiliary_loss_clip": 0.03024953,
      "auxiliary_loss_mlp": 0.01337104,
      "balance_loss_clip": 1.56464064,
      "balance_loss_mlp": 1.13454401,
      "epoch": 0.0026453435940600014,
      "flos": 37851857458560.0,
      "grad_norm": 4.159550303111269,
      "language_loss": 0.89440864,
      "learning_rate": 2.2392931865974923e-06,
      "loss": 0.93802917,
      "num_input_tokens_seen": 578450,
      "step": 22,
      "time_per_iteration": 2.7724180221557617
    },
    {
      "auxiliary_loss_clip": 0.03026118,
      "auxiliary_loss_mlp": 0.01303982,
      "balance_loss_clip": 1.56200361,
      "balance_loss_mlp": 1.10208929,
      "epoch": 0.002765586484699092,
      "flos": 21141976821120.0,
      "grad_norm": 4.356844440513057,
      "language_loss": 1.01692307,
      "learning_rate": 2.271496085962064e-06,
      "loss": 1.06022406,
      "num_input_tokens_seen": 596145,
      "step": 23,
      "time_per_iteration": 2.6588985919952393
    },
    {
      "auxiliary_loss_clip": 0.02960346,
      "auxiliary_loss_mlp": 0.01295657,
      "balance_loss_clip": 1.55110884,
      "balance_loss_mlp": 1.09309649,
      "epoch": 0.002885829375338183,
      "flos": 20667381396480.0,
      "grad_norm": 3.7172172735020097,
      "language_loss": 1.02769744,
      "learning_rate": 2.3023282262611022e-06,
      "loss": 1.07025743,
      "num_input_tokens_seen": 614920,
      "step": 24,
      "time_per_iteration": 2.625957727432251
    },
    {
      "auxiliary_loss_clip": 0.02953816,
      "auxiliary_loss_mlp": 0.01285576,
      "balance_loss_clip": 1.55001831,
      "balance_loss_mlp": 1.09465003,
      "epoch": 0.003006072265977274,
      "flos": 34823869873920.0,
      "grad_norm": 3.718087701156514,
      "language_loss": 0.925192,
      "learning_rate": 2.3319015548620114e-06,
      "loss": 0.96758592,
      "num_input_tokens_seen": 636060,
      "step": 25,
      "time_per_iteration": 2.782104969024658
    },
    {
      "auxiliary_loss_clip": 0.02953851,
      "auxiliary_loss_mlp": 0.0130588,
      "balance_loss_clip": 1.54961729,
      "balance_loss_mlp": 1.10284245,
      "epoch": 0.003126315156616365,
      "flos": 24422021118720.0,
      "grad_norm": 2.1123687812523606,
      "language_loss": 0.9296093,
      "learning_rate": 2.3603148416618152e-06,
      "loss": 0.97220659,
      "num_input_tokens_seen": 655575,
      "step": 26,
      "time_per_iteration": 2.744178056716919
    },
    {
      "auxiliary_loss_clip": 0.02893513,
      "auxiliary_loss_mlp": 0.01279401,
      "balance_loss_clip": 1.54653335,
      "balance_loss_mlp": 1.08284855,
      "epoch": 0.003246558047255456,
      "flos": 23622326674560.0,
      "grad_norm": 2.3453359149862854,
      "language_loss": 1.00922859,
      "learning_rate": 2.3876556694204647e-06,
      "loss": 1.05095768,
      "num_input_tokens_seen": 675730,
      "step": 27,
      "time_per_iteration": 2.784576177597046
    },
    {
      "auxiliary_loss_clip": 0.02851188,
      "auxiliary_loss_mlp": 0.01291519,
      "balance_loss_clip": 1.53105366,
      "balance_loss_mlp": 1.09916306,
      "epoch": 0.003366800937894547,
      "flos": 17820275725440.0,
      "grad_norm": 2.8130831364628386,
      "language_loss": 0.90652895,
      "learning_rate": 2.414002061950908e-06,
      "loss": 0.94795597,
      "num_input_tokens_seen": 694605,
      "step": 28,
      "time_per_iteration": 2.7183852195739746
    },
    {
      "auxiliary_loss_clip": 0.02842421,
      "auxiliary_loss_mlp": 0.01286896,
      "balance_loss_clip": 1.53033519,
      "balance_loss_mlp": 1.0966382,
      "epoch": 0.003487043828533638,
      "flos": 24426115269120.0,
      "grad_norm": 2.321106806316836,
      "language_loss": 0.99701285,
      "learning_rate": 2.4394238264681557e-06,
      "loss": 1.03830612,
      "num_input_tokens_seen": 714340,
      "step": 29,
      "time_per_iteration": 2.7113656997680664
    },
    {
      "auxiliary_loss_clip": 0.02849551,
      "auxiliary_loss_mlp": 0.01251704,
      "balance_loss_clip": 1.53163314,
      "balance_loss_mlp": 1.07632351,
      "epoch": 0.003607286719172729,
      "flos": 26140311002880.0,
      "grad_norm": 2.23102532260135,
      "language_loss": 0.99556005,
      "learning_rate": 2.4639836682781433e-06,
      "loss": 1.03657258,
      "num_input_tokens_seen": 734470,
      "step": 30,
      "time_per_iteration": 2.733118772506714
    },
    {
      "auxiliary_loss_clip": 0.02800612,
      "auxiliary_loss_mlp": 0.01252605,
      "balance_loss_clip": 1.52106225,
      "balance_loss_mlp": 1.0782733,
      "epoch": 0.00372752960981182,
      "flos": 20593082113920.0,
      "grad_norm": 2.6031654197911154,
      "language_loss": 1.00234509,
      "learning_rate": 2.487738122623307e-06,
      "loss": 1.04287732,
      "num_input_tokens_seen": 753380,
      "step": 31,
      "time_per_iteration": 2.639840602874756
    },
    {
      "auxiliary_loss_clip": 0.02730956,
      "auxiliary_loss_mlp": 0.01269211,
      "balance_loss_clip": 1.50536776,
      "balance_loss_mlp": 1.0801928,
      "epoch": 0.003847772500450911,
      "flos": 22674608282880.0,
      "grad_norm": 2.5107703003494652,
      "language_loss": 0.98916757,
      "learning_rate": 2.510738338534912e-06,
      "loss": 1.0291692,
      "num_input_tokens_seen": 772105,
      "step": 32,
      "time_per_iteration": 2.7080843448638916
    },
    {
      "auxiliary_loss_clip": 0.02676956,
      "auxiliary_loss_mlp": 0.01250241,
      "balance_loss_clip": 1.48908234,
      "balance_loss_mlp": 1.07133174,
      "epoch": 0.003968015391090002,
      "flos": 17967796882560.0,
      "grad_norm": 2.6116982395605985,
      "language_loss": 1.02511954,
      "learning_rate": 2.5330307420306648e-06,
      "loss": 1.06439161,
      "num_input_tokens_seen": 788955,
      "step": 33,
      "time_per_iteration": 2.6340301036834717
    },
    {
      "auxiliary_loss_clip": 0.02633222,
      "auxiliary_loss_mlp": 0.0126895,
      "balance_loss_clip": 1.47956562,
      "balance_loss_mlp": 1.08593965,
      "epoch": 0.004088258281729093,
      "flos": 27304103658240.0,
      "grad_norm": 4.51222339252788,
      "language_loss": 0.88114727,
      "learning_rate": 2.554657600279796e-06,
      "loss": 0.920169,
      "num_input_tokens_seen": 810230,
      "step": 34,
      "time_per_iteration": 2.724830150604248
    },
    {
      "auxiliary_loss_clip": 0.02551436,
      "auxiliary_loss_mlp": 0.01237731,
      "balance_loss_clip": 1.46114039,
      "balance_loss_mlp": 1.07083786,
      "epoch": 0.004208501172368184,
      "flos": 23258587599360.0,
      "grad_norm": 2.1750515826516494,
      "language_loss": 1.03274202,
      "learning_rate": 2.5756575039679493e-06,
      "loss": 1.07063377,
      "num_input_tokens_seen": 829780,
      "step": 35,
      "time_per_iteration": 2.663585662841797
    },
    {
      "auxiliary_loss_clip": 0.0253613,
      "auxiliary_loss_mlp": 0.0126914,
      "balance_loss_clip": 1.45268869,
      "balance_loss_mlp": 1.09032583,
      "epoch": 0.0043287440630072746,
      "flos": 17312104062720.0,
      "grad_norm": 1.8707606647659145,
      "language_loss": 0.95025724,
      "learning_rate": 2.5960657816942747e-06,
      "loss": 0.98830992,
      "num_input_tokens_seen": 848695,
      "step": 36,
      "time_per_iteration": 2.6781909465789795
    },
    {
      "auxiliary_loss_clip": 0.02335059,
      "auxiliary_loss_mlp": 0.01229429,
      "balance_loss_clip": 1.51892543,
      "balance_loss_mlp": 1.130247,
      "epoch": 0.004448986953646365,
      "flos": 53092491160320.0,
      "grad_norm": 1.4048744057290041,
      "language_loss": 0.60968435,
      "learning_rate": 2.6159148575788668e-06,
      "loss": 0.64532924,
      "num_input_tokens_seen": 906730,
      "step": 37,
      "time_per_iteration": 3.1554055213928223
    },
    {
      "auxiliary_loss_clip": 0.02453547,
      "auxiliary_loss_mlp": 0.01279358,
      "balance_loss_clip": 1.43425667,
      "balance_loss_mlp": 1.10483563,
      "epoch": 0.004569229844285457,
      "flos": 13444165866240.0,
      "grad_norm": 2.7718108300585858,
      "language_loss": 0.9864037,
      "learning_rate": 2.635234561171e-06,
      "loss": 1.02373278,
      "num_input_tokens_seen": 925125,
      "step": 38,
      "time_per_iteration": 2.666367769241333
    },
    {
      "auxiliary_loss_clip": 0.02431874,
      "auxiliary_loss_mlp": 0.01235588,
      "balance_loss_clip": 1.43749022,
      "balance_loss_mlp": 1.06640649,
      "epoch": 0.0046894727349245475,
      "flos": 16209609966720.0,
      "grad_norm": 2.331339898382537,
      "language_loss": 0.9418813,
      "learning_rate": 2.6540523970949877e-06,
      "loss": 0.97855592,
      "num_input_tokens_seen": 939970,
      "step": 39,
      "time_per_iteration": 2.670243501663208
    },
    {
      "auxiliary_loss_clip": 0.02410703,
      "auxiliary_loss_mlp": 0.012417,
      "balance_loss_clip": 1.42974162,
      "balance_loss_mlp": 1.08014727,
      "epoch": 0.004809715625563638,
      "flos": 23914244505600.0,
      "grad_norm": 2.6696682227032653,
      "language_loss": 0.92517465,
      "learning_rate": 2.6723937805519533e-06,
      "loss": 0.96169871,
      "num_input_tokens_seen": 957470,
      "step": 40,
      "time_per_iteration": 3.509359359741211
    },
    {
      "auxiliary_loss_clip": 0.02401928,
      "auxiliary_loss_mlp": 0.01228995,
      "balance_loss_clip": 1.4225893,
      "balance_loss_mlp": 1.0744046,
      "epoch": 0.00492995851620273,
      "flos": 20773030273920.0,
      "grad_norm": 2.387773784094591,
      "language_loss": 0.92971754,
      "learning_rate": 2.690282243737839e-06,
      "loss": 0.96602672,
      "num_input_tokens_seen": 976405,
      "step": 41,
      "time_per_iteration": 3.4301517009735107
    },
    {
      "auxiliary_loss_clip": 0.02336054,
      "auxiliary_loss_mlp": 0.01241003,
      "balance_loss_clip": 1.40155196,
      "balance_loss_mlp": 1.08116686,
      "epoch": 0.0050502014068418205,
      "flos": 20338655103360.0,
      "grad_norm": 3.1684386353437444,
      "language_loss": 0.99301088,
      "learning_rate": 2.7077396173840807e-06,
      "loss": 1.02878141,
      "num_input_tokens_seen": 994690,
      "step": 42,
      "time_per_iteration": 2.6879496574401855
    },
    {
      "auxiliary_loss_clip": 0.02335765,
      "auxiliary_loss_mlp": 0.01238816,
      "balance_loss_clip": 1.40957808,
      "balance_loss_mlp": 1.078408,
      "epoch": 0.005170444297480911,
      "flos": 25994872834560.0,
      "grad_norm": 3.1619378935936573,
      "language_loss": 0.92803317,
      "learning_rate": 2.7247861909342594e-06,
      "loss": 0.96377897,
      "num_input_tokens_seen": 1015615,
      "step": 43,
      "time_per_iteration": 2.718348503112793
    },
    {
      "auxiliary_loss_clip": 0.02303502,
      "auxiliary_loss_mlp": 0.01216371,
      "balance_loss_clip": 1.40111566,
      "balance_loss_mlp": 1.06330585,
      "epoch": 0.005290687188120003,
      "flos": 20954055841920.0,
      "grad_norm": 2.3169774312669373,
      "language_loss": 0.83036232,
      "learning_rate": 2.7414408543044743e-06,
      "loss": 0.86556101,
      "num_input_tokens_seen": 1031255,
      "step": 44,
      "time_per_iteration": 2.705641746520996
    },
    {
      "auxiliary_loss_clip": 0.02258889,
      "auxiliary_loss_mlp": 0.01210843,
      "balance_loss_clip": 1.38939393,
      "balance_loss_mlp": 1.05520368,
      "epoch": 0.005410930078759093,
      "flos": 15851401585920.0,
      "grad_norm": 4.235121524790522,
      "language_loss": 0.79133725,
      "learning_rate": 2.7577212237113157e-06,
      "loss": 0.82603467,
      "num_input_tokens_seen": 1048295,
      "step": 45,
      "time_per_iteration": 2.614469051361084
    },
    {
      "auxiliary_loss_clip": 0.02244704,
      "auxiliary_loss_mlp": 0.01209598,
      "balance_loss_clip": 1.38951397,
      "balance_loss_mlp": 1.06416225,
      "epoch": 0.005531172969398184,
      "flos": 21104988791040.0,
      "grad_norm": 2.425211912821761,
      "language_loss": 1.0416863,
      "learning_rate": 2.7736437536690466e-06,
      "loss": 1.07622933,
      "num_input_tokens_seen": 1067925,
      "step": 46,
      "time_per_iteration": 2.662057399749756
    },
    {
      "auxiliary_loss_clip": 0.0223589,
      "auxiliary_loss_mlp": 0.01215962,
      "balance_loss_clip": 1.3838706,
      "balance_loss_mlp": 1.07205296,
      "epoch": 0.005651415860037276,
      "flos": 20844887431680.0,
      "grad_norm": 2.001310035414953,
      "language_loss": 1.07784915,
      "learning_rate": 2.789223836941131e-06,
      "loss": 1.11236763,
      "num_input_tokens_seen": 1088060,
      "step": 47,
      "time_per_iteration": 2.667766809463501
    },
    {
      "auxiliary_loss_clip": 0.02204592,
      "auxiliary_loss_mlp": 0.01219969,
      "balance_loss_clip": 1.37509871,
      "balance_loss_mlp": 1.07481992,
      "epoch": 0.005771658750676366,
      "flos": 13260195383040.0,
      "grad_norm": 2.3723719715788563,
      "language_loss": 1.0880394,
      "learning_rate": 2.8044758939680847e-06,
      "loss": 1.12228489,
      "num_input_tokens_seen": 1104130,
      "step": 48,
      "time_per_iteration": 2.6212713718414307
    },
    {
      "auxiliary_loss_clip": 0.02167197,
      "auxiliary_loss_mlp": 0.01215579,
      "balance_loss_clip": 1.37099004,
      "balance_loss_mlp": 1.08072948,
      "epoch": 0.005891901641315457,
      "flos": 24425396997120.0,
      "grad_norm": 3.2283571462991545,
      "language_loss": 1.0180223,
      "learning_rate": 2.8194134530738863e-06,
      "loss": 1.05185008,
      "num_input_tokens_seen": 1122900,
      "step": 49,
      "time_per_iteration": 2.643406867980957
    },
    {
      "auxiliary_loss_clip": 0.02140832,
      "auxiliary_loss_mlp": 0.01212232,
      "balance_loss_clip": 1.36449337,
      "balance_loss_mlp": 1.08014846,
      "epoch": 0.006012144531954548,
      "flos": 23076197314560.0,
      "grad_norm": 4.250806309686046,
      "language_loss": 0.90274978,
      "learning_rate": 2.834049222568994e-06,
      "loss": 0.93628043,
      "num_input_tokens_seen": 1140250,
      "step": 50,
      "time_per_iteration": 2.6331348419189453
    },
    {
      "auxiliary_loss_clip": 0.02149088,
      "auxiliary_loss_mlp": 0.01210894,
      "balance_loss_clip": 1.36088276,
      "balance_loss_mlp": 1.07690275,
      "epoch": 0.006132387422593639,
      "flos": 22528775064960.0,
      "grad_norm": 3.3185420554971565,
      "language_loss": 0.92510939,
      "learning_rate": 2.848395155712969e-06,
      "loss": 0.95870924,
      "num_input_tokens_seen": 1160470,
      "step": 51,
      "time_per_iteration": 2.623654842376709
    },
    {
      "auxiliary_loss_clip": 0.02119848,
      "auxiliary_loss_mlp": 0.01211518,
      "balance_loss_clip": 1.35791373,
      "balance_loss_mlp": 1.08105576,
      "epoch": 0.00625263031323273,
      "flos": 27628340751360.0,
      "grad_norm": 2.3549538104501275,
      "language_loss": 0.97784567,
      "learning_rate": 2.8624625093687977e-06,
      "loss": 1.0111593,
      "num_input_tokens_seen": 1177605,
      "step": 52,
      "time_per_iteration": 2.6782426834106445
    },
    {
      "auxiliary_loss_clip": 0.02119765,
      "auxiliary_loss_mlp": 0.01209272,
      "balance_loss_clip": 1.35756218,
      "balance_loss_mlp": 1.08043051,
      "epoch": 0.006372873203871821,
      "flos": 23110671392640.0,
      "grad_norm": 2.7979322053929705,
      "language_loss": 0.89082581,
      "learning_rate": 2.876261897070029e-06,
      "loss": 0.92411613,
      "num_input_tokens_seen": 1197735,
      "step": 53,
      "time_per_iteration": 2.6332337856292725
    },
    {
      "auxiliary_loss_clip": 0.0210693,
      "auxiliary_loss_mlp": 0.01196524,
      "balance_loss_clip": 1.35462236,
      "balance_loss_mlp": 1.06873155,
      "epoch": 0.006493116094510912,
      "flos": 22856028900480.0,
      "grad_norm": 2.340559889093807,
      "language_loss": 0.92575788,
      "learning_rate": 2.889803337127447e-06,
      "loss": 0.95879239,
      "num_input_tokens_seen": 1216335,
      "step": 54,
      "time_per_iteration": 2.6453781127929688
    },
    {
      "auxiliary_loss_clip": 0.02075212,
      "auxiliary_loss_mlp": 0.01217384,
      "balance_loss_clip": 1.34608614,
      "balance_loss_mlp": 1.09254766,
      "epoch": 0.006613358985150003,
      "flos": 23071708114560.0,
      "grad_norm": 2.499119240416735,
      "language_loss": 0.84730953,
      "learning_rate": 2.903096296321516e-06,
      "loss": 0.88023549,
      "num_input_tokens_seen": 1234480,
      "step": 55,
      "time_per_iteration": 2.6286933422088623
    },
    {
      "auxiliary_loss_clip": 0.02053994,
      "auxiliary_loss_mlp": 0.01198798,
      "balance_loss_clip": 1.34349608,
      "balance_loss_mlp": 1.07653666,
      "epoch": 0.006733601875789094,
      "flos": 26537662229760.0,
      "grad_norm": 1.9640366710844803,
      "language_loss": 0.91531092,
      "learning_rate": 2.9161497296578907e-06,
      "loss": 0.94783884,
      "num_input_tokens_seen": 1253870,
      "step": 56,
      "time_per_iteration": 2.703251361846924
    },
    {
      "auxiliary_loss_clip": 0.02052004,
      "auxiliary_loss_mlp": 0.01199723,
      "balance_loss_clip": 1.33968043,
      "balance_loss_mlp": 1.08156323,
      "epoch": 0.006853844766428185,
      "flos": 15523178083200.0,
      "grad_norm": 2.9955183099440696,
      "language_loss": 0.85976094,
      "learning_rate": 2.928972116604173e-06,
      "loss": 0.89227831,
      "num_input_tokens_seen": 1270145,
      "step": 57,
      "time_per_iteration": 2.6440587043762207
    },
    {
      "auxiliary_loss_clip": 0.02027551,
      "auxiliary_loss_mlp": 0.01184574,
      "balance_loss_clip": 1.32832968,
      "balance_loss_mlp": 1.06555593,
      "epoch": 0.006974087657067276,
      "flos": 24243760897920.0,
      "grad_norm": 2.281978007295325,
      "language_loss": 1.02157617,
      "learning_rate": 2.9415714941751377e-06,
      "loss": 1.05369735,
      "num_input_tokens_seen": 1291365,
      "step": 58,
      "time_per_iteration": 2.663088798522949
    },
    {
      "auxiliary_loss_clip": 0.02022483,
      "auxiliary_loss_mlp": 0.01205806,
      "balance_loss_clip": 1.32719791,
      "balance_loss_mlp": 1.0892669,
      "epoch": 0.007094330547706367,
      "flos": 25772513690880.0,
      "grad_norm": 1.9678687188826345,
      "language_loss": 0.93580437,
      "learning_rate": 2.9539554871897396e-06,
      "loss": 0.96808732,
      "num_input_tokens_seen": 1311535,
      "step": 59,
      "time_per_iteration": 2.649562358856201
    },
    {
      "auxiliary_loss_clip": 0.01990422,
      "auxiliary_loss_mlp": 0.01195514,
      "balance_loss_clip": 1.32149422,
      "balance_loss_mlp": 1.08565044,
      "epoch": 0.007214573438345458,
      "flos": 21319015979520.0,
      "grad_norm": 2.049598517044452,
      "language_loss": 0.97361022,
      "learning_rate": 2.9661313359851253e-06,
      "loss": 1.00546956,
      "num_input_tokens_seen": 1329420,
      "step": 60,
      "time_per_iteration": 2.6267647743225098
    },
    {
      "auxiliary_loss_clip": 0.0197324,
      "auxiliary_loss_mlp": 0.01185337,
      "balance_loss_clip": 1.3135643,
      "balance_loss_mlp": 1.07966971,
      "epoch": 0.007334816328984549,
      "flos": 24937088192640.0,
      "grad_norm": 4.5420674050303935,
      "language_loss": 0.94012964,
      "learning_rate": 2.978105921839922e-06,
      "loss": 0.97171545,
      "num_input_tokens_seen": 1349965,
      "step": 61,
      "time_per_iteration": 2.6749684810638428
    },
    {
      "auxiliary_loss_clip": 0.01959505,
      "auxiliary_loss_mlp": 0.01166968,
      "balance_loss_clip": 1.31642532,
      "balance_loss_mlp": 1.06420946,
      "epoch": 0.00745505921962364,
      "flos": 18510586277760.0,
      "grad_norm": 2.4023068207806926,
      "language_loss": 0.72311914,
      "learning_rate": 2.9898857903302893e-06,
      "loss": 0.7543838,
      "num_input_tokens_seen": 1368915,
      "step": 62,
      "time_per_iteration": 2.6423768997192383
    },
    {
      "auxiliary_loss_clip": 0.01947438,
      "auxiliary_loss_mlp": 0.01184091,
      "balance_loss_clip": 1.30920172,
      "balance_loss_mlp": 1.07761288,
      "epoch": 0.007575302110262731,
      "flos": 18477656484480.0,
      "grad_norm": 3.027365462666879,
      "language_loss": 0.87883735,
      "learning_rate": 3.001477172817253e-06,
      "loss": 0.91015261,
      "num_input_tokens_seen": 1386805,
      "step": 63,
      "time_per_iteration": 2.617429256439209
    },
    {
      "auxiliary_loss_clip": 0.01930127,
      "auxiliary_loss_mlp": 0.01190527,
      "balance_loss_clip": 1.30227172,
      "balance_loss_mlp": 1.08342946,
      "epoch": 0.007695545000901822,
      "flos": 24973178382720.0,
      "grad_norm": 2.2528603802102496,
      "language_loss": 0.96358716,
      "learning_rate": 3.012886006241894e-06,
      "loss": 0.99479365,
      "num_input_tokens_seen": 1406190,
      "step": 64,
      "time_per_iteration": 2.6402382850646973
    },
    {
      "auxiliary_loss_clip": 0.01932982,
      "auxiliary_loss_mlp": 0.01179333,
      "balance_loss_clip": 1.30915737,
      "balance_loss_mlp": 1.07833874,
      "epoch": 0.007815787891540913,
      "flos": 21324223451520.0,
      "grad_norm": 1.868596309629138,
      "language_loss": 0.88130593,
      "learning_rate": 3.0241179513858383e-06,
      "loss": 0.91242909,
      "num_input_tokens_seen": 1425500,
      "step": 65,
      "time_per_iteration": 2.6497275829315186
    },
    {
      "auxiliary_loss_clip": 0.01906888,
      "auxiliary_loss_mlp": 0.01151698,
      "balance_loss_clip": 1.29048944,
      "balance_loss_mlp": 1.05232513,
      "epoch": 0.007936030782180003,
      "flos": 21575777374080.0,
      "grad_norm": 5.766846321945436,
      "language_loss": 0.87761009,
      "learning_rate": 3.035178409737647e-06,
      "loss": 0.90819597,
      "num_input_tokens_seen": 1442950,
      "step": 66,
      "time_per_iteration": 3.536262035369873
    },
    {
      "auxiliary_loss_clip": 0.0189392,
      "auxiliary_loss_mlp": 0.0117598,
      "balance_loss_clip": 1.28699124,
      "balance_loss_mlp": 1.07474685,
      "epoch": 0.008056273672819095,
      "flos": 20120785159680.0,
      "grad_norm": 2.5265376964167174,
      "language_loss": 0.88755441,
      "learning_rate": 3.046072539090907e-06,
      "loss": 0.91825342,
      "num_input_tokens_seen": 1460915,
      "step": 67,
      "time_per_iteration": 4.280246734619141
    },
    {
      "auxiliary_loss_clip": 0.01868238,
      "auxiliary_loss_mlp": 0.0116643,
      "balance_loss_clip": 1.28048182,
      "balance_loss_mlp": 1.06705761,
      "epoch": 0.008176516563458186,
      "flos": 18333116156160.0,
      "grad_norm": 2.2665359022515696,
      "language_loss": 1.04749846,
      "learning_rate": 3.056805267986779e-06,
      "loss": 1.07784522,
      "num_input_tokens_seen": 1478385,
      "step": 68,
      "time_per_iteration": 2.6140341758728027
    },
    {
      "auxiliary_loss_clip": 0.01861962,
      "auxiliary_loss_mlp": 0.01169469,
      "balance_loss_clip": 1.27846122,
      "balance_loss_mlp": 1.07228994,
      "epoch": 0.008296759454097276,
      "flos": 21872076664320.0,
      "grad_norm": 2.8743211876702204,
      "language_loss": 0.95459169,
      "learning_rate": 3.0673813091022194e-06,
      "loss": 0.98490602,
      "num_input_tokens_seen": 1497605,
      "step": 69,
      "time_per_iteration": 2.6104815006256104
    },
    {
      "auxiliary_loss_clip": 0.01696683,
      "auxiliary_loss_mlp": 0.010526,
      "balance_loss_clip": 1.30511189,
      "balance_loss_mlp": 1.0148344,
      "epoch": 0.008417002344736368,
      "flos": 63408228036480.0,
      "grad_norm": 1.3078048472863455,
      "language_loss": 0.62139696,
      "learning_rate": 3.0778051716749317e-06,
      "loss": 0.64888978,
      "num_input_tokens_seen": 1561150,
      "step": 70,
      "time_per_iteration": 3.251384973526001
    },
    {
      "auxiliary_loss_clip": 0.0182163,
      "auxiliary_loss_mlp": 0.0115885,
      "balance_loss_clip": 1.25820971,
      "balance_loss_mlp": 1.06395936,
      "epoch": 0.008537245235375458,
      "flos": 22966454286720.0,
      "grad_norm": 3.6799912470802356,
      "language_loss": 0.9033252,
      "learning_rate": 3.0880811730470094e-06,
      "loss": 0.93313003,
      "num_input_tokens_seen": 1580605,
      "step": 71,
      "time_per_iteration": 2.606001377105713
    },
    {
      "auxiliary_loss_clip": 0.0166769,
      "auxiliary_loss_mlp": 0.01042049,
      "balance_loss_clip": 1.29177213,
      "balance_loss_mlp": 1.00695348,
      "epoch": 0.008657488126014549,
      "flos": 61984046712960.0,
      "grad_norm": 1.1979577701255546,
      "language_loss": 0.58597147,
      "learning_rate": 3.098213449401257e-06,
      "loss": 0.61306882,
      "num_input_tokens_seen": 1647535,
      "step": 72,
      "time_per_iteration": 3.138117790222168
    },
    {
      "auxiliary_loss_clip": 0.01803025,
      "auxiliary_loss_mlp": 0.01165256,
      "balance_loss_clip": 1.25482249,
      "balance_loss_mlp": 1.07398975,
      "epoch": 0.00877773101665364,
      "flos": 30296791152000.0,
      "grad_norm": 2.4667256319691373,
      "language_loss": 0.98773623,
      "learning_rate": 3.1082059657570015e-06,
      "loss": 1.0174191,
      "num_input_tokens_seen": 1666770,
      "step": 73,
      "time_per_iteration": 2.762204170227051
    },
    {
      "auxiliary_loss_clip": 0.01798488,
      "auxiliary_loss_mlp": 0.0117941,
      "balance_loss_clip": 1.25453186,
      "balance_loss_mlp": 1.08962166,
      "epoch": 0.00889797390729273,
      "flos": 23514056104320.0,
      "grad_norm": 2.2211573237492193,
      "language_loss": 0.96656609,
      "learning_rate": 3.1180625252858496e-06,
      "loss": 0.99634498,
      "num_input_tokens_seen": 1685200,
      "step": 74,
      "time_per_iteration": 2.7041971683502197
    },
    {
      "auxiliary_loss_clip": 0.01774675,
      "auxiliary_loss_mlp": 0.01145069,
      "balance_loss_clip": 1.24564385,
      "balance_loss_mlp": 1.06815553,
      "epoch": 0.009018216797931822,
      "flos": 23075838178560.0,
      "grad_norm": 2.857009616960915,
      "language_loss": 0.80012983,
      "learning_rate": 3.1277867780021663e-06,
      "loss": 0.82932729,
      "num_input_tokens_seen": 1701835,
      "step": 75,
      "time_per_iteration": 2.6951398849487305
    },
    {
      "auxiliary_loss_clip": 0.01756258,
      "auxiliary_loss_mlp": 0.01147956,
      "balance_loss_clip": 1.23863006,
      "balance_loss_mlp": 1.06908679,
      "epoch": 0.009138459688570914,
      "flos": 15918877284480.0,
      "grad_norm": 2.2759806739176773,
      "language_loss": 0.95767343,
      "learning_rate": 3.1373822288779824e-06,
      "loss": 0.98671556,
      "num_input_tokens_seen": 1718415,
      "step": 76,
      "time_per_iteration": 2.620680570602417
    },
    {
      "auxiliary_loss_clip": 0.01755157,
      "auxiliary_loss_mlp": 0.01163957,
      "balance_loss_clip": 1.24133897,
      "balance_loss_mlp": 1.08804488,
      "epoch": 0.009258702579210003,
      "flos": 27016531372800.0,
      "grad_norm": 2.453429217808873,
      "language_loss": 0.79604065,
      "learning_rate": 3.1468522454274533e-06,
      "loss": 0.82523179,
      "num_input_tokens_seen": 1738770,
      "step": 77,
      "time_per_iteration": 2.6683523654937744
    },
    {
      "auxiliary_loss_clip": 0.01735096,
      "auxiliary_loss_mlp": 0.01148421,
      "balance_loss_clip": 1.23238397,
      "balance_loss_mlp": 1.07236588,
      "epoch": 0.009378945469849095,
      "flos": 26903196984960.0,
      "grad_norm": 2.2516331397481473,
      "language_loss": 0.91843367,
      "learning_rate": 3.15620006480197e-06,
      "loss": 0.94726884,
      "num_input_tokens_seen": 1758040,
      "step": 78,
      "time_per_iteration": 2.628598690032959
    },
    {
      "auxiliary_loss_clip": 0.01722777,
      "auxiliary_loss_mlp": 0.01152976,
      "balance_loss_clip": 1.22714949,
      "balance_loss_mlp": 1.07844615,
      "epoch": 0.009499188360488187,
      "flos": 35694236327040.0,
      "grad_norm": 3.44995959619227,
      "language_loss": 0.75064266,
      "learning_rate": 3.1654288004333087e-06,
      "loss": 0.77940023,
      "num_input_tokens_seen": 1776705,
      "step": 79,
      "time_per_iteration": 2.735701322555542
    },
    {
      "auxiliary_loss_clip": 0.01722334,
      "auxiliary_loss_mlp": 0.01130356,
      "balance_loss_clip": 1.22725534,
      "balance_loss_mlp": 1.05692279,
      "epoch": 0.009619431251127276,
      "flos": 21503201944320.0,
      "grad_norm": 2.473543331226019,
      "language_loss": 0.76173967,
      "learning_rate": 3.1745414482589353e-06,
      "loss": 0.79026651,
      "num_input_tokens_seen": 1795915,
      "step": 80,
      "time_per_iteration": 2.596764087677002
    },
    {
      "auxiliary_loss_clip": 0.01707214,
      "auxiliary_loss_mlp": 0.01136657,
      "balance_loss_clip": 1.22245479,
      "balance_loss_mlp": 1.06570315,
      "epoch": 0.009739674141766368,
      "flos": 17421056991360.0,
      "grad_norm": 52.57095678243634,
      "language_loss": 0.87097412,
      "learning_rate": 3.1835408925606204e-06,
      "loss": 0.89941281,
      "num_input_tokens_seen": 1814055,
      "step": 81,
      "time_per_iteration": 2.6182422637939453
    },
    {
      "auxiliary_loss_clip": 0.01697828,
      "auxiliary_loss_mlp": 0.01130204,
      "balance_loss_clip": 1.22360909,
      "balance_loss_mlp": 1.06244516,
      "epoch": 0.00985991703240546,
      "flos": 27527109246720.0,
      "grad_norm": 2.449771706006641,
      "language_loss": 0.89386612,
      "learning_rate": 3.1924299114448214e-06,
      "loss": 0.92214644,
      "num_input_tokens_seen": 1834535,
      "step": 82,
      "time_per_iteration": 2.659130573272705
    },
    {
      "auxiliary_loss_clip": 0.01694038,
      "auxiliary_loss_mlp": 0.01148796,
      "balance_loss_clip": 1.22047734,
      "balance_loss_mlp": 1.07994056,
      "epoch": 0.00998015992304455,
      "flos": 13808084509440.0,
      "grad_norm": 2.6534820905003613,
      "language_loss": 0.83321601,
      "learning_rate": 3.2012111819909055e-06,
      "loss": 0.86164439,
      "num_input_tokens_seen": 1851865,
      "step": 83,
      "time_per_iteration": 2.6104140281677246
    },
    {
      "auxiliary_loss_clip": 0.01679679,
      "auxiliary_loss_mlp": 0.01125962,
      "balance_loss_clip": 1.21525764,
      "balance_loss_mlp": 1.06087339,
      "epoch": 0.010100402813683641,
      "flos": 20191385341440.0,
      "grad_norm": 2.4150544968692365,
      "language_loss": 0.95006216,
      "learning_rate": 3.2098872850910627e-06,
      "loss": 0.97811848,
      "num_input_tokens_seen": 1868540,
      "step": 84,
      "time_per_iteration": 2.612663984298706
    },
    {
      "auxiliary_loss_clip": 0.01682143,
      "auxiliary_loss_mlp": 0.0112846,
      "balance_loss_clip": 1.21499741,
      "balance_loss_mlp": 1.05860376,
      "epoch": 0.010220645704322733,
      "flos": 17201642762880.0,
      "grad_norm": 3.3081700609859275,
      "language_loss": 0.8918491,
      "learning_rate": 3.2184607100038194e-06,
      "loss": 0.91995513,
      "num_input_tokens_seen": 1887180,
      "step": 85,
      "time_per_iteration": 2.61582350730896
    },
    {
      "auxiliary_loss_clip": 0.01683408,
      "auxiliary_loss_mlp": 0.01138015,
      "balance_loss_clip": 1.21688366,
      "balance_loss_mlp": 1.07292712,
      "epoch": 0.010340888594961822,
      "flos": 21470415805440.0,
      "grad_norm": 2.0090636295607385,
      "language_loss": 0.93072778,
      "learning_rate": 3.2269338586412414e-06,
      "loss": 0.95894206,
      "num_input_tokens_seen": 1904765,
      "step": 86,
      "time_per_iteration": 2.6697964668273926
    },
    {
      "auxiliary_loss_clip": 0.01666922,
      "auxiliary_loss_mlp": 0.01127021,
      "balance_loss_clip": 1.21045959,
      "balance_loss_mlp": 1.06736875,
      "epoch": 0.010461131485600914,
      "flos": 23002831785600.0,
      "grad_norm": 3.2065077979377437,
      "language_loss": 0.96568251,
      "learning_rate": 3.2353090496083106e-06,
      "loss": 0.99362195,
      "num_input_tokens_seen": 1922600,
      "step": 87,
      "time_per_iteration": 2.5955700874328613
    },
    {
      "auxiliary_loss_clip": 0.01652881,
      "auxiliary_loss_mlp": 0.01133292,
      "balance_loss_clip": 1.20469308,
      "balance_loss_mlp": 1.06910944,
      "epoch": 0.010581374376240005,
      "flos": 33546850571520.0,
      "grad_norm": 1.969711958388999,
      "language_loss": 0.81489229,
      "learning_rate": 3.2435885220114572e-06,
      "loss": 0.84275401,
      "num_input_tokens_seen": 1943950,
      "step": 88,
      "time_per_iteration": 2.7678210735321045
    },
    {
      "auxiliary_loss_clip": 0.01654163,
      "auxiliary_loss_mlp": 0.01136926,
      "balance_loss_clip": 1.20783734,
      "balance_loss_mlp": 1.07651079,
      "epoch": 0.010701617266879095,
      "flos": 21763087822080.0,
      "grad_norm": 2.050978301637826,
      "language_loss": 0.93889338,
      "learning_rate": 3.2517744390519113e-06,
      "loss": 0.96680427,
      "num_input_tokens_seen": 1962815,
      "step": 89,
      "time_per_iteration": 2.575794219970703
    },
    {
      "auxiliary_loss_clip": 0.01641301,
      "auxiliary_loss_mlp": 0.0111392,
      "balance_loss_clip": 1.19831264,
      "balance_loss_mlp": 1.05851138,
      "epoch": 0.010821860157518187,
      "flos": 19060199256960.0,
      "grad_norm": 1.9197973314752943,
      "language_loss": 0.75333911,
      "learning_rate": 3.259868891418298e-06,
      "loss": 0.7808913,
      "num_input_tokens_seen": 1980580,
      "step": 90,
      "time_per_iteration": 2.6197564601898193
    },
    {
      "auxiliary_loss_clip": 0.01647607,
      "auxiliary_loss_mlp": 0.01117331,
      "balance_loss_clip": 1.20769215,
      "balance_loss_mlp": 1.06182742,
      "epoch": 0.010942103048157278,
      "flos": 25447378757760.0,
      "grad_norm": 1.9585524923089532,
      "language_loss": 0.85209805,
      "learning_rate": 3.2678739004917757e-06,
      "loss": 0.87974739,
      "num_input_tokens_seen": 2000315,
      "step": 91,
      "time_per_iteration": 2.6192240715026855
    },
    {
      "auxiliary_loss_clip": 0.01627726,
      "auxiliary_loss_mlp": 0.01116929,
      "balance_loss_clip": 1.20025921,
      "balance_loss_mlp": 1.0605197,
      "epoch": 0.011062345938796368,
      "flos": 27493928058240.0,
      "grad_norm": 1.753230572550286,
      "language_loss": 0.9222917,
      "learning_rate": 3.275791421376029e-06,
      "loss": 0.94973826,
      "num_input_tokens_seen": 2023760,
      "step": 92,
      "time_per_iteration": 3.434644937515259
    },
    {
      "auxiliary_loss_clip": 0.01629189,
      "auxiliary_loss_mlp": 0.01122177,
      "balance_loss_clip": 1.19942498,
      "balance_loss_mlp": 1.06433678,
      "epoch": 0.01118258882943546,
      "flos": 16071210864000.0,
      "grad_norm": 2.520805703205127,
      "language_loss": 0.95918185,
      "learning_rate": 3.2836233457634622e-06,
      "loss": 0.98669553,
      "num_input_tokens_seen": 2041895,
      "step": 93,
      "time_per_iteration": 4.795124292373657
    },
    {
      "auxiliary_loss_clip": 0.01624645,
      "auxiliary_loss_mlp": 0.01120945,
      "balance_loss_clip": 1.19853473,
      "balance_loss_mlp": 1.06663322,
      "epoch": 0.011302831720074551,
      "flos": 20668602458880.0,
      "grad_norm": 1.950314419045031,
      "language_loss": 0.85309523,
      "learning_rate": 3.2913715046481135e-06,
      "loss": 0.8805511,
      "num_input_tokens_seen": 2061640,
      "step": 94,
      "time_per_iteration": 2.612501859664917
    },
    {
      "auxiliary_loss_clip": 0.01612217,
      "auxiliary_loss_mlp": 0.01105653,
      "balance_loss_clip": 1.19238138,
      "balance_loss_mlp": 1.04976797,
      "epoch": 0.011423074610713641,
      "flos": 13072238490240.0,
      "grad_norm": 2.0115840429806724,
      "language_loss": 0.89022559,
      "learning_rate": 3.299037670895023e-06,
      "loss": 0.91740429,
      "num_input_tokens_seen": 2078255,
      "step": 95,
      "time_per_iteration": 2.581861734390259
    },
    {
      "auxiliary_loss_clip": 0.0162201,
      "auxiliary_loss_mlp": 0.01109103,
      "balance_loss_clip": 1.20257556,
      "balance_loss_mlp": 1.06053698,
      "epoch": 0.011543317501352733,
      "flos": 30335646689280.0,
      "grad_norm": 2.335774190296398,
      "language_loss": 0.80262685,
      "learning_rate": 3.3066235616750667e-06,
      "loss": 0.82993793,
      "num_input_tokens_seen": 2099490,
      "step": 96,
      "time_per_iteration": 2.6745665073394775
    },
    {
      "auxiliary_loss_clip": 0.015998,
      "auxiliary_loss_mlp": 0.01111173,
      "balance_loss_clip": 1.18832874,
      "balance_loss_mlp": 1.06303644,
      "epoch": 0.011663560391991824,
      "flos": 15522962601600.0,
      "grad_norm": 2.1509186643232474,
      "language_loss": 0.92303348,
      "learning_rate": 3.3141308407736276e-06,
      "loss": 0.95014316,
      "num_input_tokens_seen": 2116125,
      "step": 97,
      "time_per_iteration": 2.577507972717285
    },
    {
      "auxiliary_loss_clip": 0.01594088,
      "auxiliary_loss_mlp": 0.01110039,
      "balance_loss_clip": 1.18061233,
      "balance_loss_mlp": 1.05916059,
      "epoch": 0.011783803282630914,
      "flos": 19902125116800.0,
      "grad_norm": 1.9118960346543807,
      "language_loss": 0.8669939,
      "learning_rate": 3.321561120780869e-06,
      "loss": 0.89403516,
      "num_input_tokens_seen": 2134835,
      "step": 98,
      "time_per_iteration": 2.5909221172332764
    },
    {
      "auxiliary_loss_clip": 0.01595471,
      "auxiliary_loss_mlp": 0.0111065,
      "balance_loss_clip": 1.19286859,
      "balance_loss_mlp": 1.06494498,
      "epoch": 0.011904046173270006,
      "flos": 22340674517760.0,
      "grad_norm": 2.695088146677972,
      "language_loss": 1.01540196,
      "learning_rate": 3.3289159651708192e-06,
      "loss": 1.04246306,
      "num_input_tokens_seen": 2152410,
      "step": 99,
      "time_per_iteration": 2.5690414905548096
    },
    {
      "auxiliary_loss_clip": 0.01589214,
      "auxiliary_loss_mlp": 0.01108249,
      "balance_loss_clip": 1.18928754,
      "balance_loss_mlp": 1.06225824,
      "epoch": 0.012024289063909096,
      "flos": 19100060375040.0,
      "grad_norm": 2.0058858725524455,
      "language_loss": 0.97671264,
      "learning_rate": 3.3361968902759768e-06,
      "loss": 1.00368738,
      "num_input_tokens_seen": 2172090,
      "step": 100,
      "time_per_iteration": 2.5943081378936768
    },
    {
      "auxiliary_loss_clip": 0.01577268,
      "auxiliary_loss_mlp": 0.01109525,
      "balance_loss_clip": 1.18775868,
      "balance_loss_mlp": 1.06436861,
      "epoch": 0.012144531954548187,
      "flos": 15012205159680.0,
      "grad_norm": 2.1074223861283157,
      "language_loss": 0.93874091,
      "learning_rate": 3.343405367163663e-06,
      "loss": 0.96560884,
      "num_input_tokens_seen": 2189020,
      "step": 101,
      "time_per_iteration": 2.6027944087982178
    },
    {
      "auxiliary_loss_clip": 0.01580876,
      "auxiliary_loss_mlp": 0.01109176,
      "balance_loss_clip": 1.18550253,
      "balance_loss_mlp": 1.06618881,
      "epoch": 0.012264774845187279,
      "flos": 15122020014720.0,
      "grad_norm": 2.267818428956463,
      "language_loss": 0.81062484,
      "learning_rate": 3.350542823419951e-06,
      "loss": 0.83752531,
      "num_input_tokens_seen": 2205620,
      "step": 102,
      "time_per_iteration": 2.576003313064575
    },
    {
      "auxiliary_loss_clip": 0.01575112,
      "auxiliary_loss_mlp": 0.01106025,
      "balance_loss_clip": 1.18440056,
      "balance_loss_mlp": 1.05943799,
      "epoch": 0.012385017735826368,
      "flos": 13949248959360.0,
      "grad_norm": 4.169865642013476,
      "language_loss": 0.87617517,
      "learning_rate": 3.3576106448465615e-06,
      "loss": 0.90298653,
      "num_input_tokens_seen": 2219000,
      "step": 103,
      "time_per_iteration": 2.578104019165039
    },
    {
      "auxiliary_loss_clip": 0.01570133,
      "auxiliary_loss_mlp": 0.01122099,
      "balance_loss_clip": 1.17977023,
      "balance_loss_mlp": 1.07634616,
      "epoch": 0.01250526062646546,
      "flos": 23623260428160.0,
      "grad_norm": 2.4388233641723254,
      "language_loss": 0.8803333,
      "learning_rate": 3.3646101770757797e-06,
      "loss": 0.90725553,
      "num_input_tokens_seen": 2237790,
      "step": 104,
      "time_per_iteration": 2.667443037033081
    },
    {
      "auxiliary_loss_clip": 0.01568706,
      "auxiliary_loss_mlp": 0.01095717,
      "balance_loss_clip": 1.18405902,
      "balance_loss_mlp": 1.05344534,
      "epoch": 0.012625503517104552,
      "flos": 34640078958720.0,
      "grad_norm": 2.1792645868865708,
      "language_loss": 0.85637951,
      "learning_rate": 3.371542727108104e-06,
      "loss": 0.88302374,
      "num_input_tokens_seen": 2259965,
      "step": 105,
      "time_per_iteration": 2.7457308769226074
    },
    {
      "auxiliary_loss_clip": 0.01568199,
      "auxiliary_loss_mlp": 0.01095988,
      "balance_loss_clip": 1.18159127,
      "balance_loss_mlp": 1.05507517,
      "epoch": 0.012745746407743641,
      "flos": 17821891837440.0,
      "grad_norm": 3.3076253332945047,
      "language_loss": 0.90158856,
      "learning_rate": 3.3784095647770114e-06,
      "loss": 0.92823046,
      "num_input_tokens_seen": 2278610,
      "step": 106,
      "time_per_iteration": 2.6095962524414062
    },
    {
      "auxiliary_loss_clip": 0.01556989,
      "auxiliary_loss_mlp": 0.0111149,
      "balance_loss_clip": 1.17649579,
      "balance_loss_mlp": 1.06824064,
      "epoch": 0.012865989298382733,
      "flos": 20595057361920.0,
      "grad_norm": 3.4918115954453315,
      "language_loss": 0.88824439,
      "learning_rate": 3.3852119241449547e-06,
      "loss": 0.91492915,
      "num_input_tokens_seen": 2297730,
      "step": 107,
      "time_per_iteration": 2.5979321002960205
    },
    {
      "auxiliary_loss_clip": 0.01554947,
      "auxiliary_loss_mlp": 0.01105896,
      "balance_loss_clip": 1.17780042,
      "balance_loss_mlp": 1.06631827,
      "epoch": 0.012986232189021825,
      "flos": 23948969978880.0,
      "grad_norm": 2.2252672285471156,
      "language_loss": 0.96202177,
      "learning_rate": 3.3919510048344295e-06,
      "loss": 0.98863024,
      "num_input_tokens_seen": 2315740,
      "step": 108,
      "time_per_iteration": 2.627350330352783
    },
    {
      "auxiliary_loss_clip": 0.0154982,
      "auxiliary_loss_mlp": 0.01099091,
      "balance_loss_clip": 1.17764211,
      "balance_loss_mlp": 1.06256485,
      "epoch": 0.013106475079660914,
      "flos": 23725425686400.0,
      "grad_norm": 2.342722546609425,
      "language_loss": 0.86791027,
      "learning_rate": 3.3986279732976907e-06,
      "loss": 0.8943994,
      "num_input_tokens_seen": 2334215,
      "step": 109,
      "time_per_iteration": 2.6053249835968018
    },
    {
      "auxiliary_loss_clip": 0.01547528,
      "auxiliary_loss_mlp": 0.01086079,
      "balance_loss_clip": 1.17781806,
      "balance_loss_mlp": 1.05007803,
      "epoch": 0.013226717970300006,
      "flos": 21102438925440.0,
      "grad_norm": 2.207749537151733,
      "language_loss": 0.95349258,
      "learning_rate": 3.4052439640284983e-06,
      "loss": 0.9798286,
      "num_input_tokens_seen": 2353130,
      "step": 110,
      "time_per_iteration": 2.6069517135620117
    },
    {
      "auxiliary_loss_clip": 0.01538109,
      "auxiliary_loss_mlp": 0.01096616,
      "balance_loss_clip": 1.17132282,
      "balance_loss_mlp": 1.05973268,
      "epoch": 0.013346960860939098,
      "flos": 24863902231680.0,
      "grad_norm": 1.7341826275082421,
      "language_loss": 0.81045955,
      "learning_rate": 3.4118000807190217e-06,
      "loss": 0.83680677,
      "num_input_tokens_seen": 2374010,
      "step": 111,
      "time_per_iteration": 2.596102476119995
    },
    {
      "auxiliary_loss_clip": 0.01541809,
      "auxiliary_loss_mlp": 0.01095109,
      "balance_loss_clip": 1.17327666,
      "balance_loss_mlp": 1.05875039,
      "epoch": 0.013467203751578187,
      "flos": 28181940140160.0,
      "grad_norm": 2.191465930464939,
      "language_loss": 0.76036549,
      "learning_rate": 3.4182973973648723e-06,
      "loss": 0.78673464,
      "num_input_tokens_seen": 2395220,
      "step": 112,
      "time_per_iteration": 2.653881072998047
    },
    {
      "auxiliary_loss_clip": 0.0153594,
      "auxiliary_loss_mlp": 0.01094859,
      "balance_loss_clip": 1.1702354,
      "balance_loss_mlp": 1.05690265,
      "epoch": 0.013587446642217279,
      "flos": 18916233546240.0,
      "grad_norm": 2.670462361131529,
      "language_loss": 0.95204437,
      "learning_rate": 3.424736959321014e-06,
      "loss": 0.97835237,
      "num_input_tokens_seen": 2413025,
      "step": 113,
      "time_per_iteration": 2.543848991394043
    },
    {
      "auxiliary_loss_clip": 0.01532884,
      "auxiliary_loss_mlp": 0.01093502,
      "balance_loss_clip": 1.17091274,
      "balance_loss_mlp": 1.05871677,
      "epoch": 0.01370768953285637,
      "flos": 23988615615360.0,
      "grad_norm": 2.1531393542132413,
      "language_loss": 0.88732773,
      "learning_rate": 3.431119784311155e-06,
      "loss": 0.91359162,
      "num_input_tokens_seen": 2432700,
      "step": 114,
      "time_per_iteration": 2.6215999126434326
    },
    {
      "auxiliary_loss_clip": 0.01528723,
      "auxiliary_loss_mlp": 0.01112311,
      "balance_loss_clip": 1.16879427,
      "balance_loss_mlp": 1.0796473,
      "epoch": 0.01382793242349546,
      "flos": 39202565512320.0,
      "grad_norm": 3.667384507034687,
      "language_loss": 0.77449298,
      "learning_rate": 3.43744686339307e-06,
      "loss": 0.80090332,
      "num_input_tokens_seen": 2455020,
      "step": 115,
      "time_per_iteration": 2.7326204776763916
    },
    {
      "auxiliary_loss_clip": 0.01527806,
      "auxiliary_loss_mlp": 0.01105652,
      "balance_loss_clip": 1.16489601,
      "balance_loss_mlp": 1.06841052,
      "epoch": 0.013948175314134552,
      "flos": 41353506714240.0,
      "grad_norm": 2.1083415486986317,
      "language_loss": 0.91046405,
      "learning_rate": 3.44371916188212e-06,
      "loss": 0.93679863,
      "num_input_tokens_seen": 2475775,
      "step": 116,
      "time_per_iteration": 2.753671407699585
    },
    {
      "auxiliary_loss_clip": 0.0152828,
      "auxiliary_loss_mlp": 0.01094853,
      "balance_loss_clip": 1.16817081,
      "balance_loss_mlp": 1.06254745,
      "epoch": 0.014068418204773643,
      "flos": 22453542028800.0,
      "grad_norm": 2.331354152164771,
      "language_loss": 0.86081684,
      "learning_rate": 3.449937620235143e-06,
      "loss": 0.88704818,
      "num_input_tokens_seen": 2496370,
      "step": 117,
      "time_per_iteration": 3.39797043800354
    },
    {
      "auxiliary_loss_clip": 0.01521551,
      "auxiliary_loss_mlp": 0.01103766,
      "balance_loss_clip": 1.16536784,
      "balance_loss_mlp": 1.07291484,
      "epoch": 0.014188661095412733,
      "flos": 23805147922560.0,
      "grad_norm": 1.7247211910075553,
      "language_loss": 0.89481235,
      "learning_rate": 3.456103154896722e-06,
      "loss": 0.92106551,
      "num_input_tokens_seen": 2517645,
      "step": 118,
      "time_per_iteration": 2.6332106590270996
    },
    {
      "auxiliary_loss_clip": 0.01511365,
      "auxiliary_loss_mlp": 0.01089684,
      "balance_loss_clip": 1.16303468,
      "balance_loss_mlp": 1.05683017,
      "epoch": 0.014308903986051825,
      "flos": 23660248458240.0,
      "grad_norm": 2.3849491942730485,
      "language_loss": 0.92460579,
      "learning_rate": 3.462216659109757e-06,
      "loss": 0.95061624,
      "num_input_tokens_seen": 2537825,
      "step": 119,
      "time_per_iteration": 4.125097990036011
    },
    {
      "auxiliary_loss_clip": 0.01516607,
      "auxiliary_loss_mlp": 0.01093069,
      "balance_loss_clip": 1.1648972,
      "balance_loss_mlp": 1.06209815,
      "epoch": 0.014429146876690916,
      "flos": 20667991927680.0,
      "grad_norm": 2.27335723178321,
      "language_loss": 0.85126913,
      "learning_rate": 3.4682790036921077e-06,
      "loss": 0.87736595,
      "num_input_tokens_seen": 2556485,
      "step": 120,
      "time_per_iteration": 2.5914762020111084
    },
    {
      "auxiliary_loss_clip": 0.01509401,
      "auxiliary_loss_mlp": 0.01086409,
      "balance_loss_clip": 1.16299069,
      "balance_loss_mlp": 1.05704761,
      "epoch": 0.014549389767330006,
      "flos": 20229199384320.0,
      "grad_norm": 2.0301343346690324,
      "language_loss": 0.83341813,
      "learning_rate": 3.4742910377810193e-06,
      "loss": 0.85937619,
      "num_input_tokens_seen": 2573945,
      "step": 121,
      "time_per_iteration": 2.587437391281128
    },
    {
      "auxiliary_loss_clip": 0.01511513,
      "auxiliary_loss_mlp": 0.01090445,
      "balance_loss_clip": 1.16439426,
      "balance_loss_mlp": 1.0614531,
      "epoch": 0.014669632657969098,
      "flos": 18004174381440.0,
      "grad_norm": 6.26819121268211,
      "language_loss": 0.88623309,
      "learning_rate": 3.4802535895469042e-06,
      "loss": 0.91225272,
      "num_input_tokens_seen": 2592695,
      "step": 122,
      "time_per_iteration": 2.537217617034912
    },
    {
      "auxiliary_loss_clip": 0.01508156,
      "auxiliary_loss_mlp": 0.01084531,
      "balance_loss_clip": 1.16541624,
      "balance_loss_mlp": 1.05663586,
      "epoch": 0.01478987554860819,
      "flos": 22741796672640.0,
      "grad_norm": 1.9301833841520135,
      "language_loss": 0.89698803,
      "learning_rate": 3.4861674668779934e-06,
      "loss": 0.92291492,
      "num_input_tokens_seen": 2610925,
      "step": 123,
      "time_per_iteration": 2.587790012359619
    },
    {
      "auxiliary_loss_clip": 0.01501833,
      "auxiliary_loss_mlp": 0.01086957,
      "balance_loss_clip": 1.15893316,
      "balance_loss_mlp": 1.05826378,
      "epoch": 0.01491011843924728,
      "flos": 17198590106880.0,
      "grad_norm": 1.8233780493803151,
      "language_loss": 0.84132445,
      "learning_rate": 3.492033458037272e-06,
      "loss": 0.8672123,
      "num_input_tokens_seen": 2629495,
      "step": 124,
      "time_per_iteration": 2.5837199687957764
    },
    {
      "auxiliary_loss_clip": 0.01499895,
      "auxiliary_loss_mlp": 0.01082574,
      "balance_loss_clip": 1.15823507,
      "balance_loss_mlp": 1.05455995,
      "epoch": 0.01503036132988637,
      "flos": 17673867889920.0,
      "grad_norm": 3.05631289621366,
      "language_loss": 0.86898708,
      "learning_rate": 3.497852332293018e-06,
      "loss": 0.89481175,
      "num_input_tokens_seen": 2645070,
      "step": 125,
      "time_per_iteration": 2.5391080379486084
    },
    {
      "auxiliary_loss_clip": 0.01504806,
      "auxiliary_loss_mlp": 0.01079669,
      "balance_loss_clip": 1.16009521,
      "balance_loss_mlp": 1.05115438,
      "epoch": 0.015150604220525462,
      "flos": 18878239935360.0,
      "grad_norm": 4.610784062353718,
      "language_loss": 0.96682847,
      "learning_rate": 3.5036248405242356e-06,
      "loss": 0.99267322,
      "num_input_tokens_seen": 2663825,
      "step": 126,
      "time_per_iteration": 2.545975685119629
    },
    {
      "auxiliary_loss_clip": 0.01499331,
      "auxiliary_loss_mlp": 0.01092533,
      "balance_loss_clip": 1.15824389,
      "balance_loss_mlp": 1.06464982,
      "epoch": 0.015270847111164552,
      "flos": 39420184060800.0,
      "grad_norm": 2.0763385236895,
      "language_loss": 0.82519841,
      "learning_rate": 3.509351715802146e-06,
      "loss": 0.85111707,
      "num_input_tokens_seen": 2684710,
      "step": 127,
      "time_per_iteration": 2.727416515350342
    },
    {
      "auxiliary_loss_clip": 0.01495719,
      "auxiliary_loss_mlp": 0.01080554,
      "balance_loss_clip": 1.15637851,
      "balance_loss_mlp": 1.05389857,
      "epoch": 0.015391090001803644,
      "flos": 43762466286720.0,
      "grad_norm": 2.205606749855949,
      "language_loss": 0.78285837,
      "learning_rate": 3.5150336739488763e-06,
      "loss": 0.80862111,
      "num_input_tokens_seen": 2706995,
      "step": 128,
      "time_per_iteration": 2.779639482498169
    },
    {
      "auxiliary_loss_clip": 0.01489636,
      "auxiliary_loss_mlp": 0.01070541,
      "balance_loss_clip": 1.15665209,
      "balance_loss_mlp": 1.04666317,
      "epoch": 0.015511332892442733,
      "flos": 18916341287040.0,
      "grad_norm": 1.9280132084612605,
      "language_loss": 0.84154439,
      "learning_rate": 3.5206714140744143e-06,
      "loss": 0.86714619,
      "num_input_tokens_seen": 2727050,
      "step": 129,
      "time_per_iteration": 2.5853497982025146
    },
    {
      "auxiliary_loss_clip": 0.01496167,
      "auxiliary_loss_mlp": 0.01108898,
      "balance_loss_clip": 1.16463065,
      "balance_loss_mlp": 1.08212316,
      "epoch": 0.015631575783081827,
      "flos": 24535283679360.0,
      "grad_norm": 2.378796961498555,
      "language_loss": 0.87463653,
      "learning_rate": 3.5262656190928208e-06,
      "loss": 0.90068722,
      "num_input_tokens_seen": 2745350,
      "step": 130,
      "time_per_iteration": 2.602620840072632
    },
    {
      "auxiliary_loss_clip": 0.01476459,
      "auxiliary_loss_mlp": 0.01066746,
      "balance_loss_clip": 1.23059487,
      "balance_loss_mlp": 1.05830622,
      "epoch": 0.015751818673720917,
      "flos": 62328536098560.0,
      "grad_norm": 1.076836062950782,
      "language_loss": 0.71515107,
      "learning_rate": 3.5318169562186737e-06,
      "loss": 0.74058312,
      "num_input_tokens_seen": 2814195,
      "step": 131,
      "time_per_iteration": 3.1902523040771484
    },
    {
      "auxiliary_loss_clip": 0.01488477,
      "auxiliary_loss_mlp": 0.01082521,
      "balance_loss_clip": 1.15625262,
      "balance_loss_mlp": 1.05627131,
      "epoch": 0.015872061564360006,
      "flos": 23878549365120.0,
      "grad_norm": 2.762149221162883,
      "language_loss": 0.82204282,
      "learning_rate": 3.5373260774446292e-06,
      "loss": 0.84775281,
      "num_input_tokens_seen": 2834645,
      "step": 132,
      "time_per_iteration": 2.5911712646484375
    },
    {
      "auxiliary_loss_clip": 0.01479214,
      "auxiliary_loss_mlp": 0.010884,
      "balance_loss_clip": 1.14925098,
      "balance_loss_mlp": 1.06374753,
      "epoch": 0.0159923044549991,
      "flos": 23367899664000.0,
      "grad_norm": 1.9345753706105984,
      "language_loss": 0.90561771,
      "learning_rate": 3.542793620000961e-06,
      "loss": 0.93129385,
      "num_input_tokens_seen": 2854120,
      "step": 133,
      "time_per_iteration": 2.5576682090759277
    },
    {
      "auxiliary_loss_clip": 0.01480723,
      "auxiliary_loss_mlp": 0.01073027,
      "balance_loss_clip": 1.15646648,
      "balance_loss_mlp": 1.05099702,
      "epoch": 0.01611254734563819,
      "flos": 17858305249920.0,
      "grad_norm": 2.733308464703463,
      "language_loss": 0.86870593,
      "learning_rate": 3.5482202067978894e-06,
      "loss": 0.89424342,
      "num_input_tokens_seen": 2871330,
      "step": 134,
      "time_per_iteration": 2.548170566558838
    },
    {
      "auxiliary_loss_clip": 0.01486711,
      "auxiliary_loss_mlp": 0.01078825,
      "balance_loss_clip": 1.15762949,
      "balance_loss_mlp": 1.05436289,
      "epoch": 0.01623279023627728,
      "flos": 20954774113920.0,
      "grad_norm": 1.927889197472661,
      "language_loss": 0.75770152,
      "learning_rate": 3.553606446851471e-06,
      "loss": 0.78335679,
      "num_input_tokens_seen": 2888070,
      "step": 135,
      "time_per_iteration": 2.5574491024017334
    },
    {
      "auxiliary_loss_clip": 0.01472967,
      "auxiliary_loss_mlp": 0.01080618,
      "balance_loss_clip": 1.15282369,
      "balance_loss_mlp": 1.05831432,
      "epoch": 0.016353033126916373,
      "flos": 15742412743680.0,
      "grad_norm": 1.799429396959038,
      "language_loss": 0.83424121,
      "learning_rate": 3.5589529356937613e-06,
      "loss": 0.85977703,
      "num_input_tokens_seen": 2906465,
      "step": 136,
      "time_per_iteration": 2.577397346496582
    },
    {
      "auxiliary_loss_clip": 0.0147336,
      "auxiliary_loss_mlp": 0.01088117,
      "balance_loss_clip": 1.15184355,
      "balance_loss_mlp": 1.06407273,
      "epoch": 0.016473276017555463,
      "flos": 18807280617600.0,
      "grad_norm": 1.7127651406520321,
      "language_loss": 0.76989675,
      "learning_rate": 3.5642602557679627e-06,
      "loss": 0.79551148,
      "num_input_tokens_seen": 2924915,
      "step": 137,
      "time_per_iteration": 2.5387141704559326
    },
    {
      "auxiliary_loss_clip": 0.01484932,
      "auxiliary_loss_mlp": 0.0108215,
      "balance_loss_clip": 1.16264665,
      "balance_loss_mlp": 1.06143165,
      "epoch": 0.016593518908194552,
      "flos": 24352641999360.0,
      "grad_norm": 2.2864913619152,
      "language_loss": 0.8439154,
      "learning_rate": 3.569528976809202e-06,
      "loss": 0.86958623,
      "num_input_tokens_seen": 2942130,
      "step": 138,
      "time_per_iteration": 2.602400779724121
    },
    {
      "auxiliary_loss_clip": 0.01470971,
      "auxiliary_loss_mlp": 0.01077691,
      "balance_loss_clip": 1.14954877,
      "balance_loss_mlp": 1.05433774,
      "epoch": 0.016713761798833646,
      "flos": 22346133384960.0,
      "grad_norm": 1.7369319450492473,
      "language_loss": 0.89912593,
      "learning_rate": 3.5747596562115522e-06,
      "loss": 0.92461246,
      "num_input_tokens_seen": 2962745,
      "step": 139,
      "time_per_iteration": 2.567641019821167
    },
    {
      "auxiliary_loss_clip": 0.0146965,
      "auxiliary_loss_mlp": 0.01073377,
      "balance_loss_clip": 1.14831567,
      "balance_loss_mlp": 1.04865289,
      "epoch": 0.016834004689472735,
      "flos": 17821820010240.0,
      "grad_norm": 2.6647306671690867,
      "language_loss": 0.90891987,
      "learning_rate": 3.5799528393819138e-06,
      "loss": 0.93435013,
      "num_input_tokens_seen": 2981825,
      "step": 140,
      "time_per_iteration": 2.5600266456604004
    },
    {
      "auxiliary_loss_clip": 0.01463805,
      "auxiliary_loss_mlp": 0.01083064,
      "balance_loss_clip": 1.14604521,
      "balance_loss_mlp": 1.05904377,
      "epoch": 0.016954247580111825,
      "flos": 20519501103360.0,
      "grad_norm": 1.830327226729161,
      "language_loss": 0.87970334,
      "learning_rate": 3.585109060081286e-06,
      "loss": 0.90517199,
      "num_input_tokens_seen": 3001625,
      "step": 141,
      "time_per_iteration": 2.5528087615966797
    },
    {
      "auxiliary_loss_clip": 0.01468021,
      "auxiliary_loss_mlp": 0.01083845,
      "balance_loss_clip": 1.14799333,
      "balance_loss_mlp": 1.06081367,
      "epoch": 0.017074490470750915,
      "flos": 22088869200000.0,
      "grad_norm": 4.323909885520045,
      "language_loss": 0.78534317,
      "learning_rate": 3.590228840753992e-06,
      "loss": 0.81086183,
      "num_input_tokens_seen": 3022055,
      "step": 142,
      "time_per_iteration": 2.6188130378723145
    },
    {
      "auxiliary_loss_clip": 0.01465754,
      "auxiliary_loss_mlp": 0.01074089,
      "balance_loss_clip": 1.1535064,
      "balance_loss_mlp": 1.05198812,
      "epoch": 0.01719473336139001,
      "flos": 15997270717440.0,
      "grad_norm": 2.2018648885133576,
      "language_loss": 0.87477881,
      "learning_rate": 3.5953126928453423e-06,
      "loss": 0.90017724,
      "num_input_tokens_seen": 3039605,
      "step": 143,
      "time_per_iteration": 2.5492570400238037
    },
    {
      "auxiliary_loss_clip": 0.01456476,
      "auxiliary_loss_mlp": 0.01075666,
      "balance_loss_clip": 1.1464963,
      "balance_loss_mlp": 1.05387425,
      "epoch": 0.017314976252029098,
      "flos": 22492038430080.0,
      "grad_norm": 1.9994867971396193,
      "language_loss": 0.80514395,
      "learning_rate": 3.600361117108239e-06,
      "loss": 0.83046538,
      "num_input_tokens_seen": 3059405,
      "step": 144,
      "time_per_iteration": 3.3276917934417725
    },
    {
      "auxiliary_loss_clip": 0.01458583,
      "auxiliary_loss_mlp": 0.01068948,
      "balance_loss_clip": 1.14614332,
      "balance_loss_mlp": 1.04710889,
      "epoch": 0.017435219142668188,
      "flos": 22018053536640.0,
      "grad_norm": 1.8683916270628007,
      "language_loss": 0.97108692,
      "learning_rate": 3.6053746038991616e-06,
      "loss": 0.99636221,
      "num_input_tokens_seen": 3078490,
      "step": 145,
      "time_per_iteration": 4.07668662071228
    },
    {
      "auxiliary_loss_clip": 0.01417444,
      "auxiliary_loss_mlp": 0.01055835,
      "balance_loss_clip": 1.19005215,
      "balance_loss_mlp": 1.04787219,
      "epoch": 0.01755546203330728,
      "flos": 72240526149120.0,
      "grad_norm": 1.0745818248686478,
      "language_loss": 0.58418965,
      "learning_rate": 3.6103536334639843e-06,
      "loss": 0.60892248,
      "num_input_tokens_seen": 3131755,
      "step": 146,
      "time_per_iteration": 3.801753044128418
    },
    {
      "auxiliary_loss_clip": 0.01450057,
      "auxiliary_loss_mlp": 0.01069313,
      "balance_loss_clip": 1.14471602,
      "balance_loss_mlp": 1.05012,
      "epoch": 0.01767570492394637,
      "flos": 25337061112320.0,
      "grad_norm": 2.9486675317738795,
      "language_loss": 0.85580921,
      "learning_rate": 3.615298676214041e-06,
      "loss": 0.8810029,
      "num_input_tokens_seen": 3152035,
      "step": 147,
      "time_per_iteration": 2.574413299560547
    },
    {
      "auxiliary_loss_clip": 0.01450675,
      "auxiliary_loss_mlp": 0.01078457,
      "balance_loss_clip": 1.13923168,
      "balance_loss_mlp": 1.05808425,
      "epoch": 0.01779594781458546,
      "flos": 20449188230400.0,
      "grad_norm": 2.112017758271063,
      "language_loss": 0.88778573,
      "learning_rate": 3.6202101929928317e-06,
      "loss": 0.91307712,
      "num_input_tokens_seen": 3170625,
      "step": 148,
      "time_per_iteration": 2.5598063468933105
    },
    {
      "auxiliary_loss_clip": 0.01447156,
      "auxiliary_loss_mlp": 0.01080671,
      "balance_loss_clip": 1.1403861,
      "balance_loss_mlp": 1.06025004,
      "epoch": 0.017916190705224554,
      "flos": 16253601148800.0,
      "grad_norm": 1.8918304167550135,
      "language_loss": 0.88464844,
      "learning_rate": 3.6250886353337413e-06,
      "loss": 0.90992671,
      "num_input_tokens_seen": 3188155,
      "step": 149,
      "time_per_iteration": 2.5236122608184814
    },
    {
      "auxiliary_loss_clip": 0.01449003,
      "auxiliary_loss_mlp": 0.01080387,
      "balance_loss_clip": 1.14243829,
      "balance_loss_mlp": 1.06026459,
      "epoch": 0.018036433595863644,
      "flos": 23330588411520.0,
      "grad_norm": 1.8209729436042834,
      "language_loss": 0.86479127,
      "learning_rate": 3.6299344457091488e-06,
      "loss": 0.89008522,
      "num_input_tokens_seen": 3209015,
      "step": 150,
      "time_per_iteration": 2.5801820755004883
    },
    {
      "auxiliary_loss_clip": 0.01445774,
      "auxiliary_loss_mlp": 0.01068364,
      "balance_loss_clip": 1.14245319,
      "balance_loss_mlp": 1.04831314,
      "epoch": 0.018156676486502734,
      "flos": 18588010043520.0,
      "grad_norm": 2.2351663049974424,
      "language_loss": 0.93964016,
      "learning_rate": 3.634748057771256e-06,
      "loss": 0.96478152,
      "num_input_tokens_seen": 3224955,
      "step": 151,
      "time_per_iteration": 2.539951801300049
    },
    {
      "auxiliary_loss_clip": 0.01435242,
      "auxiliary_loss_mlp": 0.01074169,
      "balance_loss_clip": 1.13546216,
      "balance_loss_mlp": 1.0544163,
      "epoch": 0.018276919377141827,
      "flos": 25448707560960.0,
      "grad_norm": 1.6348035898079096,
      "language_loss": 0.85596395,
      "learning_rate": 3.639529896584965e-06,
      "loss": 0.88105798,
      "num_input_tokens_seen": 3246330,
      "step": 152,
      "time_per_iteration": 2.613234519958496
    },
    {
      "auxiliary_loss_clip": 0.01443738,
      "auxiliary_loss_mlp": 0.01067455,
      "balance_loss_clip": 1.14062595,
      "balance_loss_mlp": 1.04634309,
      "epoch": 0.018397162267780917,
      "flos": 20047311889920.0,
      "grad_norm": 2.606444504678349,
      "language_loss": 0.88838601,
      "learning_rate": 3.6442803788531233e-06,
      "loss": 0.91349792,
      "num_input_tokens_seen": 3264290,
      "step": 153,
      "time_per_iteration": 2.563673734664917
    },
    {
      "auxiliary_loss_clip": 0.01445411,
      "auxiliary_loss_mlp": 0.01078054,
      "balance_loss_clip": 1.1394124,
      "balance_loss_mlp": 1.05768085,
      "epoch": 0.018517405158420007,
      "flos": 27565282425600.0,
      "grad_norm": 2.060910793302648,
      "language_loss": 0.96093982,
      "learning_rate": 3.6489999131344357e-06,
      "loss": 0.98617446,
      "num_input_tokens_seen": 3287065,
      "step": 154,
      "time_per_iteration": 2.63460373878479
    },
    {
      "auxiliary_loss_clip": 0.01438094,
      "auxiliary_loss_mlp": 0.01067942,
      "balance_loss_clip": 1.14096057,
      "balance_loss_mlp": 1.05000114,
      "epoch": 0.0186376480490591,
      "flos": 19354056422400.0,
      "grad_norm": 2.122365780209798,
      "language_loss": 0.90655541,
      "learning_rate": 3.653688900054313e-06,
      "loss": 0.93161583,
      "num_input_tokens_seen": 3305595,
      "step": 155,
      "time_per_iteration": 2.568570613861084
    },
    {
      "auxiliary_loss_clip": 0.01433327,
      "auxiliary_loss_mlp": 0.01071085,
      "balance_loss_clip": 1.13061881,
      "balance_loss_mlp": 1.05038989,
      "epoch": 0.01875789093969819,
      "flos": 26687840993280.0,
      "grad_norm": 2.2629009078284237,
      "language_loss": 0.76199782,
      "learning_rate": 3.6583477325089526e-06,
      "loss": 0.78704196,
      "num_input_tokens_seen": 3326135,
      "step": 156,
      "time_per_iteration": 2.62390398979187
    },
    {
      "auxiliary_loss_clip": 0.0143138,
      "auxiliary_loss_mlp": 0.01070991,
      "balance_loss_clip": 1.13526893,
      "balance_loss_mlp": 1.05366945,
      "epoch": 0.01887813383033728,
      "flos": 24353001135360.0,
      "grad_norm": 2.3531410029512436,
      "language_loss": 1.04149842,
      "learning_rate": 3.6629767958628916e-06,
      "loss": 1.066522,
      "num_input_tokens_seen": 3343510,
      "step": 157,
      "time_per_iteration": 2.6013338565826416
    },
    {
      "auxiliary_loss_clip": 0.01432496,
      "auxiliary_loss_mlp": 0.01069986,
      "balance_loss_clip": 1.13698876,
      "balance_loss_mlp": 1.04974377,
      "epoch": 0.018998376720976373,
      "flos": 14647532330880.0,
      "grad_norm": 2.637442910487536,
      "language_loss": 0.85747558,
      "learning_rate": 3.667576468140291e-06,
      "loss": 0.88250035,
      "num_input_tokens_seen": 3361325,
      "step": 158,
      "time_per_iteration": 2.626528024673462
    },
    {
      "auxiliary_loss_clip": 0.01424745,
      "auxiliary_loss_mlp": 0.01067383,
      "balance_loss_clip": 1.12962055,
      "balance_loss_mlp": 1.04751086,
      "epoch": 0.019118619611615463,
      "flos": 29305261146240.0,
      "grad_norm": 2.4630778475537918,
      "language_loss": 0.88749409,
      "learning_rate": 3.672147120210184e-06,
      "loss": 0.91241533,
      "num_input_tokens_seen": 3377925,
      "step": 159,
      "time_per_iteration": 2.6341779232025146
    },
    {
      "auxiliary_loss_clip": 0.01428859,
      "auxiliary_loss_mlp": 0.01075031,
      "balance_loss_clip": 1.13594997,
      "balance_loss_mlp": 1.05729234,
      "epoch": 0.019238862502254553,
      "flos": 20886723797760.0,
      "grad_norm": 1.797663650971881,
      "language_loss": 0.8628124,
      "learning_rate": 3.6766891159659177e-06,
      "loss": 0.88785136,
      "num_input_tokens_seen": 3396335,
      "step": 160,
      "time_per_iteration": 2.596874475479126
    },
    {
      "auxiliary_loss_clip": 0.01421546,
      "auxiliary_loss_mlp": 0.0106988,
      "balance_loss_clip": 1.13127685,
      "balance_loss_mlp": 1.05115223,
      "epoch": 0.019359105392893646,
      "flos": 21360672777600.0,
      "grad_norm": 2.6273400874916653,
      "language_loss": 0.8772254,
      "learning_rate": 3.6812028124990075e-06,
      "loss": 0.90213966,
      "num_input_tokens_seen": 3413605,
      "step": 161,
      "time_per_iteration": 2.557422637939453
    },
    {
      "auxiliary_loss_clip": 0.0141918,
      "auxiliary_loss_mlp": 0.01073971,
      "balance_loss_clip": 1.12891114,
      "balance_loss_mlp": 1.05548131,
      "epoch": 0.019479348283532736,
      "flos": 16283729681280.0,
      "grad_norm": 2.8986242324716294,
      "language_loss": 0.81670123,
      "learning_rate": 3.6856885602676016e-06,
      "loss": 0.84163272,
      "num_input_tokens_seen": 3429640,
      "step": 162,
      "time_per_iteration": 2.571394205093384
    },
    {
      "auxiliary_loss_clip": 0.01430632,
      "auxiliary_loss_mlp": 0.01069152,
      "balance_loss_clip": 1.13442016,
      "balance_loss_mlp": 1.04999518,
      "epoch": 0.019599591174171826,
      "flos": 22091239497600.0,
      "grad_norm": 3.159001397831375,
      "language_loss": 0.94293463,
      "learning_rate": 3.6901467032597733e-06,
      "loss": 0.96793246,
      "num_input_tokens_seen": 3448125,
      "step": 163,
      "time_per_iteration": 2.5990772247314453
    },
    {
      "auxiliary_loss_clip": 0.01423009,
      "auxiliary_loss_mlp": 0.01070837,
      "balance_loss_clip": 1.12974906,
      "balance_loss_mlp": 1.051072,
      "epoch": 0.01971983406481092,
      "flos": 19609668581760.0,
      "grad_norm": 2.119640135879872,
      "language_loss": 0.8749615,
      "learning_rate": 3.694577579151804e-06,
      "loss": 0.89989996,
      "num_input_tokens_seen": 3466535,
      "step": 164,
      "time_per_iteration": 2.6559078693389893
    },
    {
      "auxiliary_loss_clip": 0.01419591,
      "auxiliary_loss_mlp": 0.01077024,
      "balance_loss_clip": 1.13154387,
      "balance_loss_mlp": 1.05790305,
      "epoch": 0.01984007695545001,
      "flos": 19099342103040.0,
      "grad_norm": 24.524710308639005,
      "language_loss": 0.73692065,
      "learning_rate": 3.6989815194616703e-06,
      "loss": 0.76188684,
      "num_input_tokens_seen": 3483730,
      "step": 165,
      "time_per_iteration": 2.657019853591919
    },
    {
      "auxiliary_loss_clip": 0.01428943,
      "auxiliary_loss_mlp": 0.01066098,
      "balance_loss_clip": 1.13362384,
      "balance_loss_mlp": 1.04690564,
      "epoch": 0.0199603198460891,
      "flos": 20848406964480.0,
      "grad_norm": 2.398816831753953,
      "language_loss": 0.79931611,
      "learning_rate": 3.703358849697888e-06,
      "loss": 0.82426655,
      "num_input_tokens_seen": 3503640,
      "step": 166,
      "time_per_iteration": 2.667200803756714
    },
    {
      "auxiliary_loss_clip": 0.01418677,
      "auxiliary_loss_mlp": 0.01068854,
      "balance_loss_clip": 1.13120079,
      "balance_loss_mlp": 1.0514617,
      "epoch": 0.020080562736728192,
      "flos": 21870747861120.0,
      "grad_norm": 3.089628923113304,
      "language_loss": 0.82608318,
      "learning_rate": 3.7077098895038803e-06,
      "loss": 0.85095853,
      "num_input_tokens_seen": 3523010,
      "step": 167,
      "time_per_iteration": 2.65098237991333
    },
    {
      "auxiliary_loss_clip": 0.01422312,
      "auxiliary_loss_mlp": 0.0106119,
      "balance_loss_clip": 1.13055468,
      "balance_loss_mlp": 1.0426048,
      "epoch": 0.020200805627367282,
      "flos": 21688788539520.0,
      "grad_norm": 2.007966795344019,
      "language_loss": 0.97143447,
      "learning_rate": 3.712034952798045e-06,
      "loss": 0.99626952,
      "num_input_tokens_seen": 3541125,
      "step": 168,
      "time_per_iteration": 2.5685195922851562
    },
    {
      "auxiliary_loss_clip": 0.01422459,
      "auxiliary_loss_mlp": 0.01071358,
      "balance_loss_clip": 1.1298666,
      "balance_loss_mlp": 1.05193865,
      "epoch": 0.02032104851800637,
      "flos": 33543043729920.0,
      "grad_norm": 2.2190764627820996,
      "language_loss": 0.8470583,
      "learning_rate": 3.7163343479096656e-06,
      "loss": 0.87199652,
      "num_input_tokens_seen": 3562700,
      "step": 169,
      "time_per_iteration": 2.702793836593628
    },
    {
      "auxiliary_loss_clip": 0.01405235,
      "auxiliary_loss_mlp": 0.01071161,
      "balance_loss_clip": 1.12324238,
      "balance_loss_mlp": 1.05370879,
      "epoch": 0.020441291408645465,
      "flos": 31686965274240.0,
      "grad_norm": 4.002684992177676,
      "language_loss": 0.82808733,
      "learning_rate": 3.720608377710802e-06,
      "loss": 0.85285127,
      "num_input_tokens_seen": 3582790,
      "step": 170,
      "time_per_iteration": 3.3532819747924805
    },
    {
      "auxiliary_loss_clip": 0.01410675,
      "auxiliary_loss_mlp": 0.01068737,
      "balance_loss_clip": 1.12391329,
      "balance_loss_mlp": 1.04986596,
      "epoch": 0.020561534299284555,
      "flos": 20886687884160.0,
      "grad_norm": 4.945543180689287,
      "language_loss": 0.86576974,
      "learning_rate": 3.7248573397443277e-06,
      "loss": 0.89056385,
      "num_input_tokens_seen": 3601715,
      "step": 171,
      "time_per_iteration": 2.631410837173462
    },
    {
      "auxiliary_loss_clip": 0.0141607,
      "auxiliary_loss_mlp": 0.01060235,
      "balance_loss_clip": 1.13143539,
      "balance_loss_mlp": 1.04336691,
      "epoch": 0.020681777189923645,
      "flos": 20996610480000.0,
      "grad_norm": 2.021790414086321,
      "language_loss": 0.97713482,
      "learning_rate": 3.729081526348224e-06,
      "loss": 1.00189781,
      "num_input_tokens_seen": 3620245,
      "step": 172,
      "time_per_iteration": 4.148843765258789
    },
    {
      "auxiliary_loss_clip": 0.01408567,
      "auxiliary_loss_mlp": 0.01066341,
      "balance_loss_clip": 1.12354517,
      "balance_loss_mlp": 1.04916263,
      "epoch": 0.020802020080562738,
      "flos": 28257532312320.0,
      "grad_norm": 1.8095065813638709,
      "language_loss": 0.84940994,
      "learning_rate": 3.7332812247762777e-06,
      "loss": 0.87415898,
      "num_input_tokens_seen": 3641545,
      "step": 173,
      "time_per_iteration": 3.777919292449951
    },
    {
      "auxiliary_loss_clip": 0.01411052,
      "auxiliary_loss_mlp": 0.01069095,
      "balance_loss_clip": 1.12820864,
      "balance_loss_mlp": 1.05124903,
      "epoch": 0.020922262971201828,
      "flos": 19681274344320.0,
      "grad_norm": 2.2808659113988505,
      "language_loss": 0.95120305,
      "learning_rate": 3.737456717315293e-06,
      "loss": 0.97600454,
      "num_input_tokens_seen": 3660510,
      "step": 174,
      "time_per_iteration": 2.6159298419952393
    },
    {
      "auxiliary_loss_clip": 0.01401879,
      "auxiliary_loss_mlp": 0.01077649,
      "balance_loss_clip": 1.12322628,
      "balance_loss_mlp": 1.05982709,
      "epoch": 0.021042505861840918,
      "flos": 15666353694720.0,
      "grad_norm": 1.7428217854056662,
      "language_loss": 0.90690953,
      "learning_rate": 3.7416082813989552e-06,
      "loss": 0.93170476,
      "num_input_tokens_seen": 3677505,
      "step": 175,
      "time_per_iteration": 2.648185968399048
    },
    {
      "auxiliary_loss_clip": 0.01407783,
      "auxiliary_loss_mlp": 0.01066788,
      "balance_loss_clip": 1.12161112,
      "balance_loss_mlp": 1.04852498,
      "epoch": 0.02116274875248001,
      "flos": 21142012734720.0,
      "grad_norm": 1.9710511398244437,
      "language_loss": 0.89327276,
      "learning_rate": 3.745736189718439e-06,
      "loss": 0.91801846,
      "num_input_tokens_seen": 3696760,
      "step": 176,
      "time_per_iteration": 2.5951342582702637
    },
    {
      "auxiliary_loss_clip": 0.01397411,
      "auxiliary_loss_mlp": 0.01063139,
      "balance_loss_clip": 1.11888957,
      "balance_loss_mlp": 1.04647326,
      "epoch": 0.0212829916431191,
      "flos": 24715770543360.0,
      "grad_norm": 6.055573310511585,
      "language_loss": 0.72492576,
      "learning_rate": 3.749840710329894e-06,
      "loss": 0.74953127,
      "num_input_tokens_seen": 3717465,
      "step": 177,
      "time_per_iteration": 2.589200258255005
    },
    {
      "auxiliary_loss_clip": 0.01406321,
      "auxiliary_loss_mlp": 0.01063112,
      "balance_loss_clip": 1.12352002,
      "balance_loss_mlp": 1.04408622,
      "epoch": 0.02140323453375819,
      "flos": 16645493508480.0,
      "grad_norm": 2.6676046724362608,
      "language_loss": 0.98060441,
      "learning_rate": 3.7539221067588938e-06,
      "loss": 1.00529873,
      "num_input_tokens_seen": 3731440,
      "step": 178,
      "time_per_iteration": 2.5173823833465576
    },
    {
      "auxiliary_loss_clip": 0.0140629,
      "auxiliary_loss_mlp": 0.01084874,
      "balance_loss_clip": 1.12297571,
      "balance_loss_mlp": 1.06589556,
      "epoch": 0.021523477424397284,
      "flos": 20299332689280.0,
      "grad_norm": 5.430752276085542,
      "language_loss": 0.93840122,
      "learning_rate": 3.757980638101964e-06,
      "loss": 0.96331292,
      "num_input_tokens_seen": 3744935,
      "step": 179,
      "time_per_iteration": 2.5842807292938232
    },
    {
      "auxiliary_loss_clip": 0.01403066,
      "auxiliary_loss_mlp": 0.01067692,
      "balance_loss_clip": 1.1224407,
      "balance_loss_mlp": 1.04937005,
      "epoch": 0.021643720315036374,
      "flos": 26104005331200.0,
      "grad_norm": 2.0669926176854263,
      "language_loss": 0.89775437,
      "learning_rate": 3.7620165591252806e-06,
      "loss": 0.92246187,
      "num_input_tokens_seen": 3763035,
      "step": 180,
      "time_per_iteration": 2.5774319171905518
    },
    {
      "auxiliary_loss_clip": 0.01400148,
      "auxiliary_loss_mlp": 0.0107207,
      "balance_loss_clip": 1.12314165,
      "balance_loss_mlp": 1.05657303,
      "epoch": 0.021763963205675464,
      "flos": 24787663614720.0,
      "grad_norm": 2.0648149266139626,
      "language_loss": 0.94337201,
      "learning_rate": 3.766030120360636e-06,
      "loss": 0.96809411,
      "num_input_tokens_seen": 3782665,
      "step": 181,
      "time_per_iteration": 2.6220462322235107
    },
    {
      "auxiliary_loss_clip": 0.01398727,
      "auxiliary_loss_mlp": 0.01060231,
      "balance_loss_clip": 1.11917794,
      "balance_loss_mlp": 1.04252851,
      "epoch": 0.021884206096314557,
      "flos": 25813559957760.0,
      "grad_norm": 1.9933825787311061,
      "language_loss": 0.9031933,
      "learning_rate": 3.7700215681987578e-06,
      "loss": 0.92778289,
      "num_input_tokens_seen": 3802435,
      "step": 182,
      "time_per_iteration": 2.568166494369507
    },
    {
      "auxiliary_loss_clip": 0.01397827,
      "auxiliary_loss_mlp": 0.01063926,
      "balance_loss_clip": 1.12121618,
      "balance_loss_mlp": 1.04586601,
      "epoch": 0.022004448986953647,
      "flos": 20082719721600.0,
      "grad_norm": 1.7083174013032125,
      "language_loss": 0.82206285,
      "learning_rate": 3.7739911449800767e-06,
      "loss": 0.84668034,
      "num_input_tokens_seen": 3822490,
      "step": 183,
      "time_per_iteration": 2.6015920639038086
    },
    {
      "auxiliary_loss_clip": 0.01395285,
      "auxiliary_loss_mlp": 0.0106668,
      "balance_loss_clip": 1.11631227,
      "balance_loss_mlp": 1.04956162,
      "epoch": 0.022124691877592736,
      "flos": 20480609652480.0,
      "grad_norm": 3.3593239422888828,
      "language_loss": 0.80787718,
      "learning_rate": 3.7779390890830114e-06,
      "loss": 0.83249688,
      "num_input_tokens_seen": 3841140,
      "step": 184,
      "time_per_iteration": 2.541952133178711
    },
    {
      "auxiliary_loss_clip": 0.01401212,
      "auxiliary_loss_mlp": 0.01063504,
      "balance_loss_clip": 1.11930692,
      "balance_loss_mlp": 1.04748225,
      "epoch": 0.02224493476823183,
      "flos": 23586847015680.0,
      "grad_norm": 1.9649897730512753,
      "language_loss": 0.85946041,
      "learning_rate": 3.7818656350098723e-06,
      "loss": 0.88410759,
      "num_input_tokens_seen": 3862090,
      "step": 185,
      "time_per_iteration": 2.633087158203125
    },
    {
      "auxiliary_loss_clip": 0.01394853,
      "auxiliary_loss_mlp": 0.01067037,
      "balance_loss_clip": 1.11646259,
      "balance_loss_mlp": 1.04962111,
      "epoch": 0.02236517765887092,
      "flos": 16909940413440.0,
      "grad_norm": 2.5645972941324366,
      "language_loss": 0.77138811,
      "learning_rate": 3.7857710134704447e-06,
      "loss": 0.79600704,
      "num_input_tokens_seen": 3881025,
      "step": 186,
      "time_per_iteration": 2.531020164489746
    },
    {
      "auxiliary_loss_clip": 0.0138888,
      "auxiliary_loss_mlp": 0.01068104,
      "balance_loss_clip": 1.11589277,
      "balance_loss_mlp": 1.05022252,
      "epoch": 0.02248542054951001,
      "flos": 43508182930560.0,
      "grad_norm": 2.2528827725049276,
      "language_loss": 0.79152197,
      "learning_rate": 3.7896554514633234e-06,
      "loss": 0.81609178,
      "num_input_tokens_seen": 3905310,
      "step": 187,
      "time_per_iteration": 2.7522711753845215
    },
    {
      "auxiliary_loss_clip": 0.01384519,
      "auxiliary_loss_mlp": 0.01059279,
      "balance_loss_clip": 1.11232233,
      "balance_loss_mlp": 1.04256582,
      "epoch": 0.022605663440149103,
      "flos": 23367648268800.0,
      "grad_norm": 2.40896698141146,
      "language_loss": 0.83972859,
      "learning_rate": 3.7935191723550955e-06,
      "loss": 0.86416656,
      "num_input_tokens_seen": 3924265,
      "step": 188,
      "time_per_iteration": 2.551440715789795
    },
    {
      "auxiliary_loss_clip": 0.01380961,
      "auxiliary_loss_mlp": 0.01057129,
      "balance_loss_clip": 1.11154723,
      "balance_loss_mlp": 1.04225135,
      "epoch": 0.022725906330788193,
      "flos": 29019915504000.0,
      "grad_norm": 2.202555841232679,
      "language_loss": 0.88383079,
      "learning_rate": 3.797362395957408e-06,
      "loss": 0.90821171,
      "num_input_tokens_seen": 3944830,
      "step": 189,
      "time_per_iteration": 2.626603603363037
    },
    {
      "auxiliary_loss_clip": 0.01389374,
      "auxiliary_loss_mlp": 0.01060635,
      "balance_loss_clip": 1.11681533,
      "balance_loss_mlp": 1.04280114,
      "epoch": 0.022846149221427282,
      "flos": 24496176746880.0,
      "grad_norm": 4.161102080520976,
      "language_loss": 0.78231037,
      "learning_rate": 3.8011853386020055e-06,
      "loss": 0.80681044,
      "num_input_tokens_seen": 3965735,
      "step": 190,
      "time_per_iteration": 2.5811970233917236
    },
    {
      "auxiliary_loss_clip": 0.01384862,
      "auxiliary_loss_mlp": 0.01057414,
      "balance_loss_clip": 1.1134975,
      "balance_loss_mlp": 1.03993785,
      "epoch": 0.022966392112066376,
      "flos": 15523537219200.0,
      "grad_norm": 2.540935842084887,
      "language_loss": 0.89846992,
      "learning_rate": 3.804988213213804e-06,
      "loss": 0.92289269,
      "num_input_tokens_seen": 3983975,
      "step": 191,
      "time_per_iteration": 2.5411465167999268
    },
    {
      "auxiliary_loss_clip": 0.01355599,
      "auxiliary_loss_mlp": 0.01018091,
      "balance_loss_clip": 1.15855932,
      "balance_loss_mlp": 1.0087446,
      "epoch": 0.023086635002705466,
      "flos": 55650408433920.0,
      "grad_norm": 1.0216338981293598,
      "language_loss": 0.63189375,
      "learning_rate": 3.808771229382049e-06,
      "loss": 0.65563071,
      "num_input_tokens_seen": 4043440,
      "step": 192,
      "time_per_iteration": 3.0271992683410645
    },
    {
      "auxiliary_loss_clip": 0.0137893,
      "auxiliary_loss_mlp": 0.01052618,
      "balance_loss_clip": 1.10847664,
      "balance_loss_mlp": 1.03685844,
      "epoch": 0.023206877893344555,
      "flos": 19313441118720.0,
      "grad_norm": 2.026556519369347,
      "language_loss": 0.84456909,
      "learning_rate": 3.8125345934296324e-06,
      "loss": 0.86888462,
      "num_input_tokens_seen": 4061750,
      "step": 193,
      "time_per_iteration": 2.5397098064422607
    },
    {
      "auxiliary_loss_clip": 0.01387198,
      "auxiliary_loss_mlp": 0.0106634,
      "balance_loss_clip": 1.11238658,
      "balance_loss_mlp": 1.0498054,
      "epoch": 0.02332712078398365,
      "flos": 23072965090560.0,
      "grad_norm": 2.117772883896705,
      "language_loss": 0.8771137,
      "learning_rate": 3.81627850848061e-06,
      "loss": 0.90164912,
      "num_input_tokens_seen": 4082345,
      "step": 194,
      "time_per_iteration": 2.5548768043518066
    },
    {
      "auxiliary_loss_clip": 0.01377763,
      "auxiliary_loss_mlp": 0.01070273,
      "balance_loss_clip": 1.10996997,
      "balance_loss_mlp": 1.05549145,
      "epoch": 0.02344736367462274,
      "flos": 24425971614720.0,
      "grad_norm": 2.175690275204565,
      "language_loss": 0.86286664,
      "learning_rate": 3.820003174525994e-06,
      "loss": 0.88734704,
      "num_input_tokens_seen": 4101770,
      "step": 195,
      "time_per_iteration": 2.577684164047241
    },
    {
      "auxiliary_loss_clip": 0.01372954,
      "auxiliary_loss_mlp": 0.0105794,
      "balance_loss_clip": 1.10774279,
      "balance_loss_mlp": 1.04268706,
      "epoch": 0.02356760656526183,
      "flos": 21579799697280.0,
      "grad_norm": 2.380772844356794,
      "language_loss": 0.83011055,
      "learning_rate": 3.823708788487851e-06,
      "loss": 0.85441947,
      "num_input_tokens_seen": 4118770,
      "step": 196,
      "time_per_iteration": 2.514930486679077
    },
    {
      "auxiliary_loss_clip": 0.01379229,
      "auxiliary_loss_mlp": 0.01074726,
      "balance_loss_clip": 1.1118772,
      "balance_loss_mlp": 1.05888867,
      "epoch": 0.02368784945590092,
      "flos": 25193598192000.0,
      "grad_norm": 1.888798046041017,
      "language_loss": 0.84661829,
      "learning_rate": 3.827395544281781e-06,
      "loss": 0.87115788,
      "num_input_tokens_seen": 4141110,
      "step": 197,
      "time_per_iteration": 3.3496015071868896
    },
    {
      "auxiliary_loss_clip": 0.0137764,
      "auxiliary_loss_mlp": 0.01059592,
      "balance_loss_clip": 1.11045456,
      "balance_loss_mlp": 1.04311728,
      "epoch": 0.02380809234654001,
      "flos": 27562481164800.0,
      "grad_norm": 2.0412373988252392,
      "language_loss": 0.78996813,
      "learning_rate": 3.831063632877802e-06,
      "loss": 0.81434047,
      "num_input_tokens_seen": 4161430,
      "step": 198,
      "time_per_iteration": 2.620748281478882
    },
    {
      "auxiliary_loss_clip": 0.01373728,
      "auxiliary_loss_mlp": 0.01060566,
      "balance_loss_clip": 1.11249316,
      "balance_loss_mlp": 1.04548562,
      "epoch": 0.0239283352371791,
      "flos": 18259786540800.0,
      "grad_norm": 2.4973835053823414,
      "language_loss": 0.75715494,
      "learning_rate": 3.834713242359712e-06,
      "loss": 0.78149784,
      "num_input_tokens_seen": 4179260,
      "step": 199,
      "time_per_iteration": 4.188012599945068
    },
    {
      "auxiliary_loss_clip": 0.01372148,
      "auxiliary_loss_mlp": 0.01055516,
      "balance_loss_clip": 1.10652089,
      "balance_loss_mlp": 1.04019213,
      "epoch": 0.02404857812781819,
      "flos": 21395110942080.0,
      "grad_norm": 1.8406229148544657,
      "language_loss": 0.87191784,
      "learning_rate": 3.838344557982959e-06,
      "loss": 0.89619452,
      "num_input_tokens_seen": 4200640,
      "step": 200,
      "time_per_iteration": 2.5780439376831055
    },
    {
      "auxiliary_loss_clip": 0.01367217,
      "auxiliary_loss_mlp": 0.01048915,
      "balance_loss_clip": 1.10380507,
      "balance_loss_mlp": 1.03335786,
      "epoch": 0.024168821018457284,
      "flos": 16654256426880.0,
      "grad_norm": 2.5556274138500843,
      "language_loss": 0.84963012,
      "learning_rate": 3.841957762231063e-06,
      "loss": 0.87379134,
      "num_input_tokens_seen": 4218170,
      "step": 201,
      "time_per_iteration": 2.53367018699646
    },
    {
      "auxiliary_loss_clip": 0.01367393,
      "auxiliary_loss_mlp": 0.01065034,
      "balance_loss_clip": 1.10361958,
      "balance_loss_mlp": 1.04995394,
      "epoch": 0.024289063909096374,
      "flos": 22820872464000.0,
      "grad_norm": 2.372302389341696,
      "language_loss": 0.87688529,
      "learning_rate": 3.8455530348706454e-06,
      "loss": 0.90120959,
      "num_input_tokens_seen": 4237770,
      "step": 202,
      "time_per_iteration": 2.5297961235046387
    },
    {
      "auxiliary_loss_clip": 0.01367834,
      "auxiliary_loss_mlp": 0.01056776,
      "balance_loss_clip": 1.10619771,
      "balance_loss_mlp": 1.04123139,
      "epoch": 0.024409306799735464,
      "flos": 17748598135680.0,
      "grad_norm": 2.291040841664121,
      "language_loss": 0.77661228,
      "learning_rate": 3.849130553005099e-06,
      "loss": 0.80085838,
      "num_input_tokens_seen": 4255985,
      "step": 203,
      "time_per_iteration": 2.510061025619507
    },
    {
      "auxiliary_loss_clip": 0.01370629,
      "auxiliary_loss_mlp": 0.01058855,
      "balance_loss_clip": 1.10799456,
      "balance_loss_mlp": 1.04173672,
      "epoch": 0.024529549690374557,
      "flos": 21616213109760.0,
      "grad_norm": 1.6852667136055821,
      "language_loss": 0.83667123,
      "learning_rate": 3.852690491126933e-06,
      "loss": 0.86096609,
      "num_input_tokens_seen": 4276035,
      "step": 204,
      "time_per_iteration": 2.5535197257995605
    },
    {
      "auxiliary_loss_clip": 0.01369096,
      "auxiliary_loss_mlp": 0.01065778,
      "balance_loss_clip": 1.10478115,
      "balance_loss_mlp": 1.04926753,
      "epoch": 0.024649792581013647,
      "flos": 25551662918400.0,
      "grad_norm": 2.269993467498683,
      "language_loss": 0.91096079,
      "learning_rate": 3.856233021168845e-06,
      "loss": 0.93530953,
      "num_input_tokens_seen": 4295730,
      "step": 205,
      "time_per_iteration": 2.6161463260650635
    },
    {
      "auxiliary_loss_clip": 0.01361517,
      "auxiliary_loss_mlp": 0.01053096,
      "balance_loss_clip": 1.10084772,
      "balance_loss_mlp": 1.03875542,
      "epoch": 0.024770035471652737,
      "flos": 34495574544000.0,
      "grad_norm": 2.1339780736405505,
      "language_loss": 0.91457516,
      "learning_rate": 3.859758312553544e-06,
      "loss": 0.9387213,
      "num_input_tokens_seen": 4317950,
      "step": 206,
      "time_per_iteration": 2.6812021732330322
    },
    {
      "auxiliary_loss_clip": 0.01367583,
      "auxiliary_loss_mlp": 0.01062321,
      "balance_loss_clip": 1.10557389,
      "balance_loss_mlp": 1.04638267,
      "epoch": 0.02489027836229183,
      "flos": 21505428587520.0,
      "grad_norm": 1.7928459392664282,
      "language_loss": 0.91850138,
      "learning_rate": 3.8632665322423735e-06,
      "loss": 0.9428004,
      "num_input_tokens_seen": 4337605,
      "step": 207,
      "time_per_iteration": 2.5833051204681396
    },
    {
      "auxiliary_loss_clip": 0.01362237,
      "auxiliary_loss_mlp": 0.01078059,
      "balance_loss_clip": 1.10268044,
      "balance_loss_mlp": 1.0629549,
      "epoch": 0.02501052125293092,
      "flos": 23219013790080.0,
      "grad_norm": 1.647325932826466,
      "language_loss": 0.8595947,
      "learning_rate": 3.866757844782762e-06,
      "loss": 0.88399762,
      "num_input_tokens_seen": 4358110,
      "step": 208,
      "time_per_iteration": 2.582639217376709
    },
    {
      "auxiliary_loss_clip": 0.01360574,
      "auxiliary_loss_mlp": 0.01072155,
      "balance_loss_clip": 1.10277367,
      "balance_loss_mlp": 1.05597878,
      "epoch": 0.02513076414357001,
      "flos": 26388920010240.0,
      "grad_norm": 2.4193652798123213,
      "language_loss": 0.91475713,
      "learning_rate": 3.870232412354527e-06,
      "loss": 0.93908441,
      "num_input_tokens_seen": 4374955,
      "step": 209,
      "time_per_iteration": 2.6014459133148193
    },
    {
      "auxiliary_loss_clip": 0.01362131,
      "auxiliary_loss_mlp": 0.01055153,
      "balance_loss_clip": 1.10086262,
      "balance_loss_mlp": 1.04022813,
      "epoch": 0.025251007034209103,
      "flos": 13590430047360.0,
      "grad_norm": 2.3922685349550874,
      "language_loss": 0.92197978,
      "learning_rate": 3.873690394815086e-06,
      "loss": 0.94615269,
      "num_input_tokens_seen": 4391535,
      "step": 210,
      "time_per_iteration": 2.541599988937378
    },
    {
      "auxiliary_loss_clip": 0.01360421,
      "auxiliary_loss_mlp": 0.01059907,
      "balance_loss_clip": 1.10148454,
      "balance_loss_mlp": 1.04430223,
      "epoch": 0.025371249924848193,
      "flos": 15049229103360.0,
      "grad_norm": 2.979794761301918,
      "language_loss": 0.91024959,
      "learning_rate": 3.877131949743587e-06,
      "loss": 0.93445277,
      "num_input_tokens_seen": 4408400,
      "step": 211,
      "time_per_iteration": 2.559276819229126
    },
    {
      "auxiliary_loss_clip": 0.01358608,
      "auxiliary_loss_mlp": 0.01064336,
      "balance_loss_clip": 1.10031843,
      "balance_loss_mlp": 1.04845762,
      "epoch": 0.025491492815487283,
      "flos": 25553853648000.0,
      "grad_norm": 3.1962614535762737,
      "language_loss": 0.78099656,
      "learning_rate": 3.880557232483993e-06,
      "loss": 0.80522597,
      "num_input_tokens_seen": 4427840,
      "step": 212,
      "time_per_iteration": 2.5640130043029785
    },
    {
      "auxiliary_loss_clip": 0.01353685,
      "auxiliary_loss_mlp": 0.01054975,
      "balance_loss_clip": 1.09528208,
      "balance_loss_mlp": 1.03931117,
      "epoch": 0.025611735706126376,
      "flos": 20630752502400.0,
      "grad_norm": 1.9011805707700036,
      "language_loss": 0.86806357,
      "learning_rate": 3.883966396187164e-06,
      "loss": 0.89215016,
      "num_input_tokens_seen": 4447110,
      "step": 213,
      "time_per_iteration": 2.551992654800415
    },
    {
      "auxiliary_loss_clip": 0.01357932,
      "auxiliary_loss_mlp": 0.01055588,
      "balance_loss_clip": 1.10243177,
      "balance_loss_mlp": 1.04056728,
      "epoch": 0.025731978596765466,
      "flos": 19062282245760.0,
      "grad_norm": 2.1001890165969774,
      "language_loss": 0.89962023,
      "learning_rate": 3.887359591851937e-06,
      "loss": 0.92375541,
      "num_input_tokens_seen": 4464715,
      "step": 214,
      "time_per_iteration": 2.522515296936035
    },
    {
      "auxiliary_loss_clip": 0.01349348,
      "auxiliary_loss_mlp": 0.0105735,
      "balance_loss_clip": 1.09714103,
      "balance_loss_mlp": 1.04279423,
      "epoch": 0.025852221487404556,
      "flos": 22163814927360.0,
      "grad_norm": 1.6143044097267194,
      "language_loss": 0.92470986,
      "learning_rate": 3.890736968365265e-06,
      "loss": 0.94877684,
      "num_input_tokens_seen": 4485030,
      "step": 215,
      "time_per_iteration": 2.5590529441833496
    },
    {
      "auxiliary_loss_clip": 0.01351818,
      "auxiliary_loss_mlp": 0.01051289,
      "balance_loss_clip": 1.0962857,
      "balance_loss_mlp": 1.03622103,
      "epoch": 0.02597246437804365,
      "flos": 26541971861760.0,
      "grad_norm": 31.792986009628954,
      "language_loss": 0.85065651,
      "learning_rate": 3.894098672541412e-06,
      "loss": 0.87468767,
      "num_input_tokens_seen": 4505935,
      "step": 216,
      "time_per_iteration": 2.563410997390747
    },
    {
      "auxiliary_loss_clip": 0.01352634,
      "auxiliary_loss_mlp": 0.01054062,
      "balance_loss_clip": 1.09759808,
      "balance_loss_mlp": 1.03844547,
      "epoch": 0.02609270726868274,
      "flos": 32671671696000.0,
      "grad_norm": 2.2379833621656857,
      "language_loss": 0.75454122,
      "learning_rate": 3.89744484916025e-06,
      "loss": 0.7786082,
      "num_input_tokens_seen": 4527045,
      "step": 217,
      "time_per_iteration": 2.686225414276123
    },
    {
      "auxiliary_loss_clip": 0.01354128,
      "auxiliary_loss_mlp": 0.01069194,
      "balance_loss_clip": 1.09946012,
      "balance_loss_mlp": 1.05229056,
      "epoch": 0.02621295015932183,
      "flos": 26243553669120.0,
      "grad_norm": 1.8798467979218316,
      "language_loss": 0.87242806,
      "learning_rate": 3.900775641004673e-06,
      "loss": 0.89666128,
      "num_input_tokens_seen": 4546360,
      "step": 218,
      "time_per_iteration": 2.5786094665527344
    },
    {
      "auxiliary_loss_clip": 0.01358598,
      "auxiliary_loss_mlp": 0.01054199,
      "balance_loss_clip": 1.10063612,
      "balance_loss_mlp": 1.03767681,
      "epoch": 0.026333193049960922,
      "flos": 42921402353280.0,
      "grad_norm": 3.5744107738675703,
      "language_loss": 0.74149132,
      "learning_rate": 3.904091188897156e-06,
      "loss": 0.76561928,
      "num_input_tokens_seen": 4565495,
      "step": 219,
      "time_per_iteration": 2.706536293029785
    },
    {
      "auxiliary_loss_clip": 0.01352853,
      "auxiliary_loss_mlp": 0.01069619,
      "balance_loss_clip": 1.09920883,
      "balance_loss_mlp": 1.05455053,
      "epoch": 0.026453435940600012,
      "flos": 17963846386560.0,
      "grad_norm": 2.0829457919395837,
      "language_loss": 0.82206857,
      "learning_rate": 3.90739163173548e-06,
      "loss": 0.84629327,
      "num_input_tokens_seen": 4583330,
      "step": 220,
      "time_per_iteration": 2.52390193939209
    },
    {
      "auxiliary_loss_clip": 0.01348959,
      "auxiliary_loss_mlp": 0.01058568,
      "balance_loss_clip": 1.09491587,
      "balance_loss_mlp": 1.04258204,
      "epoch": 0.026573678831239102,
      "flos": 18984319776000.0,
      "grad_norm": 2.2873343918620503,
      "language_loss": 0.88522542,
      "learning_rate": 3.910677106527646e-06,
      "loss": 0.90930068,
      "num_input_tokens_seen": 4600520,
      "step": 221,
      "time_per_iteration": 2.527245044708252
    },
    {
      "auxiliary_loss_clip": 0.01348308,
      "auxiliary_loss_mlp": 0.01043507,
      "balance_loss_clip": 1.09672129,
      "balance_loss_mlp": 1.02927351,
      "epoch": 0.026693921721878195,
      "flos": 29241448634880.0,
      "grad_norm": 2.493035504738007,
      "language_loss": 0.84181643,
      "learning_rate": 3.913947748426004e-06,
      "loss": 0.86573458,
      "num_input_tokens_seen": 4617340,
      "step": 222,
      "time_per_iteration": 2.5639421939849854
    },
    {
      "auxiliary_loss_clip": 0.01350787,
      "auxiliary_loss_mlp": 0.01058914,
      "balance_loss_clip": 1.09930432,
      "balance_loss_mlp": 1.0441798,
      "epoch": 0.026814164612517285,
      "flos": 14128083797760.0,
      "grad_norm": 2.8062866278377845,
      "language_loss": 0.76366138,
      "learning_rate": 3.9172036907606136e-06,
      "loss": 0.78775847,
      "num_input_tokens_seen": 4630820,
      "step": 223,
      "time_per_iteration": 3.2294790744781494
    },
    {
      "auxiliary_loss_clip": 0.01348281,
      "auxiliary_loss_mlp": 0.01057183,
      "balance_loss_clip": 1.09565723,
      "balance_loss_mlp": 1.04123282,
      "epoch": 0.026934407503156375,
      "flos": 23511973115520.0,
      "grad_norm": 1.7010636237267593,
      "language_loss": 0.95084202,
      "learning_rate": 3.920445065071855e-06,
      "loss": 0.97489661,
      "num_input_tokens_seen": 4651985,
      "step": 224,
      "time_per_iteration": 2.599256753921509
    },
    {
      "auxiliary_loss_clip": 0.01346965,
      "auxiliary_loss_mlp": 0.01050364,
      "balance_loss_clip": 1.09693027,
      "balance_loss_mlp": 1.03624368,
      "epoch": 0.027054650393795468,
      "flos": 28950356816640.0,
      "grad_norm": 3.41104465251325,
      "language_loss": 0.79954422,
      "learning_rate": 3.923672001142322e-06,
      "loss": 0.82351744,
      "num_input_tokens_seen": 4672295,
      "step": 225,
      "time_per_iteration": 3.3217508792877197
    },
    {
      "auxiliary_loss_clip": 0.01345738,
      "auxiliary_loss_mlp": 0.01065658,
      "balance_loss_clip": 1.09679198,
      "balance_loss_mlp": 1.05164504,
      "epoch": 0.027174893284434558,
      "flos": 31431568596480.0,
      "grad_norm": 2.06788590509748,
      "language_loss": 0.84489751,
      "learning_rate": 3.926884627027996e-06,
      "loss": 0.86901146,
      "num_input_tokens_seen": 4696065,
      "step": 226,
      "time_per_iteration": 3.4337024688720703
    },
    {
      "auxiliary_loss_clip": 0.01345855,
      "auxiliary_loss_mlp": 0.01062661,
      "balance_loss_clip": 1.09613228,
      "balance_loss_mlp": 1.04742622,
      "epoch": 0.027295136175073648,
      "flos": 22054466949120.0,
      "grad_norm": 1.9168353711792765,
      "language_loss": 0.77478689,
      "learning_rate": 3.930083069088744e-06,
      "loss": 0.79887205,
      "num_input_tokens_seen": 4716065,
      "step": 227,
      "time_per_iteration": 2.621284246444702
    },
    {
      "auxiliary_loss_clip": 0.01332799,
      "auxiliary_loss_mlp": 0.01052352,
      "balance_loss_clip": 1.14926231,
      "balance_loss_mlp": 1.04462719,
      "epoch": 0.02741537906571274,
      "flos": 60800752972800.0,
      "grad_norm": 1.0095291513436848,
      "language_loss": 0.59378207,
      "learning_rate": 3.933267452018137e-06,
      "loss": 0.61763352,
      "num_input_tokens_seen": 4775860,
      "step": 228,
      "time_per_iteration": 3.1514360904693604
    },
    {
      "auxiliary_loss_clip": 0.01346603,
      "auxiliary_loss_mlp": 0.01054454,
      "balance_loss_clip": 1.09743333,
      "balance_loss_mlp": 1.04082239,
      "epoch": 0.02753562195635183,
      "flos": 24606278910720.0,
      "grad_norm": 2.754758500916086,
      "language_loss": 0.84264541,
      "learning_rate": 3.936437898872622e-06,
      "loss": 0.86665606,
      "num_input_tokens_seen": 4795835,
      "step": 229,
      "time_per_iteration": 2.5734877586364746
    },
    {
      "auxiliary_loss_clip": 0.01343899,
      "auxiliary_loss_mlp": 0.01059172,
      "balance_loss_clip": 1.09582353,
      "balance_loss_mlp": 1.04387748,
      "epoch": 0.02765586484699092,
      "flos": 34094236907520.0,
      "grad_norm": 2.8586147508221447,
      "language_loss": 0.79840773,
      "learning_rate": 3.9395945311000525e-06,
      "loss": 0.82243842,
      "num_input_tokens_seen": 4817460,
      "step": 230,
      "time_per_iteration": 2.6784613132476807
    },
    {
      "auxiliary_loss_clip": 0.01347301,
      "auxiliary_loss_mlp": 0.01053951,
      "balance_loss_clip": 1.0970757,
      "balance_loss_mlp": 1.03782201,
      "epoch": 0.027776107737630014,
      "flos": 14829922615680.0,
      "grad_norm": 2.0717027991475674,
      "language_loss": 0.90772343,
      "learning_rate": 3.942737468567608e-06,
      "loss": 0.93173593,
      "num_input_tokens_seen": 4835475,
      "step": 231,
      "time_per_iteration": 2.5196847915649414
    },
    {
      "auxiliary_loss_clip": 0.01340656,
      "auxiliary_loss_mlp": 0.01053785,
      "balance_loss_clip": 1.0944767,
      "balance_loss_mlp": 1.03961062,
      "epoch": 0.027896350628269104,
      "flos": 47920347066240.0,
      "grad_norm": 2.2758480233354046,
      "language_loss": 0.86153758,
      "learning_rate": 3.9458668295891026e-06,
      "loss": 0.88548195,
      "num_input_tokens_seen": 4857760,
      "step": 232,
      "time_per_iteration": 2.766575574874878
    },
    {
      "auxiliary_loss_clip": 0.01336423,
      "auxiliary_loss_mlp": 0.01051812,
      "balance_loss_clip": 1.09038424,
      "balance_loss_mlp": 1.03586173,
      "epoch": 0.028016593518908194,
      "flos": 21684550734720.0,
      "grad_norm": 2.311099590148783,
      "language_loss": 0.86897814,
      "learning_rate": 3.948982730951712e-06,
      "loss": 0.89286053,
      "num_input_tokens_seen": 4875855,
      "step": 233,
      "time_per_iteration": 2.513349771499634
    },
    {
      "auxiliary_loss_clip": 0.01342793,
      "auxiliary_loss_mlp": 0.01062367,
      "balance_loss_clip": 1.09354448,
      "balance_loss_mlp": 1.04621458,
      "epoch": 0.028136836409547287,
      "flos": 18439483305600.0,
      "grad_norm": 2.315154096288921,
      "language_loss": 0.81818986,
      "learning_rate": 3.9520852879421254e-06,
      "loss": 0.84224153,
      "num_input_tokens_seen": 4893200,
      "step": 234,
      "time_per_iteration": 2.486950397491455
    },
    {
      "auxiliary_loss_clip": 0.01341031,
      "auxiliary_loss_mlp": 0.01055023,
      "balance_loss_clip": 1.09608364,
      "balance_loss_mlp": 1.03999102,
      "epoch": 0.028257079300186377,
      "flos": 31576934937600.0,
      "grad_norm": 2.197228528694378,
      "language_loss": 0.81636643,
      "learning_rate": 3.955174614372137e-06,
      "loss": 0.84032691,
      "num_input_tokens_seen": 4912965,
      "step": 235,
      "time_per_iteration": 2.6110587120056152
    },
    {
      "auxiliary_loss_clip": 0.01339849,
      "auxiliary_loss_mlp": 0.01058273,
      "balance_loss_clip": 1.09528136,
      "balance_loss_mlp": 1.04339015,
      "epoch": 0.028377322190825467,
      "flos": 23513337832320.0,
      "grad_norm": 3.6045207715356704,
      "language_loss": 0.84285545,
      "learning_rate": 3.9582508226037045e-06,
      "loss": 0.86683667,
      "num_input_tokens_seen": 4933105,
      "step": 236,
      "time_per_iteration": 2.539132833480835
    },
    {
      "auxiliary_loss_clip": 0.01342671,
      "auxiliary_loss_mlp": 0.01060793,
      "balance_loss_clip": 1.09393406,
      "balance_loss_mlp": 1.04388952,
      "epoch": 0.02849756508146456,
      "flos": 20479604071680.0,
      "grad_norm": 2.331757055494177,
      "language_loss": 0.94015527,
      "learning_rate": 3.9613140235734636e-06,
      "loss": 0.96418989,
      "num_input_tokens_seen": 4950085,
      "step": 237,
      "time_per_iteration": 2.526366710662842
    },
    {
      "auxiliary_loss_clip": 0.01342218,
      "auxiliary_loss_mlp": 0.01065478,
      "balance_loss_clip": 1.09648585,
      "balance_loss_mlp": 1.05064869,
      "epoch": 0.02861780797210365,
      "flos": 14283362292480.0,
      "grad_norm": 1.8613490746181172,
      "language_loss": 0.81386042,
      "learning_rate": 3.96436432681674e-06,
      "loss": 0.83793736,
      "num_input_tokens_seen": 4968075,
      "step": 238,
      "time_per_iteration": 2.4888927936553955
    },
    {
      "auxiliary_loss_clip": 0.01340708,
      "auxiliary_loss_mlp": 0.01066483,
      "balance_loss_clip": 1.0949719,
      "balance_loss_mlp": 1.05084896,
      "epoch": 0.02873805086274274,
      "flos": 25808532053760.0,
      "grad_norm": 2.1996843196600864,
      "language_loss": 0.89110518,
      "learning_rate": 3.967401840491044e-06,
      "loss": 0.91517705,
      "num_input_tokens_seen": 4987355,
      "step": 239,
      "time_per_iteration": 2.601609945297241
    },
    {
      "auxiliary_loss_clip": 0.01339571,
      "auxiliary_loss_mlp": 0.01065391,
      "balance_loss_clip": 1.09678435,
      "balance_loss_mlp": 1.04994774,
      "epoch": 0.028858293753381833,
      "flos": 17304238984320.0,
      "grad_norm": 2.440108628282799,
      "language_loss": 0.87792104,
      "learning_rate": 3.97042667139909e-06,
      "loss": 0.90197068,
      "num_input_tokens_seen": 5004680,
      "step": 240,
      "time_per_iteration": 2.5748629570007324
    },
    {
      "auxiliary_loss_clip": 0.0134225,
      "auxiliary_loss_mlp": 0.01061305,
      "balance_loss_clip": 1.09688139,
      "balance_loss_mlp": 1.04628491,
      "epoch": 0.028978536644020923,
      "flos": 23038347358080.0,
      "grad_norm": 1.970998699805061,
      "language_loss": 0.87439859,
      "learning_rate": 3.973438925011327e-06,
      "loss": 0.89843416,
      "num_input_tokens_seen": 5022965,
      "step": 241,
      "time_per_iteration": 2.561873435974121
    },
    {
      "auxiliary_loss_clip": 0.01333938,
      "auxiliary_loss_mlp": 0.01049484,
      "balance_loss_clip": 1.08921599,
      "balance_loss_mlp": 1.03589988,
      "epoch": 0.029098779534660012,
      "flos": 28329712692480.0,
      "grad_norm": 2.6350301267290575,
      "language_loss": 0.91208136,
      "learning_rate": 3.976438705488002e-06,
      "loss": 0.93591559,
      "num_input_tokens_seen": 5042625,
      "step": 242,
      "time_per_iteration": 2.5613090991973877
    },
    {
      "auxiliary_loss_clip": 0.01338429,
      "auxiliary_loss_mlp": 0.01064035,
      "balance_loss_clip": 1.09511232,
      "balance_loss_mlp": 1.04922879,
      "epoch": 0.029219022425299106,
      "flos": 13881665520000.0,
      "grad_norm": 2.7389999747801,
      "language_loss": 0.93144369,
      "learning_rate": 3.9794261157007744e-06,
      "loss": 0.9554683,
      "num_input_tokens_seen": 5060380,
      "step": 243,
      "time_per_iteration": 2.5738015174865723
    },
    {
      "auxiliary_loss_clip": 0.01341395,
      "auxiliary_loss_mlp": 0.01057949,
      "balance_loss_clip": 1.09806585,
      "balance_loss_mlp": 1.04179573,
      "epoch": 0.029339265315938196,
      "flos": 19422501788160.0,
      "grad_norm": 2.211819095413116,
      "language_loss": 0.8448602,
      "learning_rate": 3.982401257253887e-06,
      "loss": 0.86885363,
      "num_input_tokens_seen": 5078720,
      "step": 244,
      "time_per_iteration": 2.546577215194702
    },
    {
      "auxiliary_loss_clip": 0.01334843,
      "auxiliary_loss_mlp": 0.01060225,
      "balance_loss_clip": 1.09381151,
      "balance_loss_mlp": 1.0452404,
      "epoch": 0.029459508206577285,
      "flos": 15669550005120.0,
      "grad_norm": 2.4162360068181763,
      "language_loss": 0.90021223,
      "learning_rate": 3.985364230504893e-06,
      "loss": 0.92416298,
      "num_input_tokens_seen": 5096605,
      "step": 245,
      "time_per_iteration": 2.565833806991577
    },
    {
      "auxiliary_loss_clip": 0.01337186,
      "auxiliary_loss_mlp": 0.01055832,
      "balance_loss_clip": 1.09351432,
      "balance_loss_mlp": 1.04033518,
      "epoch": 0.02957975109721638,
      "flos": 28220975245440.0,
      "grad_norm": 1.9058354586502553,
      "language_loss": 0.8429026,
      "learning_rate": 3.988315134584976e-06,
      "loss": 0.86683273,
      "num_input_tokens_seen": 5116285,
      "step": 246,
      "time_per_iteration": 2.5868866443634033
    },
    {
      "auxiliary_loss_clip": 0.01339891,
      "auxiliary_loss_mlp": 0.01054593,
      "balance_loss_clip": 1.09684002,
      "balance_loss_mlp": 1.03995454,
      "epoch": 0.02969999398785547,
      "flos": 24315869450880.0,
      "grad_norm": 2.2153913035327677,
      "language_loss": 0.80170518,
      "learning_rate": 3.991254067418851e-06,
      "loss": 0.82564998,
      "num_input_tokens_seen": 5136825,
      "step": 247,
      "time_per_iteration": 2.5695528984069824
    },
    {
      "auxiliary_loss_clip": 0.01334671,
      "auxiliary_loss_mlp": 0.01057435,
      "balance_loss_clip": 1.0934633,
      "balance_loss_mlp": 1.0428319,
      "epoch": 0.02982023687849456,
      "flos": 35078584193280.0,
      "grad_norm": 2.325894301640739,
      "language_loss": 0.82863331,
      "learning_rate": 3.994181125744254e-06,
      "loss": 0.85255432,
      "num_input_tokens_seen": 5158630,
      "step": 248,
      "time_per_iteration": 2.63382625579834
    },
    {
      "auxiliary_loss_clip": 0.01337917,
      "auxiliary_loss_mlp": 0.01060942,
      "balance_loss_clip": 1.09625649,
      "balance_loss_mlp": 1.04687572,
      "epoch": 0.02994047976913365,
      "flos": 26177155378560.0,
      "grad_norm": 1.7884191148268913,
      "language_loss": 0.73850513,
      "learning_rate": 3.99709640513106e-06,
      "loss": 0.76249373,
      "num_input_tokens_seen": 5179510,
      "step": 249,
      "time_per_iteration": 3.318661689758301
    },
    {
      "auxiliary_loss_clip": 0.01338753,
      "auxiliary_loss_mlp": 0.0104968,
      "balance_loss_clip": 1.09465575,
      "balance_loss_mlp": 1.03401566,
      "epoch": 0.03006072265977274,
      "flos": 25625028447360.0,
      "grad_norm": 1.9110357881473876,
      "language_loss": 0.85603315,
      "learning_rate": 4e-06,
      "loss": 0.87991744,
      "num_input_tokens_seen": 5199345,
      "step": 250,
      "time_per_iteration": 2.5853569507598877
    },
    {
      "auxiliary_loss_clip": 0.01339696,
      "auxiliary_loss_mlp": 0.01068033,
      "balance_loss_clip": 1.09865189,
      "balance_loss_mlp": 1.05344224,
      "epoch": 0.03018096555041183,
      "flos": 22127078292480.0,
      "grad_norm": 5.981262042134408,
      "language_loss": 0.88129616,
      "learning_rate": 3.999999848300794e-06,
      "loss": 0.90537345,
      "num_input_tokens_seen": 5218330,
      "step": 251,
      "time_per_iteration": 3.255072832107544
    },
    {
      "auxiliary_loss_clip": 0.01330771,
      "auxiliary_loss_mlp": 0.01052639,
      "balance_loss_clip": 1.0906322,
      "balance_loss_mlp": 1.03764296,
      "epoch": 0.030301208441050925,
      "flos": 30188197359360.0,
      "grad_norm": 1.6568302554024308,
      "language_loss": 0.89170271,
      "learning_rate": 3.999999393203203e-06,
      "loss": 0.91553676,
      "num_input_tokens_seen": 5240740,
      "step": 252,
      "time_per_iteration": 4.091420888900757
    },
    {
      "auxiliary_loss_clip": 0.01328956,
      "auxiliary_loss_mlp": 0.01057486,
      "balance_loss_clip": 1.09000897,
      "balance_loss_mlp": 1.04259706,
      "epoch": 0.030421451331690014,
      "flos": 23621392920960.0,
      "grad_norm": 2.001010413322784,
      "language_loss": 0.85005176,
      "learning_rate": 3.999998634707293e-06,
      "loss": 0.87391627,
      "num_input_tokens_seen": 5260290,
      "step": 253,
      "time_per_iteration": 2.520881175994873
    },
    {
      "auxiliary_loss_clip": 0.01337847,
      "auxiliary_loss_mlp": 0.01063578,
      "balance_loss_clip": 1.09796453,
      "balance_loss_mlp": 1.04772305,
      "epoch": 0.030541694222329104,
      "flos": 27928446883200.0,
      "grad_norm": 2.3288344351860335,
      "language_loss": 0.96367872,
      "learning_rate": 3.999997572813182e-06,
      "loss": 0.98769295,
      "num_input_tokens_seen": 5278100,
      "step": 254,
      "time_per_iteration": 2.561330795288086
    },
    {
      "auxiliary_loss_clip": 0.01336374,
      "auxiliary_loss_mlp": 0.01055275,
      "balance_loss_clip": 1.09416366,
      "balance_loss_mlp": 1.04055882,
      "epoch": 0.030661937112968194,
      "flos": 18588441006720.0,
      "grad_norm": 1.963566879368455,
      "language_loss": 0.87522596,
      "learning_rate": 3.999996207521028e-06,
      "loss": 0.89914238,
      "num_input_tokens_seen": 5296810,
      "step": 255,
      "time_per_iteration": 2.544182777404785
    },
    {
      "auxiliary_loss_clip": 0.01333335,
      "auxiliary_loss_mlp": 0.0105334,
      "balance_loss_clip": 1.09279597,
      "balance_loss_mlp": 1.0371753,
      "epoch": 0.030782180003607287,
      "flos": 12969139478400.0,
      "grad_norm": 2.586974589679778,
      "language_loss": 0.82402045,
      "learning_rate": 3.999994538831039e-06,
      "loss": 0.84788722,
      "num_input_tokens_seen": 5313395,
      "step": 256,
      "time_per_iteration": 2.546891927719116
    },
    {
      "auxiliary_loss_clip": 0.01331392,
      "auxiliary_loss_mlp": 0.01065701,
      "balance_loss_clip": 1.09196126,
      "balance_loss_mlp": 1.04934525,
      "epoch": 0.030902422894246377,
      "flos": 23335364920320.0,
      "grad_norm": 2.5537645168596086,
      "language_loss": 0.85794568,
      "learning_rate": 3.99999256674347e-06,
      "loss": 0.88191664,
      "num_input_tokens_seen": 5333545,
      "step": 257,
      "time_per_iteration": 2.5956575870513916
    },
    {
      "auxiliary_loss_clip": 0.01307763,
      "auxiliary_loss_mlp": 0.01021069,
      "balance_loss_clip": 1.14026558,
      "balance_loss_mlp": 1.01377344,
      "epoch": 0.031022665784885467,
      "flos": 55094151438720.0,
      "grad_norm": 1.0142093281151994,
      "language_loss": 0.53576285,
      "learning_rate": 3.999990291258618e-06,
      "loss": 0.55905116,
      "num_input_tokens_seen": 5392235,
      "step": 258,
      "time_per_iteration": 3.090458631515503
    },
    {
      "auxiliary_loss_clip": 0.01332934,
      "auxiliary_loss_mlp": 0.01055239,
      "balance_loss_clip": 1.09403229,
      "balance_loss_mlp": 1.04191768,
      "epoch": 0.03114290867552456,
      "flos": 19317786664320.0,
      "grad_norm": 2.224334568279,
      "language_loss": 0.86729789,
      "learning_rate": 3.999987712376829e-06,
      "loss": 0.89117956,
      "num_input_tokens_seen": 5410555,
      "step": 259,
      "time_per_iteration": 2.526338815689087
    },
    {
      "auxiliary_loss_clip": 0.0133238,
      "auxiliary_loss_mlp": 0.01051043,
      "balance_loss_clip": 1.09578979,
      "balance_loss_mlp": 1.03623128,
      "epoch": 0.031263151566163654,
      "flos": 20959442881920.0,
      "grad_norm": 1.8755625018424293,
      "language_loss": 0.81756967,
      "learning_rate": 3.999984830098494e-06,
      "loss": 0.84140396,
      "num_input_tokens_seen": 5430135,
      "step": 260,
      "time_per_iteration": 2.5190436840057373
    },
    {
      "auxiliary_loss_clip": 0.01332976,
      "auxiliary_loss_mlp": 0.01068433,
      "balance_loss_clip": 1.09351707,
      "balance_loss_mlp": 1.05230451,
      "epoch": 0.03138339445680274,
      "flos": 14793006412800.0,
      "grad_norm": 2.966091894560407,
      "language_loss": 0.97791523,
      "learning_rate": 3.999981644424051e-06,
      "loss": 1.00192928,
      "num_input_tokens_seen": 5444935,
      "step": 261,
      "time_per_iteration": 2.499943494796753
    },
    {
      "auxiliary_loss_clip": 0.0133133,
      "auxiliary_loss_mlp": 0.01054499,
      "balance_loss_clip": 1.09331894,
      "balance_loss_mlp": 1.03965735,
      "epoch": 0.03150363734744183,
      "flos": 11655599022720.0,
      "grad_norm": 2.3578316503244006,
      "language_loss": 0.86097372,
      "learning_rate": 3.999978155353982e-06,
      "loss": 0.88483196,
      "num_input_tokens_seen": 5462080,
      "step": 262,
      "time_per_iteration": 2.4861016273498535
    },
    {
      "auxiliary_loss_clip": 0.01328989,
      "auxiliary_loss_mlp": 0.01054803,
      "balance_loss_clip": 1.09129882,
      "balance_loss_mlp": 1.04084957,
      "epoch": 0.03162388023808092,
      "flos": 33727732485120.0,
      "grad_norm": 2.3171670857925766,
      "language_loss": 0.80099583,
      "learning_rate": 3.9999743628888186e-06,
      "loss": 0.82483381,
      "num_input_tokens_seen": 5483870,
      "step": 263,
      "time_per_iteration": 2.645575761795044
    },
    {
      "auxiliary_loss_clip": 0.01330153,
      "auxiliary_loss_mlp": 0.01044827,
      "balance_loss_clip": 1.09274089,
      "balance_loss_mlp": 1.02996206,
      "epoch": 0.03174412312872001,
      "flos": 20810952057600.0,
      "grad_norm": 2.356165196941417,
      "language_loss": 0.8960517,
      "learning_rate": 3.999970267029133e-06,
      "loss": 0.91980159,
      "num_input_tokens_seen": 5502830,
      "step": 264,
      "time_per_iteration": 2.548417568206787
    },
    {
      "auxiliary_loss_clip": 0.01327689,
      "auxiliary_loss_mlp": 0.01053892,
      "balance_loss_clip": 1.0919385,
      "balance_loss_mlp": 1.04015875,
      "epoch": 0.0318643660193591,
      "flos": 23727939638400.0,
      "grad_norm": 2.0181700796910316,
      "language_loss": 0.80020487,
      "learning_rate": 3.999965867775548e-06,
      "loss": 0.82402062,
      "num_input_tokens_seen": 5523225,
      "step": 265,
      "time_per_iteration": 2.5712244510650635
    },
    {
      "auxiliary_loss_clip": 0.01327563,
      "auxiliary_loss_mlp": 0.01057608,
      "balance_loss_clip": 1.09111202,
      "balance_loss_mlp": 1.04301703,
      "epoch": 0.0319846089099982,
      "flos": 13917863450880.0,
      "grad_norm": 3.6718332502600215,
      "language_loss": 0.86877465,
      "learning_rate": 3.9999611651287315e-06,
      "loss": 0.8926264,
      "num_input_tokens_seen": 5541380,
      "step": 266,
      "time_per_iteration": 2.5207462310791016
    },
    {
      "auxiliary_loss_clip": 0.01333813,
      "auxiliary_loss_mlp": 0.01056239,
      "balance_loss_clip": 1.09465802,
      "balance_loss_mlp": 1.04134941,
      "epoch": 0.03210485180063729,
      "flos": 14753253035520.0,
      "grad_norm": 2.4714967286756897,
      "language_loss": 0.78752661,
      "learning_rate": 3.999956159089396e-06,
      "loss": 0.81142712,
      "num_input_tokens_seen": 5558830,
      "step": 267,
      "time_per_iteration": 2.53702974319458
    },
    {
      "auxiliary_loss_clip": 0.01328852,
      "auxiliary_loss_mlp": 0.01064132,
      "balance_loss_clip": 1.09336877,
      "balance_loss_mlp": 1.04919529,
      "epoch": 0.03222509469127638,
      "flos": 28913153304960.0,
      "grad_norm": 2.1904948433913956,
      "language_loss": 0.79897857,
      "learning_rate": 3.999950849658302e-06,
      "loss": 0.8229084,
      "num_input_tokens_seen": 5577750,
      "step": 268,
      "time_per_iteration": 2.551652669906616
    },
    {
      "auxiliary_loss_clip": 0.01330574,
      "auxiliary_loss_mlp": 0.01067854,
      "balance_loss_clip": 1.09451795,
      "balance_loss_mlp": 1.05218983,
      "epoch": 0.03234533758191547,
      "flos": 16946389739520.0,
      "grad_norm": 6.056220652155281,
      "language_loss": 0.84342527,
      "learning_rate": 3.999945236836254e-06,
      "loss": 0.86740947,
      "num_input_tokens_seen": 5596715,
      "step": 269,
      "time_per_iteration": 2.5469486713409424
    },
    {
      "auxiliary_loss_clip": 0.01334459,
      "auxiliary_loss_mlp": 0.01077234,
      "balance_loss_clip": 1.09569645,
      "balance_loss_mlp": 1.06096208,
      "epoch": 0.03246558047255456,
      "flos": 18989096284800.0,
      "grad_norm": 2.8572679751800947,
      "language_loss": 0.94941884,
      "learning_rate": 3.999939320624103e-06,
      "loss": 0.97353578,
      "num_input_tokens_seen": 5611865,
      "step": 270,
      "time_per_iteration": 2.477614402770996
    },
    {
      "auxiliary_loss_clip": 0.01330925,
      "auxiliary_loss_mlp": 0.01053959,
      "balance_loss_clip": 1.09552705,
      "balance_loss_mlp": 1.03952253,
      "epoch": 0.03258582336319365,
      "flos": 23728334688000.0,
      "grad_norm": 2.917799163489746,
      "language_loss": 0.90128386,
      "learning_rate": 3.999933101022749e-06,
      "loss": 0.92513269,
      "num_input_tokens_seen": 5632270,
      "step": 271,
      "time_per_iteration": 2.5571577548980713
    },
    {
      "auxiliary_loss_clip": 0.01331097,
      "auxiliary_loss_mlp": 0.01056021,
      "balance_loss_clip": 1.0959301,
      "balance_loss_mlp": 1.04074454,
      "epoch": 0.032706066253832745,
      "flos": 27670823562240.0,
      "grad_norm": 2.119537439640371,
      "language_loss": 0.86997116,
      "learning_rate": 3.999926578033132e-06,
      "loss": 0.89384234,
      "num_input_tokens_seen": 5652085,
      "step": 272,
      "time_per_iteration": 2.542473554611206
    },
    {
      "auxiliary_loss_clip": 0.01327461,
      "auxiliary_loss_mlp": 0.01059856,
      "balance_loss_clip": 1.08978271,
      "balance_loss_mlp": 1.04360783,
      "epoch": 0.032826309144471835,
      "flos": 45624685968000.0,
      "grad_norm": 2.0921552073934606,
      "language_loss": 0.62907565,
      "learning_rate": 3.999919751656244e-06,
      "loss": 0.6529488,
      "num_input_tokens_seen": 5678985,
      "step": 273,
      "time_per_iteration": 2.789545774459839
    },
    {
      "auxiliary_loss_clip": 0.01328661,
      "auxiliary_loss_mlp": 0.01066036,
      "balance_loss_clip": 1.09290957,
      "balance_loss_mlp": 1.05144453,
      "epoch": 0.032946552035110925,
      "flos": 25812374808960.0,
      "grad_norm": 7.65831127895052,
      "language_loss": 0.7600857,
      "learning_rate": 3.9999126218931195e-06,
      "loss": 0.7840327,
      "num_input_tokens_seen": 5697020,
      "step": 274,
      "time_per_iteration": 2.532543182373047
    },
    {
      "auxiliary_loss_clip": 0.0133317,
      "auxiliary_loss_mlp": 0.01065876,
      "balance_loss_clip": 1.09775794,
      "balance_loss_mlp": 1.05059338,
      "epoch": 0.033066794925750015,
      "flos": 15121984101120.0,
      "grad_norm": 2.2202396544236116,
      "language_loss": 0.89606571,
      "learning_rate": 3.99990518874484e-06,
      "loss": 0.9200561,
      "num_input_tokens_seen": 5713460,
      "step": 275,
      "time_per_iteration": 3.283907413482666
    },
    {
      "auxiliary_loss_clip": 0.01334622,
      "auxiliary_loss_mlp": 0.01058733,
      "balance_loss_clip": 1.09773219,
      "balance_loss_mlp": 1.04416609,
      "epoch": 0.033187037816389105,
      "flos": 22776593973120.0,
      "grad_norm": 2.1791699189992433,
      "language_loss": 0.92370325,
      "learning_rate": 3.999897452212534e-06,
      "loss": 0.94763684,
      "num_input_tokens_seen": 5730790,
      "step": 276,
      "time_per_iteration": 2.5228428840637207
    },
    {
      "auxiliary_loss_clip": 0.01325674,
      "auxiliary_loss_mlp": 0.01065248,
      "balance_loss_clip": 1.09222293,
      "balance_loss_mlp": 1.04955959,
      "epoch": 0.033307280707028195,
      "flos": 23331414424320.0,
      "grad_norm": 2.1790188640933073,
      "language_loss": 1.00224972,
      "learning_rate": 3.999889412297374e-06,
      "loss": 1.02615881,
      "num_input_tokens_seen": 5750215,
      "step": 277,
      "time_per_iteration": 2.558551788330078
    },
    {
      "auxiliary_loss_clip": 0.01329355,
      "auxiliary_loss_mlp": 0.0105818,
      "balance_loss_clip": 1.09278548,
      "balance_loss_mlp": 1.04369628,
      "epoch": 0.03342752359766729,
      "flos": 28840290566400.0,
      "grad_norm": 1.9550817178253963,
      "language_loss": 0.78991795,
      "learning_rate": 3.999881069000581e-06,
      "loss": 0.8137933,
      "num_input_tokens_seen": 5769945,
      "step": 278,
      "time_per_iteration": 2.5984652042388916
    },
    {
      "auxiliary_loss_clip": 0.01324833,
      "auxiliary_loss_mlp": 0.0105935,
      "balance_loss_clip": 1.08919489,
      "balance_loss_mlp": 1.04443717,
      "epoch": 0.03354776648830638,
      "flos": 19384544090880.0,
      "grad_norm": 3.770273553066629,
      "language_loss": 0.86596984,
      "learning_rate": 3.99987242232342e-06,
      "loss": 0.88981164,
      "num_input_tokens_seen": 5784950,
      "step": 279,
      "time_per_iteration": 4.056426048278809
    },
    {
      "auxiliary_loss_clip": 0.01328504,
      "auxiliary_loss_mlp": 0.01059067,
      "balance_loss_clip": 1.09599733,
      "balance_loss_mlp": 1.04389191,
      "epoch": 0.03366800937894547,
      "flos": 17858628472320.0,
      "grad_norm": 1.7795316541151167,
      "language_loss": 0.79775548,
      "learning_rate": 3.9998634722672026e-06,
      "loss": 0.82163119,
      "num_input_tokens_seen": 5805005,
      "step": 280,
      "time_per_iteration": 2.554086446762085
    },
    {
      "auxiliary_loss_clip": 0.0133201,
      "auxiliary_loss_mlp": 0.01052625,
      "balance_loss_clip": 1.09756231,
      "balance_loss_mlp": 1.03718758,
      "epoch": 0.03378825226958456,
      "flos": 35951033635200.0,
      "grad_norm": 2.2206088407616855,
      "language_loss": 0.78592384,
      "learning_rate": 3.999854218833286e-06,
      "loss": 0.80977017,
      "num_input_tokens_seen": 5825825,
      "step": 281,
      "time_per_iteration": 2.6415319442749023
    },
    {
      "auxiliary_loss_clip": 0.01326787,
      "auxiliary_loss_mlp": 0.01061746,
      "balance_loss_clip": 1.09236896,
      "balance_loss_mlp": 1.04797077,
      "epoch": 0.03390849516022365,
      "flos": 25702488126720.0,
      "grad_norm": 2.373221073650071,
      "language_loss": 0.82169604,
      "learning_rate": 3.999844662023075e-06,
      "loss": 0.84558141,
      "num_input_tokens_seen": 5845700,
      "step": 282,
      "time_per_iteration": 2.566725492477417
    },
    {
      "auxiliary_loss_clip": 0.01322713,
      "auxiliary_loss_mlp": 0.01050863,
      "balance_loss_clip": 1.09205127,
      "balance_loss_mlp": 1.0379827,
      "epoch": 0.03402873805086274,
      "flos": 21284505987840.0,
      "grad_norm": 1.7683111561381406,
      "language_loss": 0.91957229,
      "learning_rate": 3.999834801838018e-06,
      "loss": 0.94330812,
      "num_input_tokens_seen": 5864680,
      "step": 283,
      "time_per_iteration": 2.5180513858795166
    },
    {
      "auxiliary_loss_clip": 0.01324341,
      "auxiliary_loss_mlp": 0.01049379,
      "balance_loss_clip": 1.0922296,
      "balance_loss_mlp": 1.03497267,
      "epoch": 0.03414898094150183,
      "flos": 22710913954560.0,
      "grad_norm": 1.852058272429499,
      "language_loss": 0.73851526,
      "learning_rate": 3.9998246382796115e-06,
      "loss": 0.76225239,
      "num_input_tokens_seen": 5884260,
      "step": 284,
      "time_per_iteration": 2.552903890609741
    },
    {
      "auxiliary_loss_clip": 0.01324018,
      "auxiliary_loss_mlp": 0.01054897,
      "balance_loss_clip": 1.09035015,
      "balance_loss_mlp": 1.03925633,
      "epoch": 0.03426922383214093,
      "flos": 18879927874560.0,
      "grad_norm": 2.0351008456041066,
      "language_loss": 0.90830219,
      "learning_rate": 3.999814171349399e-06,
      "loss": 0.93209136,
      "num_input_tokens_seen": 5902120,
      "step": 285,
      "time_per_iteration": 2.4935922622680664
    },
    {
      "auxiliary_loss_clip": 0.01321396,
      "auxiliary_loss_mlp": 0.01074059,
      "balance_loss_clip": 1.09098375,
      "balance_loss_mlp": 1.06048715,
      "epoch": 0.03438946672278002,
      "flos": 34752012716160.0,
      "grad_norm": 2.012974160584163,
      "language_loss": 0.73750973,
      "learning_rate": 3.9998034010489655e-06,
      "loss": 0.7614643,
      "num_input_tokens_seen": 5925810,
      "step": 286,
      "time_per_iteration": 2.6690914630889893
    },
    {
      "auxiliary_loss_clip": 0.01323152,
      "auxiliary_loss_mlp": 0.01039788,
      "balance_loss_clip": 1.0927459,
      "balance_loss_mlp": 1.02609694,
      "epoch": 0.03450970961341911,
      "flos": 22164102236160.0,
      "grad_norm": 2.0071599458032527,
      "language_loss": 0.75688303,
      "learning_rate": 3.999792327379946e-06,
      "loss": 0.78051251,
      "num_input_tokens_seen": 5945185,
      "step": 287,
      "time_per_iteration": 2.553694486618042
    },
    {
      "auxiliary_loss_clip": 0.01322951,
      "auxiliary_loss_mlp": 0.01061798,
      "balance_loss_clip": 1.0925988,
      "balance_loss_mlp": 1.04723108,
      "epoch": 0.034629952504058197,
      "flos": 21725740656000.0,
      "grad_norm": 2.113283984984685,
      "language_loss": 0.96186984,
      "learning_rate": 3.999780950344021e-06,
      "loss": 0.9857173,
      "num_input_tokens_seen": 5963375,
      "step": 288,
      "time_per_iteration": 2.5291478633880615
    },
    {
      "auxiliary_loss_clip": 0.01324364,
      "auxiliary_loss_mlp": 0.0105475,
      "balance_loss_clip": 1.09139669,
      "balance_loss_mlp": 1.03870451,
      "epoch": 0.034750195394697286,
      "flos": 20047994248320.0,
      "grad_norm": 1.9985061410834943,
      "language_loss": 0.82529783,
      "learning_rate": 3.999769269942916e-06,
      "loss": 0.84908897,
      "num_input_tokens_seen": 5983415,
      "step": 289,
      "time_per_iteration": 2.532067060470581
    },
    {
      "auxiliary_loss_clip": 0.0132125,
      "auxiliary_loss_mlp": 0.01062755,
      "balance_loss_clip": 1.09004462,
      "balance_loss_mlp": 1.04876018,
      "epoch": 0.034870438285336376,
      "flos": 27965865876480.0,
      "grad_norm": 2.1282564810361606,
      "language_loss": 0.80963022,
      "learning_rate": 3.999757286178402e-06,
      "loss": 0.83347034,
      "num_input_tokens_seen": 6005850,
      "step": 290,
      "time_per_iteration": 2.603459119796753
    },
    {
      "auxiliary_loss_clip": 0.01324975,
      "auxiliary_loss_mlp": 0.01063681,
      "balance_loss_clip": 1.09452963,
      "balance_loss_mlp": 1.04955435,
      "epoch": 0.03499068117597547,
      "flos": 22017514832640.0,
      "grad_norm": 2.057064719793104,
      "language_loss": 0.90830946,
      "learning_rate": 3.999744999052299e-06,
      "loss": 0.93219602,
      "num_input_tokens_seen": 6027240,
      "step": 291,
      "time_per_iteration": 2.5722527503967285
    },
    {
      "auxiliary_loss_clip": 0.01275081,
      "auxiliary_loss_mlp": 0.01046138,
      "balance_loss_clip": 1.11859703,
      "balance_loss_mlp": 1.03955734,
      "epoch": 0.03511092406661456,
      "flos": 57242147725440.0,
      "grad_norm": 0.9682003665864284,
      "language_loss": 0.61191434,
      "learning_rate": 3.9997324085664675e-06,
      "loss": 0.63512653,
      "num_input_tokens_seen": 6087470,
      "step": 292,
      "time_per_iteration": 3.109800100326538
    },
    {
      "auxiliary_loss_clip": 0.01318551,
      "auxiliary_loss_mlp": 0.0106171,
      "balance_loss_clip": 1.08691382,
      "balance_loss_mlp": 1.04717851,
      "epoch": 0.03523116695725365,
      "flos": 22928065626240.0,
      "grad_norm": 2.405873206927614,
      "language_loss": 0.92186075,
      "learning_rate": 3.999719514722821e-06,
      "loss": 0.94566333,
      "num_input_tokens_seen": 6107600,
      "step": 293,
      "time_per_iteration": 2.5438714027404785
    },
    {
      "auxiliary_loss_clip": 0.01318037,
      "auxiliary_loss_mlp": 0.01052422,
      "balance_loss_clip": 1.08970761,
      "balance_loss_mlp": 1.03914189,
      "epoch": 0.03535140984789274,
      "flos": 36903241226880.0,
      "grad_norm": 13.61778190660226,
      "language_loss": 0.74897373,
      "learning_rate": 3.999706317523314e-06,
      "loss": 0.77267826,
      "num_input_tokens_seen": 6126160,
      "step": 294,
      "time_per_iteration": 2.6228296756744385
    },
    {
      "auxiliary_loss_clip": 0.01320034,
      "auxiliary_loss_mlp": 0.01047499,
      "balance_loss_clip": 1.09124923,
      "balance_loss_mlp": 1.03434408,
      "epoch": 0.03547165273853183,
      "flos": 20449152316800.0,
      "grad_norm": 2.400715673874234,
      "language_loss": 0.8594321,
      "learning_rate": 3.999692816969948e-06,
      "loss": 0.88310748,
      "num_input_tokens_seen": 6145695,
      "step": 295,
      "time_per_iteration": 2.5144224166870117
    },
    {
      "auxiliary_loss_clip": 0.01266356,
      "auxiliary_loss_mlp": 0.01008995,
      "balance_loss_clip": 1.11271596,
      "balance_loss_mlp": 1.00222421,
      "epoch": 0.03559189562917092,
      "flos": 69850564871040.0,
      "grad_norm": 1.001732724736107,
      "language_loss": 0.6940074,
      "learning_rate": 3.999679013064772e-06,
      "loss": 0.71676087,
      "num_input_tokens_seen": 6212440,
      "step": 296,
      "time_per_iteration": 3.134211778640747
    },
    {
      "auxiliary_loss_clip": 0.01317803,
      "auxiliary_loss_mlp": 0.01067996,
      "balance_loss_clip": 1.08846796,
      "balance_loss_mlp": 1.05359507,
      "epoch": 0.03571213851981002,
      "flos": 21651944163840.0,
      "grad_norm": 2.714939384158462,
      "language_loss": 0.85783637,
      "learning_rate": 3.99966490580988e-06,
      "loss": 0.88169432,
      "num_input_tokens_seen": 6229800,
      "step": 297,
      "time_per_iteration": 2.532797336578369
    },
    {
      "auxiliary_loss_clip": 0.01322419,
      "auxiliary_loss_mlp": 0.01056891,
      "balance_loss_clip": 1.09243202,
      "balance_loss_mlp": 1.04200768,
      "epoch": 0.03583238141044911,
      "flos": 43945610757120.0,
      "grad_norm": 2.0070251973319766,
      "language_loss": 0.6565522,
      "learning_rate": 3.999650495207411e-06,
      "loss": 0.6803453,
      "num_input_tokens_seen": 6255825,
      "step": 298,
      "time_per_iteration": 2.711317300796509
    },
    {
      "auxiliary_loss_clip": 0.01319406,
      "auxiliary_loss_mlp": 0.01058953,
      "balance_loss_clip": 1.09167385,
      "balance_loss_mlp": 1.04460025,
      "epoch": 0.0359526243010882,
      "flos": 18910810592640.0,
      "grad_norm": 2.608592937394606,
      "language_loss": 0.90209717,
      "learning_rate": 3.999635781259553e-06,
      "loss": 0.92588079,
      "num_input_tokens_seen": 6271090,
      "step": 299,
      "time_per_iteration": 2.5290775299072266
    },
    {
      "auxiliary_loss_clip": 0.01256287,
      "auxiliary_loss_mlp": 0.01015083,
      "balance_loss_clip": 1.10583234,
      "balance_loss_mlp": 1.00845551,
      "epoch": 0.03607286719172729,
      "flos": 61668892782720.0,
      "grad_norm": 0.9175789158310258,
      "language_loss": 0.52284563,
      "learning_rate": 3.999620763968535e-06,
      "loss": 0.54555935,
      "num_input_tokens_seen": 6329965,
      "step": 300,
      "time_per_iteration": 2.944399833679199
    },
    {
      "auxiliary_loss_clip": 0.0132123,
      "auxiliary_loss_mlp": 0.01065524,
      "balance_loss_clip": 1.09450316,
      "balance_loss_mlp": 1.05258417,
      "epoch": 0.03619311008236638,
      "flos": 27819062991360.0,
      "grad_norm": 1.8334650054517432,
      "language_loss": 0.86413467,
      "learning_rate": 3.999605443336638e-06,
      "loss": 0.88800228,
      "num_input_tokens_seen": 6352095,
      "step": 301,
      "time_per_iteration": 2.5617876052856445
    },
    {
      "auxiliary_loss_clip": 0.01323598,
      "auxiliary_loss_mlp": 0.01045721,
      "balance_loss_clip": 1.09255791,
      "balance_loss_mlp": 1.03193998,
      "epoch": 0.03631335297300547,
      "flos": 13621133197440.0,
      "grad_norm": 2.827447195240601,
      "language_loss": 0.8958323,
      "learning_rate": 3.999589819366185e-06,
      "loss": 0.91952544,
      "num_input_tokens_seen": 6365885,
      "step": 302,
      "time_per_iteration": 2.4725828170776367
    },
    {
      "auxiliary_loss_clip": 0.01317267,
      "auxiliary_loss_mlp": 0.01061278,
      "balance_loss_clip": 1.087165,
      "balance_loss_mlp": 1.04630578,
      "epoch": 0.036433595863644565,
      "flos": 27631788456960.0,
      "grad_norm": 2.198552807054148,
      "language_loss": 0.84808797,
      "learning_rate": 3.999573892059547e-06,
      "loss": 0.87187338,
      "num_input_tokens_seen": 6385015,
      "step": 303,
      "time_per_iteration": 3.301636219024658
    },
    {
      "auxiliary_loss_clip": 0.0132151,
      "auxiliary_loss_mlp": 0.01055352,
      "balance_loss_clip": 1.09059536,
      "balance_loss_mlp": 1.03958058,
      "epoch": 0.036553838754283655,
      "flos": 24572020314240.0,
      "grad_norm": 1.9487025598848489,
      "language_loss": 0.81258619,
      "learning_rate": 3.999557661419138e-06,
      "loss": 0.83635473,
      "num_input_tokens_seen": 6405165,
      "step": 304,
      "time_per_iteration": 2.574895143508911
    },
    {
      "auxiliary_loss_clip": 0.01324041,
      "auxiliary_loss_mlp": 0.01047224,
      "balance_loss_clip": 1.09500861,
      "balance_loss_mlp": 1.03371191,
      "epoch": 0.036674081644922744,
      "flos": 23404313076480.0,
      "grad_norm": 1.8242542276275375,
      "language_loss": 0.81441307,
      "learning_rate": 3.9995411274474225e-06,
      "loss": 0.83812571,
      "num_input_tokens_seen": 6424445,
      "step": 305,
      "time_per_iteration": 4.004798650741577
    },
    {
      "auxiliary_loss_clip": 0.0132233,
      "auxiliary_loss_mlp": 0.01052237,
      "balance_loss_clip": 1.09319925,
      "balance_loss_mlp": 1.03687048,
      "epoch": 0.036794324535561834,
      "flos": 27489690253440.0,
      "grad_norm": 2.0142004482064397,
      "language_loss": 0.81508338,
      "learning_rate": 3.999524290146908e-06,
      "loss": 0.83882904,
      "num_input_tokens_seen": 6444650,
      "step": 306,
      "time_per_iteration": 3.3743178844451904
    },
    {
      "auxiliary_loss_clip": 0.0132331,
      "auxiliary_loss_mlp": 0.01049878,
      "balance_loss_clip": 1.09614086,
      "balance_loss_mlp": 1.03555536,
      "epoch": 0.036914567426200924,
      "flos": 19463476227840.0,
      "grad_norm": 2.1339219922187103,
      "language_loss": 0.92245024,
      "learning_rate": 3.9995071495201485e-06,
      "loss": 0.94618213,
      "num_input_tokens_seen": 6461755,
      "step": 307,
      "time_per_iteration": 2.5409111976623535
    },
    {
      "auxiliary_loss_clip": 0.01319063,
      "auxiliary_loss_mlp": 0.01057471,
      "balance_loss_clip": 1.09215701,
      "balance_loss_mlp": 1.04300463,
      "epoch": 0.037034810316840014,
      "flos": 22309324922880.0,
      "grad_norm": 2.6638064577893363,
      "language_loss": 0.97833025,
      "learning_rate": 3.999489705569744e-06,
      "loss": 1.00209558,
      "num_input_tokens_seen": 6479455,
      "step": 308,
      "time_per_iteration": 2.5393967628479004
    },
    {
      "auxiliary_loss_clip": 0.01315008,
      "auxiliary_loss_mlp": 0.01052987,
      "balance_loss_clip": 1.08826971,
      "balance_loss_mlp": 1.03765678,
      "epoch": 0.03715505320747911,
      "flos": 18588333265920.0,
      "grad_norm": 2.302844709494629,
      "language_loss": 0.8655858,
      "learning_rate": 3.999471958298341e-06,
      "loss": 0.88926584,
      "num_input_tokens_seen": 6498365,
      "step": 309,
      "time_per_iteration": 2.5324020385742188
    },
    {
      "auxiliary_loss_clip": 0.01324175,
      "auxiliary_loss_mlp": 0.0105484,
      "balance_loss_clip": 1.09257567,
      "balance_loss_mlp": 1.03959358,
      "epoch": 0.0372752960981182,
      "flos": 35955343267200.0,
      "grad_norm": 1.7614688527919695,
      "language_loss": 0.76180172,
      "learning_rate": 3.999453907708631e-06,
      "loss": 0.78559184,
      "num_input_tokens_seen": 6520770,
      "step": 310,
      "time_per_iteration": 2.6303646564483643
    },
    {
      "auxiliary_loss_clip": 0.01317565,
      "auxiliary_loss_mlp": 0.01056307,
      "balance_loss_clip": 1.09066117,
      "balance_loss_mlp": 1.04166198,
      "epoch": 0.03739553898875729,
      "flos": 20814040627200.0,
      "grad_norm": 2.113059096009357,
      "language_loss": 0.81386268,
      "learning_rate": 3.999435553803353e-06,
      "loss": 0.83760142,
      "num_input_tokens_seen": 6540170,
      "step": 311,
      "time_per_iteration": 2.5118112564086914
    },
    {
      "auxiliary_loss_clip": 0.01316403,
      "auxiliary_loss_mlp": 0.01058637,
      "balance_loss_clip": 1.08986211,
      "balance_loss_mlp": 1.04271078,
      "epoch": 0.03751578187939638,
      "flos": 20264140339200.0,
      "grad_norm": 2.7524671069512125,
      "language_loss": 0.83355778,
      "learning_rate": 3.999416896585292e-06,
      "loss": 0.85730821,
      "num_input_tokens_seen": 6557200,
      "step": 312,
      "time_per_iteration": 2.5100231170654297
    },
    {
      "auxiliary_loss_clip": 0.01317708,
      "auxiliary_loss_mlp": 0.01054512,
      "balance_loss_clip": 1.09037626,
      "balance_loss_mlp": 1.04100585,
      "epoch": 0.03763602477003547,
      "flos": 20668063754880.0,
      "grad_norm": 4.403190873109443,
      "language_loss": 0.85844338,
      "learning_rate": 3.9993979360572775e-06,
      "loss": 0.88216555,
      "num_input_tokens_seen": 6577340,
      "step": 313,
      "time_per_iteration": 2.538710355758667
    },
    {
      "auxiliary_loss_clip": 0.01327224,
      "auxiliary_loss_mlp": 0.01054335,
      "balance_loss_clip": 1.09789264,
      "balance_loss_mlp": 1.03933239,
      "epoch": 0.03775626766067456,
      "flos": 16691352197760.0,
      "grad_norm": 2.8007867887371876,
      "language_loss": 0.82909524,
      "learning_rate": 3.999378672222185e-06,
      "loss": 0.85291082,
      "num_input_tokens_seen": 6595125,
      "step": 314,
      "time_per_iteration": 2.4923055171966553
    },
    {
      "auxiliary_loss_clip": 0.01316873,
      "auxiliary_loss_mlp": 0.01059032,
      "balance_loss_clip": 1.09186316,
      "balance_loss_mlp": 1.04442859,
      "epoch": 0.03787651055131366,
      "flos": 21141797253120.0,
      "grad_norm": 2.0383167509445723,
      "language_loss": 0.82744318,
      "learning_rate": 3.9993591050829385e-06,
      "loss": 0.85120225,
      "num_input_tokens_seen": 6612990,
      "step": 315,
      "time_per_iteration": 2.517538547515869
    },
    {
      "auxiliary_loss_clip": 0.01318752,
      "auxiliary_loss_mlp": 0.01063091,
      "balance_loss_clip": 1.09249866,
      "balance_loss_mlp": 1.04840457,
      "epoch": 0.037996753441952746,
      "flos": 22018089450240.0,
      "grad_norm": 2.1169643942368284,
      "language_loss": 0.79442149,
      "learning_rate": 3.999339234642506e-06,
      "loss": 0.81823993,
      "num_input_tokens_seen": 6632740,
      "step": 316,
      "time_per_iteration": 2.515993595123291
    },
    {
      "auxiliary_loss_clip": 0.01318907,
      "auxiliary_loss_mlp": 0.01050864,
      "balance_loss_clip": 1.09165907,
      "balance_loss_mlp": 1.0359869,
      "epoch": 0.038116996332591836,
      "flos": 27709391790720.0,
      "grad_norm": 1.9164393816485623,
      "language_loss": 0.83559942,
      "learning_rate": 3.9993190609038994e-06,
      "loss": 0.8592971,
      "num_input_tokens_seen": 6651505,
      "step": 317,
      "time_per_iteration": 2.5499019622802734
    },
    {
      "auxiliary_loss_clip": 0.0131668,
      "auxiliary_loss_mlp": 0.01058768,
      "balance_loss_clip": 1.09112442,
      "balance_loss_mlp": 1.04456484,
      "epoch": 0.038237239223230926,
      "flos": 21178067011200.0,
      "grad_norm": 1.838685332448648,
      "language_loss": 0.83224481,
      "learning_rate": 3.999298583870182e-06,
      "loss": 0.85599929,
      "num_input_tokens_seen": 6671090,
      "step": 318,
      "time_per_iteration": 2.5035042762756348
    },
    {
      "auxiliary_loss_clip": 0.01319901,
      "auxiliary_loss_mlp": 0.01055421,
      "balance_loss_clip": 1.09371233,
      "balance_loss_mlp": 1.04067469,
      "epoch": 0.038357482113870016,
      "flos": 25556618995200.0,
      "grad_norm": 1.8943434422359402,
      "language_loss": 0.77685165,
      "learning_rate": 3.999277803544458e-06,
      "loss": 0.80060482,
      "num_input_tokens_seen": 6691245,
      "step": 319,
      "time_per_iteration": 2.5568432807922363
    },
    {
      "auxiliary_loss_clip": 0.01233894,
      "auxiliary_loss_mlp": 0.01031339,
      "balance_loss_clip": 1.08934855,
      "balance_loss_mlp": 1.024997,
      "epoch": 0.038477725004509106,
      "flos": 59227578034560.0,
      "grad_norm": 0.9513375427893127,
      "language_loss": 0.62348098,
      "learning_rate": 3.999256719929882e-06,
      "loss": 0.64613324,
      "num_input_tokens_seen": 6752520,
      "step": 320,
      "time_per_iteration": 3.072059154510498
    },
    {
      "auxiliary_loss_clip": 0.01231506,
      "auxiliary_loss_mlp": 0.01018007,
      "balance_loss_clip": 1.08758235,
      "balance_loss_mlp": 1.01176012,
      "epoch": 0.0385979678951482,
      "flos": 67317676398720.0,
      "grad_norm": 1.2112025694047706,
      "language_loss": 0.67128277,
      "learning_rate": 3.999235333029651e-06,
      "loss": 0.69377792,
      "num_input_tokens_seen": 6806460,
      "step": 321,
      "time_per_iteration": 3.005730390548706
    },
    {
      "auxiliary_loss_clip": 0.01318229,
      "auxiliary_loss_mlp": 0.01057503,
      "balance_loss_clip": 1.09297633,
      "balance_loss_mlp": 1.04427636,
      "epoch": 0.03871821078578729,
      "flos": 22746752749440.0,
      "grad_norm": 1.8296146687539867,
      "language_loss": 0.82062113,
      "learning_rate": 3.999213642847009e-06,
      "loss": 0.84437847,
      "num_input_tokens_seen": 6827045,
      "step": 322,
      "time_per_iteration": 2.5541253089904785
    },
    {
      "auxiliary_loss_clip": 0.01319212,
      "auxiliary_loss_mlp": 0.01051964,
      "balance_loss_clip": 1.09297347,
      "balance_loss_mlp": 1.03786135,
      "epoch": 0.03883845367642638,
      "flos": 26280613526400.0,
      "grad_norm": 1.6981699957530276,
      "language_loss": 0.90994018,
      "learning_rate": 3.999191649385247e-06,
      "loss": 0.93365192,
      "num_input_tokens_seen": 6848220,
      "step": 323,
      "time_per_iteration": 2.5618481636047363
    },
    {
      "auxiliary_loss_clip": 0.01225398,
      "auxiliary_loss_mlp": 0.01014446,
      "balance_loss_clip": 1.08278096,
      "balance_loss_mlp": 1.00834227,
      "epoch": 0.03895869656706547,
      "flos": 56962835568000.0,
      "grad_norm": 0.9110791348930323,
      "language_loss": 0.59806377,
      "learning_rate": 3.999169352647702e-06,
      "loss": 0.62046224,
      "num_input_tokens_seen": 6909400,
      "step": 324,
      "time_per_iteration": 3.046316385269165
    },
    {
      "auxiliary_loss_clip": 0.01319852,
      "auxiliary_loss_mlp": 0.01058214,
      "balance_loss_clip": 1.09322512,
      "balance_loss_mlp": 1.04365218,
      "epoch": 0.03907893945770456,
      "flos": 24863363527680.0,
      "grad_norm": 1.7755059170984826,
      "language_loss": 0.82945853,
      "learning_rate": 3.999146752637755e-06,
      "loss": 0.85323918,
      "num_input_tokens_seen": 6930445,
      "step": 325,
      "time_per_iteration": 2.539952278137207
    },
    {
      "auxiliary_loss_clip": 0.013157,
      "auxiliary_loss_mlp": 0.01051247,
      "balance_loss_clip": 1.08970344,
      "balance_loss_mlp": 1.03590441,
      "epoch": 0.03919918234834365,
      "flos": 18368595815040.0,
      "grad_norm": 2.462607271883047,
      "language_loss": 0.89657652,
      "learning_rate": 3.999123849358836e-06,
      "loss": 0.92024595,
      "num_input_tokens_seen": 6948110,
      "step": 326,
      "time_per_iteration": 2.504488229751587
    },
    {
      "auxiliary_loss_clip": 0.01312834,
      "auxiliary_loss_mlp": 0.0105805,
      "balance_loss_clip": 1.08887684,
      "balance_loss_mlp": 1.04446626,
      "epoch": 0.03931942523898275,
      "flos": 25225414663680.0,
      "grad_norm": 1.9864590551152903,
      "language_loss": 0.74770623,
      "learning_rate": 3.999100642814418e-06,
      "loss": 0.77141511,
      "num_input_tokens_seen": 6968550,
      "step": 327,
      "time_per_iteration": 2.533271074295044
    },
    {
      "auxiliary_loss_clip": 0.01314575,
      "auxiliary_loss_mlp": 0.01065208,
      "balance_loss_clip": 1.09086835,
      "balance_loss_mlp": 1.05058134,
      "epoch": 0.03943966812962184,
      "flos": 23257905240960.0,
      "grad_norm": 2.4962273177387804,
      "language_loss": 0.88747549,
      "learning_rate": 3.999077133008022e-06,
      "loss": 0.91127336,
      "num_input_tokens_seen": 6987135,
      "step": 328,
      "time_per_iteration": 2.533186674118042
    },
    {
      "auxiliary_loss_clip": 0.01318609,
      "auxiliary_loss_mlp": 0.01061055,
      "balance_loss_clip": 1.09328806,
      "balance_loss_mlp": 1.04616582,
      "epoch": 0.03955991102026093,
      "flos": 29168837291520.0,
      "grad_norm": 1.7605090985753582,
      "language_loss": 0.90584797,
      "learning_rate": 3.9990533199432145e-06,
      "loss": 0.92964464,
      "num_input_tokens_seen": 7008630,
      "step": 329,
      "time_per_iteration": 3.349200487136841
    },
    {
      "auxiliary_loss_clip": 0.01316768,
      "auxiliary_loss_mlp": 0.01048766,
      "balance_loss_clip": 1.09302616,
      "balance_loss_mlp": 1.03388894,
      "epoch": 0.03968015391090002,
      "flos": 17602441695360.0,
      "grad_norm": 2.28323905735946,
      "language_loss": 0.75842416,
      "learning_rate": 3.999029203623608e-06,
      "loss": 0.78207958,
      "num_input_tokens_seen": 7026350,
      "step": 330,
      "time_per_iteration": 2.5184061527252197
    },
    {
      "auxiliary_loss_clip": 0.01316288,
      "auxiliary_loss_mlp": 0.0106005,
      "balance_loss_clip": 1.093889,
      "balance_loss_mlp": 1.04597139,
      "epoch": 0.03980039680153911,
      "flos": 21799285752960.0,
      "grad_norm": 1.9373879238885332,
      "language_loss": 0.86733139,
      "learning_rate": 3.99900478405286e-06,
      "loss": 0.8910948,
      "num_input_tokens_seen": 7045660,
      "step": 331,
      "time_per_iteration": 3.2127890586853027
    },
    {
      "auxiliary_loss_clip": 0.01318631,
      "auxiliary_loss_mlp": 0.01057384,
      "balance_loss_clip": 1.09571314,
      "balance_loss_mlp": 1.04310274,
      "epoch": 0.0399206396921782,
      "flos": 15195134148480.0,
      "grad_norm": 2.2008021697812667,
      "language_loss": 0.82537913,
      "learning_rate": 3.998980061234676e-06,
      "loss": 0.84913921,
      "num_input_tokens_seen": 7063575,
      "step": 332,
      "time_per_iteration": 4.526222467422485
    },
    {
      "auxiliary_loss_clip": 0.0131322,
      "auxiliary_loss_mlp": 0.01053636,
      "balance_loss_clip": 1.09026265,
      "balance_loss_mlp": 1.0388546,
      "epoch": 0.040040882582817294,
      "flos": 14422910630400.0,
      "grad_norm": 2.5565098907307484,
      "language_loss": 0.7565695,
      "learning_rate": 3.9989550351728055e-06,
      "loss": 0.78023803,
      "num_input_tokens_seen": 7080505,
      "step": 333,
      "time_per_iteration": 2.531641721725464
    },
    {
      "auxiliary_loss_clip": 0.01310734,
      "auxiliary_loss_mlp": 0.01049518,
      "balance_loss_clip": 1.09068918,
      "balance_loss_mlp": 1.0348376,
      "epoch": 0.040161125473456384,
      "flos": 19280906375040.0,
      "grad_norm": 2.596717113650031,
      "language_loss": 0.8464781,
      "learning_rate": 3.998929705871046e-06,
      "loss": 0.87008059,
      "num_input_tokens_seen": 7097860,
      "step": 334,
      "time_per_iteration": 2.545999050140381
    },
    {
      "auxiliary_loss_clip": 0.01314636,
      "auxiliary_loss_mlp": 0.01053243,
      "balance_loss_clip": 1.09407616,
      "balance_loss_mlp": 1.03890789,
      "epoch": 0.040281368364095474,
      "flos": 17821101738240.0,
      "grad_norm": 2.356084721211621,
      "language_loss": 0.89058423,
      "learning_rate": 3.99890407333324e-06,
      "loss": 0.91426301,
      "num_input_tokens_seen": 7116390,
      "step": 335,
      "time_per_iteration": 2.514615297317505
    },
    {
      "auxiliary_loss_clip": 0.01313341,
      "auxiliary_loss_mlp": 0.01065751,
      "balance_loss_clip": 1.09012496,
      "balance_loss_mlp": 1.05113578,
      "epoch": 0.040401611254734564,
      "flos": 19573757959680.0,
      "grad_norm": 1.6977392657812431,
      "language_loss": 0.86971194,
      "learning_rate": 3.998878137563275e-06,
      "loss": 0.89350283,
      "num_input_tokens_seen": 7135940,
      "step": 336,
      "time_per_iteration": 2.5181312561035156
    },
    {
      "auxiliary_loss_clip": 0.01313754,
      "auxiliary_loss_mlp": 0.0105063,
      "balance_loss_clip": 1.09159148,
      "balance_loss_mlp": 1.03569269,
      "epoch": 0.040521854145373654,
      "flos": 22054466949120.0,
      "grad_norm": 2.2703484025276937,
      "language_loss": 0.85037673,
      "learning_rate": 3.998851898565085e-06,
      "loss": 0.87402058,
      "num_input_tokens_seen": 7155745,
      "step": 337,
      "time_per_iteration": 2.528578281402588
    },
    {
      "auxiliary_loss_clip": 0.01311248,
      "auxiliary_loss_mlp": 0.01054638,
      "balance_loss_clip": 1.08934045,
      "balance_loss_mlp": 1.0408392,
      "epoch": 0.04064209703601274,
      "flos": 22674644196480.0,
      "grad_norm": 1.9780194479753836,
      "language_loss": 0.83019316,
      "learning_rate": 3.998825356342653e-06,
      "loss": 0.85385203,
      "num_input_tokens_seen": 7175920,
      "step": 338,
      "time_per_iteration": 2.520033359527588
    },
    {
      "auxiliary_loss_clip": 0.01310582,
      "auxiliary_loss_mlp": 0.01055805,
      "balance_loss_clip": 1.08977628,
      "balance_loss_mlp": 1.04192293,
      "epoch": 0.04076233992665183,
      "flos": 38582172783360.0,
      "grad_norm": 2.1946952760656755,
      "language_loss": 0.7307989,
      "learning_rate": 3.998798510900003e-06,
      "loss": 0.75446284,
      "num_input_tokens_seen": 7198720,
      "step": 339,
      "time_per_iteration": 2.6788558959960938
    },
    {
      "auxiliary_loss_clip": 0.01306239,
      "auxiliary_loss_mlp": 0.01062111,
      "balance_loss_clip": 1.08742023,
      "balance_loss_mlp": 1.04769886,
      "epoch": 0.04088258281729093,
      "flos": 25885309374720.0,
      "grad_norm": 2.2647429576224005,
      "language_loss": 0.83690643,
      "learning_rate": 3.998771362241207e-06,
      "loss": 0.86058992,
      "num_input_tokens_seen": 7219125,
      "step": 340,
      "time_per_iteration": 2.5736894607543945
    },
    {
      "auxiliary_loss_clip": 0.01305508,
      "auxiliary_loss_mlp": 0.01062149,
      "balance_loss_clip": 1.08711779,
      "balance_loss_mlp": 1.04747462,
      "epoch": 0.04100282570793002,
      "flos": 19789832223360.0,
      "grad_norm": 1.8026651717589495,
      "language_loss": 0.8805961,
      "learning_rate": 3.998743910370385e-06,
      "loss": 0.90427274,
      "num_input_tokens_seen": 7237985,
      "step": 341,
      "time_per_iteration": 2.5231807231903076
    },
    {
      "auxiliary_loss_clip": 0.01314185,
      "auxiliary_loss_mlp": 0.01071864,
      "balance_loss_clip": 1.09502923,
      "balance_loss_mlp": 1.05745149,
      "epoch": 0.04112306859856911,
      "flos": 22565152563840.0,
      "grad_norm": 2.005214437669453,
      "language_loss": 0.73053145,
      "learning_rate": 3.998716155291702e-06,
      "loss": 0.75439191,
      "num_input_tokens_seen": 7255825,
      "step": 342,
      "time_per_iteration": 2.5463719367980957
    },
    {
      "auxiliary_loss_clip": 0.01314291,
      "auxiliary_loss_mlp": 0.01056984,
      "balance_loss_clip": 1.09403527,
      "balance_loss_mlp": 1.04298902,
      "epoch": 0.0412433114892082,
      "flos": 25040654081280.0,
      "grad_norm": 2.567936122791699,
      "language_loss": 0.90308261,
      "learning_rate": 3.998688097009366e-06,
      "loss": 0.92679548,
      "num_input_tokens_seen": 7276590,
      "step": 343,
      "time_per_iteration": 2.5754597187042236
    },
    {
      "auxiliary_loss_clip": 0.01307098,
      "auxiliary_loss_mlp": 0.01062659,
      "balance_loss_clip": 1.08946478,
      "balance_loss_mlp": 1.04778171,
      "epoch": 0.04136355437984729,
      "flos": 25191371548800.0,
      "grad_norm": 2.075241516392257,
      "language_loss": 0.80355072,
      "learning_rate": 3.998659735527636e-06,
      "loss": 0.82724833,
      "num_input_tokens_seen": 7295680,
      "step": 344,
      "time_per_iteration": 2.552063465118408
    },
    {
      "auxiliary_loss_clip": 0.013087,
      "auxiliary_loss_mlp": 0.01055991,
      "balance_loss_clip": 1.08985329,
      "balance_loss_mlp": 1.04239511,
      "epoch": 0.04148379727048638,
      "flos": 22966777509120.0,
      "grad_norm": 1.6717548354628131,
      "language_loss": 0.77905518,
      "learning_rate": 3.998631070850813e-06,
      "loss": 0.80270207,
      "num_input_tokens_seen": 7316300,
      "step": 345,
      "time_per_iteration": 2.5350868701934814
    },
    {
      "auxiliary_loss_clip": 0.01308274,
      "auxiliary_loss_mlp": 0.01058827,
      "balance_loss_clip": 1.0912683,
      "balance_loss_mlp": 1.04557133,
      "epoch": 0.041604040161125476,
      "flos": 14063481187200.0,
      "grad_norm": 2.6215785120684765,
      "language_loss": 0.8351419,
      "learning_rate": 3.9986021029832455e-06,
      "loss": 0.85881293,
      "num_input_tokens_seen": 7333615,
      "step": 346,
      "time_per_iteration": 2.5036513805389404
    },
    {
      "auxiliary_loss_clip": 0.01308856,
      "auxiliary_loss_mlp": 0.01051486,
      "balance_loss_clip": 1.08974135,
      "balance_loss_mlp": 1.03687668,
      "epoch": 0.041724283051764566,
      "flos": 12091877614080.0,
      "grad_norm": 3.4041727436159155,
      "language_loss": 0.91711819,
      "learning_rate": 3.9985728319293285e-06,
      "loss": 0.94072163,
      "num_input_tokens_seen": 7347590,
      "step": 347,
      "time_per_iteration": 2.5365169048309326
    },
    {
      "auxiliary_loss_clip": 0.01307541,
      "auxiliary_loss_mlp": 0.01047657,
      "balance_loss_clip": 1.08506775,
      "balance_loss_mlp": 1.03229117,
      "epoch": 0.041844525942403656,
      "flos": 12385303816320.0,
      "grad_norm": 13.071677555497837,
      "language_loss": 0.85614735,
      "learning_rate": 3.998543257693501e-06,
      "loss": 0.87969935,
      "num_input_tokens_seen": 7364345,
      "step": 348,
      "time_per_iteration": 2.501319169998169
    },
    {
      "auxiliary_loss_clip": 0.01310431,
      "auxiliary_loss_mlp": 0.01063926,
      "balance_loss_clip": 1.09192967,
      "balance_loss_mlp": 1.05043769,
      "epoch": 0.041964768833042745,
      "flos": 23769345041280.0,
      "grad_norm": 1.6547325953331997,
      "language_loss": 0.87745696,
      "learning_rate": 3.998513380280251e-06,
      "loss": 0.90120053,
      "num_input_tokens_seen": 7384625,
      "step": 349,
      "time_per_iteration": 2.5608365535736084
    },
    {
      "auxiliary_loss_clip": 0.0131299,
      "auxiliary_loss_mlp": 0.01065638,
      "balance_loss_clip": 1.09318435,
      "balance_loss_mlp": 1.05107117,
      "epoch": 0.042085011723681835,
      "flos": 11875336473600.0,
      "grad_norm": 2.26618502574991,
      "language_loss": 0.94824517,
      "learning_rate": 3.99848319969411e-06,
      "loss": 0.97203147,
      "num_input_tokens_seen": 7402225,
      "step": 350,
      "time_per_iteration": 2.504929780960083
    },
    {
      "auxiliary_loss_clip": 0.01312867,
      "auxiliary_loss_mlp": 0.0107688,
      "balance_loss_clip": 1.09304142,
      "balance_loss_mlp": 1.06300998,
      "epoch": 0.042205254614320925,
      "flos": 16873957964160.0,
      "grad_norm": 2.0352777124850583,
      "language_loss": 0.79412478,
      "learning_rate": 3.9984527159396564e-06,
      "loss": 0.81802225,
      "num_input_tokens_seen": 7420865,
      "step": 351,
      "time_per_iteration": 2.5298609733581543
    },
    {
      "auxiliary_loss_clip": 0.01303921,
      "auxiliary_loss_mlp": 0.01057811,
      "balance_loss_clip": 1.0849191,
      "balance_loss_mlp": 1.04366136,
      "epoch": 0.04232549750496002,
      "flos": 25118508810240.0,
      "grad_norm": 2.067214404279356,
      "language_loss": 0.84328985,
      "learning_rate": 3.9984219290215154e-06,
      "loss": 0.86690724,
      "num_input_tokens_seen": 7441040,
      "step": 352,
      "time_per_iteration": 2.55487322807312
    },
    {
      "auxiliary_loss_clip": 0.01305516,
      "auxiliary_loss_mlp": 0.01053343,
      "balance_loss_clip": 1.08811975,
      "balance_loss_mlp": 1.04074883,
      "epoch": 0.04244574039559911,
      "flos": 26724541714560.0,
      "grad_norm": 1.616194315758387,
      "language_loss": 0.89214641,
      "learning_rate": 3.998390838944356e-06,
      "loss": 0.91573501,
      "num_input_tokens_seen": 7462545,
      "step": 353,
      "time_per_iteration": 2.5857582092285156
    },
    {
      "auxiliary_loss_clip": 0.01307244,
      "auxiliary_loss_mlp": 0.01060927,
      "balance_loss_clip": 1.08902609,
      "balance_loss_mlp": 1.04650307,
      "epoch": 0.0425659832862382,
      "flos": 20923244951040.0,
      "grad_norm": 2.409808096594608,
      "language_loss": 0.90238363,
      "learning_rate": 3.998359445712895e-06,
      "loss": 0.92606533,
      "num_input_tokens_seen": 7481650,
      "step": 354,
      "time_per_iteration": 2.5512373447418213
    },
    {
      "auxiliary_loss_clip": 0.01305481,
      "auxiliary_loss_mlp": 0.01056678,
      "balance_loss_clip": 1.08575869,
      "balance_loss_mlp": 1.04201555,
      "epoch": 0.04268622617687729,
      "flos": 23331127115520.0,
      "grad_norm": 2.097075536437559,
      "language_loss": 0.81085157,
      "learning_rate": 3.9983277493318955e-06,
      "loss": 0.83447313,
      "num_input_tokens_seen": 7500945,
      "step": 355,
      "time_per_iteration": 3.2719600200653076
    },
    {
      "auxiliary_loss_clip": 0.01308053,
      "auxiliary_loss_mlp": 0.01054099,
      "balance_loss_clip": 1.08785927,
      "balance_loss_mlp": 1.03860188,
      "epoch": 0.04280646906751638,
      "flos": 25994010908160.0,
      "grad_norm": 2.7928955387437213,
      "language_loss": 0.81115079,
      "learning_rate": 3.998295749806165e-06,
      "loss": 0.83477235,
      "num_input_tokens_seen": 7522170,
      "step": 356,
      "time_per_iteration": 2.5935537815093994
    },
    {
      "auxiliary_loss_clip": 0.01307585,
      "auxiliary_loss_mlp": 0.01064414,
      "balance_loss_clip": 1.0898962,
      "balance_loss_mlp": 1.05015612,
      "epoch": 0.04292671195815547,
      "flos": 26906824258560.0,
      "grad_norm": 2.437499876955015,
      "language_loss": 0.83409923,
      "learning_rate": 3.998263447140558e-06,
      "loss": 0.8578192,
      "num_input_tokens_seen": 7542370,
      "step": 357,
      "time_per_iteration": 3.286508321762085
    },
    {
      "auxiliary_loss_clip": 0.01301079,
      "auxiliary_loss_mlp": 0.01057778,
      "balance_loss_clip": 1.08446372,
      "balance_loss_mlp": 1.0431397,
      "epoch": 0.04304695484879457,
      "flos": 39457315745280.0,
      "grad_norm": 1.8402279960007148,
      "language_loss": 0.81688887,
      "learning_rate": 3.998230841339976e-06,
      "loss": 0.84047747,
      "num_input_tokens_seen": 7564380,
      "step": 358,
      "time_per_iteration": 2.6848254203796387
    },
    {
      "auxiliary_loss_clip": 0.01304409,
      "auxiliary_loss_mlp": 0.01060019,
      "balance_loss_clip": 1.08799839,
      "balance_loss_mlp": 1.04742432,
      "epoch": 0.04316719773943366,
      "flos": 19646297475840.0,
      "grad_norm": 2.2890338487867936,
      "language_loss": 0.85276091,
      "learning_rate": 3.998197932409363e-06,
      "loss": 0.87640524,
      "num_input_tokens_seen": 7582390,
      "step": 359,
      "time_per_iteration": 4.163724422454834
    },
    {
      "auxiliary_loss_clip": 0.01306656,
      "auxiliary_loss_mlp": 0.01056826,
      "balance_loss_clip": 1.09133101,
      "balance_loss_mlp": 1.04312336,
      "epoch": 0.04328744063007275,
      "flos": 22452320966400.0,
      "grad_norm": 2.8074454328020577,
      "language_loss": 0.8607778,
      "learning_rate": 3.9981647203537125e-06,
      "loss": 0.88441265,
      "num_input_tokens_seen": 7599890,
      "step": 360,
      "time_per_iteration": 2.5552849769592285
    },
    {
      "auxiliary_loss_clip": 0.01306705,
      "auxiliary_loss_mlp": 0.01053239,
      "balance_loss_clip": 1.0887748,
      "balance_loss_mlp": 1.040025,
      "epoch": 0.04340768352071184,
      "flos": 21283033530240.0,
      "grad_norm": 2.2480959030467185,
      "language_loss": 0.95519155,
      "learning_rate": 3.998131205178063e-06,
      "loss": 0.978791,
      "num_input_tokens_seen": 7618360,
      "step": 361,
      "time_per_iteration": 2.648998737335205
    },
    {
      "auxiliary_loss_clip": 0.01305118,
      "auxiliary_loss_mlp": 0.01055471,
      "balance_loss_clip": 1.08607912,
      "balance_loss_mlp": 1.04102266,
      "epoch": 0.04352792641135093,
      "flos": 11583705951360.0,
      "grad_norm": 2.4925768614398502,
      "language_loss": 0.76655912,
      "learning_rate": 3.998097386887498e-06,
      "loss": 0.79016501,
      "num_input_tokens_seen": 7635435,
      "step": 362,
      "time_per_iteration": 2.5709354877471924
    },
    {
      "auxiliary_loss_clip": 0.01299339,
      "auxiliary_loss_mlp": 0.01058494,
      "balance_loss_clip": 1.08592558,
      "balance_loss_mlp": 1.04473782,
      "epoch": 0.04364816930199002,
      "flos": 23623547736960.0,
      "grad_norm": 1.732202133324105,
      "language_loss": 0.84658849,
      "learning_rate": 3.998063265487148e-06,
      "loss": 0.8701669,
      "num_input_tokens_seen": 7656485,
      "step": 363,
      "time_per_iteration": 2.6171112060546875
    },
    {
      "auxiliary_loss_clip": 0.01302672,
      "auxiliary_loss_mlp": 0.01060728,
      "balance_loss_clip": 1.08827174,
      "balance_loss_mlp": 1.04729342,
      "epoch": 0.043768412192629114,
      "flos": 14429734214400.0,
      "grad_norm": 1.828003254274323,
      "language_loss": 0.80775303,
      "learning_rate": 3.99802884098219e-06,
      "loss": 0.83138704,
      "num_input_tokens_seen": 7674595,
      "step": 364,
      "time_per_iteration": 2.5265486240386963
    },
    {
      "auxiliary_loss_clip": 0.01302109,
      "auxiliary_loss_mlp": 0.01048308,
      "balance_loss_clip": 1.08543992,
      "balance_loss_mlp": 1.03394306,
      "epoch": 0.043888655083268203,
      "flos": 26468893641600.0,
      "grad_norm": 2.166315302224309,
      "language_loss": 0.82275569,
      "learning_rate": 3.997994113377845e-06,
      "loss": 0.84625983,
      "num_input_tokens_seen": 7693495,
      "step": 365,
      "time_per_iteration": 2.613940477371216
    },
    {
      "auxiliary_loss_clip": 0.01301036,
      "auxiliary_loss_mlp": 0.01057627,
      "balance_loss_clip": 1.08585835,
      "balance_loss_mlp": 1.04401958,
      "epoch": 0.04400889797390729,
      "flos": 27235263242880.0,
      "grad_norm": 2.107104815608086,
      "language_loss": 0.83271766,
      "learning_rate": 3.9979590826793815e-06,
      "loss": 0.85630423,
      "num_input_tokens_seen": 7714685,
      "step": 366,
      "time_per_iteration": 2.5660226345062256
    },
    {
      "auxiliary_loss_clip": 0.01306218,
      "auxiliary_loss_mlp": 0.01051847,
      "balance_loss_clip": 1.09065711,
      "balance_loss_mlp": 1.0377264,
      "epoch": 0.04412914086454638,
      "flos": 20119528183680.0,
      "grad_norm": 2.4757305366761995,
      "language_loss": 0.81087899,
      "learning_rate": 3.997923748892113e-06,
      "loss": 0.8344596,
      "num_input_tokens_seen": 7734005,
      "step": 367,
      "time_per_iteration": 2.5656020641326904
    },
    {
      "auxiliary_loss_clip": 0.0129909,
      "auxiliary_loss_mlp": 0.01057027,
      "balance_loss_clip": 1.08692169,
      "balance_loss_mlp": 1.04358602,
      "epoch": 0.04424938375518547,
      "flos": 22604618632320.0,
      "grad_norm": 1.7416551695515288,
      "language_loss": 0.88569885,
      "learning_rate": 3.9978881120214015e-06,
      "loss": 0.90926003,
      "num_input_tokens_seen": 7755525,
      "step": 368,
      "time_per_iteration": 2.583942413330078
    },
    {
      "auxiliary_loss_clip": 0.01300208,
      "auxiliary_loss_mlp": 0.01058333,
      "balance_loss_clip": 1.0837276,
      "balance_loss_mlp": 1.04343176,
      "epoch": 0.04436962664582456,
      "flos": 24132365844480.0,
      "grad_norm": 1.8899362160155027,
      "language_loss": 0.79284316,
      "learning_rate": 3.997852172072652e-06,
      "loss": 0.81642854,
      "num_input_tokens_seen": 7776740,
      "step": 369,
      "time_per_iteration": 2.5564613342285156
    },
    {
      "auxiliary_loss_clip": 0.01307351,
      "auxiliary_loss_mlp": 0.01059,
      "balance_loss_clip": 1.08910441,
      "balance_loss_mlp": 1.04481399,
      "epoch": 0.04448986953646366,
      "flos": 18222906251520.0,
      "grad_norm": 3.1066548777172884,
      "language_loss": 0.89237714,
      "learning_rate": 3.9978159290513155e-06,
      "loss": 0.91604066,
      "num_input_tokens_seen": 7794820,
      "step": 370,
      "time_per_iteration": 2.5255751609802246
    },
    {
      "auxiliary_loss_clip": 0.01305853,
      "auxiliary_loss_mlp": 0.01050523,
      "balance_loss_clip": 1.08806777,
      "balance_loss_mlp": 1.0358839,
      "epoch": 0.04461011242710275,
      "flos": 30117920400000.0,
      "grad_norm": 2.4965919564950423,
      "language_loss": 0.80074084,
      "learning_rate": 3.997779382962892e-06,
      "loss": 0.82430458,
      "num_input_tokens_seen": 7817705,
      "step": 371,
      "time_per_iteration": 2.60319447517395
    },
    {
      "auxiliary_loss_clip": 0.01299034,
      "auxiliary_loss_mlp": 0.01047547,
      "balance_loss_clip": 1.08420706,
      "balance_loss_mlp": 1.03434455,
      "epoch": 0.04473035531774184,
      "flos": 29752529299200.0,
      "grad_norm": 2.573831449450958,
      "language_loss": 0.7379204,
      "learning_rate": 3.997742533812924e-06,
      "loss": 0.76138628,
      "num_input_tokens_seen": 7840970,
      "step": 372,
      "time_per_iteration": 2.5902652740478516
    },
    {
      "auxiliary_loss_clip": 0.01304295,
      "auxiliary_loss_mlp": 0.01056549,
      "balance_loss_clip": 1.08808279,
      "balance_loss_mlp": 1.0417912,
      "epoch": 0.04485059820838093,
      "flos": 13151565676800.0,
      "grad_norm": 3.3210445813353386,
      "language_loss": 0.92268425,
      "learning_rate": 3.997705381607001e-06,
      "loss": 0.94629264,
      "num_input_tokens_seen": 7857785,
      "step": 373,
      "time_per_iteration": 2.5294086933135986
    },
    {
      "auxiliary_loss_clip": 0.01213533,
      "auxiliary_loss_mlp": 0.01043804,
      "balance_loss_clip": 1.07620752,
      "balance_loss_mlp": 1.03843975,
      "epoch": 0.04497084109902002,
      "flos": 68094209548800.0,
      "grad_norm": 0.9762370706554626,
      "language_loss": 0.6031006,
      "learning_rate": 3.997667926350761e-06,
      "loss": 0.62567395,
      "num_input_tokens_seen": 7916115,
      "step": 374,
      "time_per_iteration": 3.0383386611938477
    },
    {
      "auxiliary_loss_clip": 0.01213675,
      "auxiliary_loss_mlp": 0.01032462,
      "balance_loss_clip": 1.0759573,
      "balance_loss_mlp": 1.02724099,
      "epoch": 0.04509108398965911,
      "flos": 64342263346560.0,
      "grad_norm": 0.897278684969327,
      "language_loss": 0.57758641,
      "learning_rate": 3.997630168049886e-06,
      "loss": 0.60004783,
      "num_input_tokens_seen": 7974480,
      "step": 375,
      "time_per_iteration": 3.140150547027588
    },
    {
      "auxiliary_loss_clip": 0.01305004,
      "auxiliary_loss_mlp": 0.01050691,
      "balance_loss_clip": 1.08762658,
      "balance_loss_mlp": 1.03651106,
      "epoch": 0.045211326880298205,
      "flos": 22271115830400.0,
      "grad_norm": 2.1872675984514025,
      "language_loss": 0.7732361,
      "learning_rate": 3.997592106710101e-06,
      "loss": 0.7967931,
      "num_input_tokens_seen": 7993940,
      "step": 376,
      "time_per_iteration": 2.5851426124572754
    },
    {
      "auxiliary_loss_clip": 0.01303004,
      "auxiliary_loss_mlp": 0.01053087,
      "balance_loss_clip": 1.08753943,
      "balance_loss_mlp": 1.0392884,
      "epoch": 0.045331569770937295,
      "flos": 32159441796480.0,
      "grad_norm": 2.62671355901087,
      "language_loss": 0.65748513,
      "learning_rate": 3.997553742337182e-06,
      "loss": 0.68104607,
      "num_input_tokens_seen": 8013365,
      "step": 377,
      "time_per_iteration": 2.6352899074554443
    },
    {
      "auxiliary_loss_clip": 0.0130018,
      "auxiliary_loss_mlp": 0.01055299,
      "balance_loss_clip": 1.08636856,
      "balance_loss_mlp": 1.04073715,
      "epoch": 0.045451812661576385,
      "flos": 22163455791360.0,
      "grad_norm": 2.0924302645276045,
      "language_loss": 0.9140929,
      "learning_rate": 3.997515074936949e-06,
      "loss": 0.9376477,
      "num_input_tokens_seen": 8034240,
      "step": 378,
      "time_per_iteration": 2.5796120166778564
    },
    {
      "auxiliary_loss_clip": 0.01303683,
      "auxiliary_loss_mlp": 0.01066653,
      "balance_loss_clip": 1.08977962,
      "balance_loss_mlp": 1.05145407,
      "epoch": 0.045572055552215475,
      "flos": 16581968305920.0,
      "grad_norm": 2.5431639632885843,
      "language_loss": 0.86978,
      "learning_rate": 3.997476104515268e-06,
      "loss": 0.8934834,
      "num_input_tokens_seen": 8052430,
      "step": 379,
      "time_per_iteration": 2.5220797061920166
    },
    {
      "auxiliary_loss_clip": 0.01301228,
      "auxiliary_loss_mlp": 0.01054584,
      "balance_loss_clip": 1.08850706,
      "balance_loss_mlp": 1.04154849,
      "epoch": 0.045692298442854565,
      "flos": 17603375448960.0,
      "grad_norm": 5.8900836238518375,
      "language_loss": 0.77856112,
      "learning_rate": 3.9974368310780485e-06,
      "loss": 0.80211926,
      "num_input_tokens_seen": 8069605,
      "step": 380,
      "time_per_iteration": 2.5133159160614014
    },
    {
      "auxiliary_loss_clip": 0.0130749,
      "auxiliary_loss_mlp": 0.01069462,
      "balance_loss_clip": 1.08843994,
      "balance_loss_mlp": 1.05482352,
      "epoch": 0.045812541333493655,
      "flos": 26761098781440.0,
      "grad_norm": 2.7837063261425756,
      "language_loss": 0.74422026,
      "learning_rate": 3.997397254631251e-06,
      "loss": 0.76798975,
      "num_input_tokens_seen": 8090225,
      "step": 381,
      "time_per_iteration": 2.6074421405792236
    },
    {
      "auxiliary_loss_clip": 0.01201274,
      "auxiliary_loss_mlp": 0.01112401,
      "balance_loss_clip": 1.06633997,
      "balance_loss_mlp": 1.10746562,
      "epoch": 0.04593278422413275,
      "flos": 60250349894400.0,
      "grad_norm": 0.8679140422610949,
      "language_loss": 0.60062426,
      "learning_rate": 3.997357375180878e-06,
      "loss": 0.623761,
      "num_input_tokens_seen": 8154505,
      "step": 382,
      "time_per_iteration": 3.8849434852600098
    },
    {
      "auxiliary_loss_clip": 0.01299759,
      "auxiliary_loss_mlp": 0.01047371,
      "balance_loss_clip": 1.08752191,
      "balance_loss_mlp": 1.03322077,
      "epoch": 0.04605302711477184,
      "flos": 21799249839360.0,
      "grad_norm": 2.0100369123222683,
      "language_loss": 0.75250077,
      "learning_rate": 3.997317192732979e-06,
      "loss": 0.77597213,
      "num_input_tokens_seen": 8173285,
      "step": 383,
      "time_per_iteration": 2.5444180965423584
    },
    {
      "auxiliary_loss_clip": 0.01305749,
      "auxiliary_loss_mlp": 0.01057162,
      "balance_loss_clip": 1.09006119,
      "balance_loss_mlp": 1.0434525,
      "epoch": 0.04617327000541093,
      "flos": 19459705299840.0,
      "grad_norm": 1.9247383853394031,
      "language_loss": 0.8271963,
      "learning_rate": 3.99727670729365e-06,
      "loss": 0.85082537,
      "num_input_tokens_seen": 8191845,
      "step": 384,
      "time_per_iteration": 3.246682643890381
    },
    {
      "auxiliary_loss_clip": 0.01305886,
      "auxiliary_loss_mlp": 0.01059306,
      "balance_loss_clip": 1.09291458,
      "balance_loss_mlp": 1.04671144,
      "epoch": 0.04629351289605002,
      "flos": 25411468135680.0,
      "grad_norm": 1.7594548493690716,
      "language_loss": 0.77771282,
      "learning_rate": 3.997235918869033e-06,
      "loss": 0.80136472,
      "num_input_tokens_seen": 8212880,
      "step": 385,
      "time_per_iteration": 3.4974958896636963
    },
    {
      "auxiliary_loss_clip": 0.01300716,
      "auxiliary_loss_mlp": 0.01047652,
      "balance_loss_clip": 1.0876348,
      "balance_loss_mlp": 1.03357959,
      "epoch": 0.04641375578668911,
      "flos": 20558284813440.0,
      "grad_norm": 1.8510229169930243,
      "language_loss": 0.82472646,
      "learning_rate": 3.997194827465315e-06,
      "loss": 0.84821022,
      "num_input_tokens_seen": 8231475,
      "step": 386,
      "time_per_iteration": 2.520608901977539
    },
    {
      "auxiliary_loss_clip": 0.01297269,
      "auxiliary_loss_mlp": 0.01054786,
      "balance_loss_clip": 1.08440113,
      "balance_loss_mlp": 1.04032588,
      "epoch": 0.0465339986773282,
      "flos": 13188661447680.0,
      "grad_norm": 3.342638381318503,
      "language_loss": 0.91039729,
      "learning_rate": 3.997153433088728e-06,
      "loss": 0.93391788,
      "num_input_tokens_seen": 8248600,
      "step": 387,
      "time_per_iteration": 2.4972915649414062
    },
    {
      "auxiliary_loss_clip": 0.01301794,
      "auxiliary_loss_mlp": 0.01071441,
      "balance_loss_clip": 1.08865023,
      "balance_loss_mlp": 1.05676603,
      "epoch": 0.0466542415679673,
      "flos": 25556547168000.0,
      "grad_norm": 1.9588800846267977,
      "language_loss": 0.81182247,
      "learning_rate": 3.997111735745554e-06,
      "loss": 0.83555472,
      "num_input_tokens_seen": 8271570,
      "step": 388,
      "time_per_iteration": 2.557467222213745
    },
    {
      "auxiliary_loss_clip": 0.01302379,
      "auxiliary_loss_mlp": 0.01052859,
      "balance_loss_clip": 1.08969259,
      "balance_loss_mlp": 1.0386374,
      "epoch": 0.04677448445860639,
      "flos": 22236749493120.0,
      "grad_norm": 1.8425311138282774,
      "language_loss": 0.82694167,
      "learning_rate": 3.997069735442118e-06,
      "loss": 0.85049409,
      "num_input_tokens_seen": 8291265,
      "step": 389,
      "time_per_iteration": 2.540644884109497
    },
    {
      "auxiliary_loss_clip": 0.01297284,
      "auxiliary_loss_mlp": 0.01055303,
      "balance_loss_clip": 1.08510876,
      "balance_loss_mlp": 1.04182053,
      "epoch": 0.04689472734924548,
      "flos": 28147825198080.0,
      "grad_norm": 1.7323307251928022,
      "language_loss": 0.80197328,
      "learning_rate": 3.997027432184792e-06,
      "loss": 0.82549918,
      "num_input_tokens_seen": 8315925,
      "step": 390,
      "time_per_iteration": 2.6028120517730713
    },
    {
      "auxiliary_loss_clip": 0.01299828,
      "auxiliary_loss_mlp": 0.01058999,
      "balance_loss_clip": 1.08555412,
      "balance_loss_mlp": 1.04425263,
      "epoch": 0.04701497023988457,
      "flos": 23148952312320.0,
      "grad_norm": 1.8378619195442716,
      "language_loss": 0.89240122,
      "learning_rate": 3.99698482597999e-06,
      "loss": 0.91598946,
      "num_input_tokens_seen": 8333605,
      "step": 391,
      "time_per_iteration": 2.539053440093994
    },
    {
      "auxiliary_loss_clip": 0.01188977,
      "auxiliary_loss_mlp": 0.01030567,
      "balance_loss_clip": 1.05706573,
      "balance_loss_mlp": 1.02565598,
      "epoch": 0.04713521313052366,
      "flos": 64827668764800.0,
      "grad_norm": 0.8892898289068121,
      "language_loss": 0.63951337,
      "learning_rate": 3.99694191683418e-06,
      "loss": 0.66170883,
      "num_input_tokens_seen": 8394405,
      "step": 392,
      "time_per_iteration": 3.1284377574920654
    },
    {
      "auxiliary_loss_clip": 0.01303388,
      "auxiliary_loss_mlp": 0.01064254,
      "balance_loss_clip": 1.08962464,
      "balance_loss_mlp": 1.04895353,
      "epoch": 0.047255456021162746,
      "flos": 18771585477120.0,
      "grad_norm": 1.8553547007987088,
      "language_loss": 0.8194291,
      "learning_rate": 3.996898704753867e-06,
      "loss": 0.84310549,
      "num_input_tokens_seen": 8412355,
      "step": 393,
      "time_per_iteration": 2.5147111415863037
    },
    {
      "auxiliary_loss_clip": 0.01300102,
      "auxiliary_loss_mlp": 0.01053776,
      "balance_loss_clip": 1.0856781,
      "balance_loss_mlp": 1.03821898,
      "epoch": 0.04737569891180184,
      "flos": 22053820504320.0,
      "grad_norm": 2.276158456191609,
      "language_loss": 0.87559038,
      "learning_rate": 3.996855189745609e-06,
      "loss": 0.89912921,
      "num_input_tokens_seen": 8431620,
      "step": 394,
      "time_per_iteration": 2.519475221633911
    },
    {
      "auxiliary_loss_clip": 0.01297969,
      "auxiliary_loss_mlp": 0.01056434,
      "balance_loss_clip": 1.08481884,
      "balance_loss_mlp": 1.04203379,
      "epoch": 0.04749594180244093,
      "flos": 29057370410880.0,
      "grad_norm": 2.177305733168899,
      "language_loss": 0.924519,
      "learning_rate": 3.996811371816007e-06,
      "loss": 0.94806302,
      "num_input_tokens_seen": 8454045,
      "step": 395,
      "time_per_iteration": 2.579786539077759
    },
    {
      "auxiliary_loss_clip": 0.0130009,
      "auxiliary_loss_mlp": 0.01055205,
      "balance_loss_clip": 1.08881855,
      "balance_loss_mlp": 1.04074466,
      "epoch": 0.04761618469308002,
      "flos": 35112268172160.0,
      "grad_norm": 1.8585215509662585,
      "language_loss": 0.7783581,
      "learning_rate": 3.996767250971707e-06,
      "loss": 0.80191106,
      "num_input_tokens_seen": 8476785,
      "step": 396,
      "time_per_iteration": 2.6452834606170654
    },
    {
      "auxiliary_loss_clip": 0.01302419,
      "auxiliary_loss_mlp": 0.01061188,
      "balance_loss_clip": 1.089764,
      "balance_loss_mlp": 1.04797983,
      "epoch": 0.04773642758371911,
      "flos": 25630702796160.0,
      "grad_norm": 5.387316918295607,
      "language_loss": 0.86926305,
      "learning_rate": 3.996722827219403e-06,
      "loss": 0.89289916,
      "num_input_tokens_seen": 8498400,
      "step": 397,
      "time_per_iteration": 2.6004762649536133
    },
    {
      "auxiliary_loss_clip": 0.01301981,
      "auxiliary_loss_mlp": 0.0105534,
      "balance_loss_clip": 1.08983195,
      "balance_loss_mlp": 1.04107022,
      "epoch": 0.0478566704743582,
      "flos": 20631506688000.0,
      "grad_norm": 2.371611049687816,
      "language_loss": 0.82242262,
      "learning_rate": 3.996678100565833e-06,
      "loss": 0.84599584,
      "num_input_tokens_seen": 8517455,
      "step": 398,
      "time_per_iteration": 2.5355756282806396
    },
    {
      "auxiliary_loss_clip": 0.01294029,
      "auxiliary_loss_mlp": 0.01049261,
      "balance_loss_clip": 1.08367407,
      "balance_loss_mlp": 1.03505123,
      "epoch": 0.04797691336499729,
      "flos": 18835721210880.0,
      "grad_norm": 2.4935237954883798,
      "language_loss": 0.88365811,
      "learning_rate": 3.996633071017783e-06,
      "loss": 0.90709102,
      "num_input_tokens_seen": 8534085,
      "step": 399,
      "time_per_iteration": 2.5252017974853516
    },
    {
      "auxiliary_loss_clip": 0.01298547,
      "auxiliary_loss_mlp": 0.01052037,
      "balance_loss_clip": 1.08563828,
      "balance_loss_mlp": 1.03833961,
      "epoch": 0.04809715625563638,
      "flos": 21099673578240.0,
      "grad_norm": 2.7784191773277582,
      "language_loss": 0.81459171,
      "learning_rate": 3.996587738582084e-06,
      "loss": 0.83809757,
      "num_input_tokens_seen": 8550885,
      "step": 400,
      "time_per_iteration": 2.532257080078125
    },
    {
      "auxiliary_loss_clip": 0.01296374,
      "auxiliary_loss_mlp": 0.01057945,
      "balance_loss_clip": 1.08454335,
      "balance_loss_mlp": 1.0437119,
      "epoch": 0.04821739914627548,
      "flos": 23805650712960.0,
      "grad_norm": 2.767417968868868,
      "language_loss": 0.85933489,
      "learning_rate": 3.9965421032656115e-06,
      "loss": 0.88287807,
      "num_input_tokens_seen": 8570815,
      "step": 401,
      "time_per_iteration": 2.52763295173645
    },
    {
      "auxiliary_loss_clip": 0.0129743,
      "auxiliary_loss_mlp": 0.01056337,
      "balance_loss_clip": 1.08491325,
      "balance_loss_mlp": 1.04209149,
      "epoch": 0.04833764203691457,
      "flos": 22200587475840.0,
      "grad_norm": 2.7395367186996404,
      "language_loss": 0.94259155,
      "learning_rate": 3.99649616507529e-06,
      "loss": 0.9661293,
      "num_input_tokens_seen": 8589910,
      "step": 402,
      "time_per_iteration": 2.5382044315338135
    },
    {
      "auxiliary_loss_clip": 0.01183634,
      "auxiliary_loss_mlp": 0.01009718,
      "balance_loss_clip": 1.05383909,
      "balance_loss_mlp": 1.00516403,
      "epoch": 0.04845788492755366,
      "flos": 65904376896000.0,
      "grad_norm": 0.8892393253277029,
      "language_loss": 0.63120091,
      "learning_rate": 3.996449924018088e-06,
      "loss": 0.65313447,
      "num_input_tokens_seen": 8650370,
      "step": 403,
      "time_per_iteration": 3.062020778656006
    },
    {
      "auxiliary_loss_clip": 0.01294488,
      "auxiliary_loss_mlp": 0.01051483,
      "balance_loss_clip": 1.08415258,
      "balance_loss_mlp": 1.03753591,
      "epoch": 0.04857812781819275,
      "flos": 19281301424640.0,
      "grad_norm": 2.384306696339672,
      "language_loss": 0.79400563,
      "learning_rate": 3.99640338010102e-06,
      "loss": 0.81746531,
      "num_input_tokens_seen": 8669475,
      "step": 404,
      "time_per_iteration": 2.536983013153076
    },
    {
      "auxiliary_loss_clip": 0.01291825,
      "auxiliary_loss_mlp": 0.01054814,
      "balance_loss_clip": 1.08265543,
      "balance_loss_mlp": 1.04100919,
      "epoch": 0.04869837070883184,
      "flos": 24062376193920.0,
      "grad_norm": 3.327937200990285,
      "language_loss": 0.78373933,
      "learning_rate": 3.996356533331146e-06,
      "loss": 0.80720568,
      "num_input_tokens_seen": 8691345,
      "step": 405,
      "time_per_iteration": 2.551147699356079
    },
    {
      "auxiliary_loss_clip": 0.01301255,
      "auxiliary_loss_mlp": 0.01053829,
      "balance_loss_clip": 1.08429801,
      "balance_loss_mlp": 1.03896403,
      "epoch": 0.04881861359947093,
      "flos": 25187169657600.0,
      "grad_norm": 2.9563094822042215,
      "language_loss": 0.61869043,
      "learning_rate": 3.996309383715573e-06,
      "loss": 0.64224136,
      "num_input_tokens_seen": 8710125,
      "step": 406,
      "time_per_iteration": 2.528376340866089
    },
    {
      "auxiliary_loss_clip": 0.01295865,
      "auxiliary_loss_mlp": 0.0104439,
      "balance_loss_clip": 1.08421755,
      "balance_loss_mlp": 1.02926278,
      "epoch": 0.048938856490110025,
      "flos": 16362913213440.0,
      "grad_norm": 2.9877722518969616,
      "language_loss": 0.73846602,
      "learning_rate": 3.996261931261454e-06,
      "loss": 0.7618686,
      "num_input_tokens_seen": 8728705,
      "step": 407,
      "time_per_iteration": 2.497581720352173
    },
    {
      "auxiliary_loss_clip": 0.01298022,
      "auxiliary_loss_mlp": 0.01052092,
      "balance_loss_clip": 1.08748698,
      "balance_loss_mlp": 1.03776312,
      "epoch": 0.049059099380749115,
      "flos": 29895094379520.0,
      "grad_norm": 1.5904793879279402,
      "language_loss": 0.86182833,
      "learning_rate": 3.996214175975987e-06,
      "loss": 0.88532948,
      "num_input_tokens_seen": 8749225,
      "step": 408,
      "time_per_iteration": 2.5930793285369873
    },
    {
      "auxiliary_loss_clip": 0.01298568,
      "auxiliary_loss_mlp": 0.01058239,
      "balance_loss_clip": 1.08786249,
      "balance_loss_mlp": 1.04433918,
      "epoch": 0.049179342271388204,
      "flos": 35918858027520.0,
      "grad_norm": 2.1007401857695345,
      "language_loss": 0.79300892,
      "learning_rate": 3.996166117866417e-06,
      "loss": 0.81657702,
      "num_input_tokens_seen": 8771160,
      "step": 409,
      "time_per_iteration": 3.4002935886383057
    },
    {
      "auxiliary_loss_clip": 0.01294389,
      "auxiliary_loss_mlp": 0.01043133,
      "balance_loss_clip": 1.08303809,
      "balance_loss_mlp": 1.02953076,
      "epoch": 0.049299585162027294,
      "flos": 14611226659200.0,
      "grad_norm": 1.9690163795562645,
      "language_loss": 0.86467993,
      "learning_rate": 3.996117756940035e-06,
      "loss": 0.88805509,
      "num_input_tokens_seen": 8787845,
      "step": 410,
      "time_per_iteration": 3.2073020935058594
    },
    {
      "auxiliary_loss_clip": 0.01298872,
      "auxiliary_loss_mlp": 0.01050953,
      "balance_loss_clip": 1.08781314,
      "balance_loss_mlp": 1.03717244,
      "epoch": 0.049419828052666384,
      "flos": 19567939956480.0,
      "grad_norm": 1.9783378658777173,
      "language_loss": 0.97630811,
      "learning_rate": 3.996069093204175e-06,
      "loss": 0.9998064,
      "num_input_tokens_seen": 8803805,
      "step": 411,
      "time_per_iteration": 2.5075764656066895
    },
    {
      "auxiliary_loss_clip": 0.01300983,
      "auxiliary_loss_mlp": 0.01044483,
      "balance_loss_clip": 1.08808351,
      "balance_loss_mlp": 1.03039205,
      "epoch": 0.049540070943305474,
      "flos": 13659916907520.0,
      "grad_norm": 2.592496252616047,
      "language_loss": 0.88169348,
      "learning_rate": 3.996020126666221e-06,
      "loss": 0.90514815,
      "num_input_tokens_seen": 8820785,
      "step": 412,
      "time_per_iteration": 3.264265775680542
    },
    {
      "auxiliary_loss_clip": 0.01296108,
      "auxiliary_loss_mlp": 0.01051977,
      "balance_loss_clip": 1.0858047,
      "balance_loss_mlp": 1.0378623,
      "epoch": 0.04966031383394457,
      "flos": 21832035978240.0,
      "grad_norm": 4.377381791143581,
      "language_loss": 0.82002091,
      "learning_rate": 3.995970857333601e-06,
      "loss": 0.84350175,
      "num_input_tokens_seen": 8841195,
      "step": 413,
      "time_per_iteration": 3.3601770401000977
    },
    {
      "auxiliary_loss_clip": 0.01296794,
      "auxiliary_loss_mlp": 0.0105326,
      "balance_loss_clip": 1.08526587,
      "balance_loss_mlp": 1.039819,
      "epoch": 0.04978055672458366,
      "flos": 28618793349120.0,
      "grad_norm": 2.2475093934647594,
      "language_loss": 0.79724431,
      "learning_rate": 3.995921285213789e-06,
      "loss": 0.82074487,
      "num_input_tokens_seen": 8861455,
      "step": 414,
      "time_per_iteration": 2.57621693611145
    },
    {
      "auxiliary_loss_clip": 0.01295517,
      "auxiliary_loss_mlp": 0.01049913,
      "balance_loss_clip": 1.08663404,
      "balance_loss_mlp": 1.03616202,
      "epoch": 0.04990079961522275,
      "flos": 19828220883840.0,
      "grad_norm": 2.1102372743679094,
      "language_loss": 0.80552173,
      "learning_rate": 3.995871410314305e-06,
      "loss": 0.82897604,
      "num_input_tokens_seen": 8880015,
      "step": 415,
      "time_per_iteration": 2.4838385581970215
    },
    {
      "auxiliary_loss_clip": 0.01167306,
      "auxiliary_loss_mlp": 0.01020829,
      "balance_loss_clip": 1.05264688,
      "balance_loss_mlp": 1.01629925,
      "epoch": 0.05002104250586184,
      "flos": 62735045293440.0,
      "grad_norm": 0.9071841059237318,
      "language_loss": 0.59636384,
      "learning_rate": 3.995821232642714e-06,
      "loss": 0.61824512,
      "num_input_tokens_seen": 8938420,
      "step": 416,
      "time_per_iteration": 3.1995089054107666
    },
    {
      "auxiliary_loss_clip": 0.01276868,
      "auxiliary_loss_mlp": 0.01065635,
      "balance_loss_clip": 1.08548546,
      "balance_loss_mlp": 1.05259371,
      "epoch": 0.05014128539650093,
      "flos": 27928518710400.0,
      "grad_norm": 2.1814105868005726,
      "language_loss": 0.82372421,
      "learning_rate": 3.995770752206629e-06,
      "loss": 0.84714925,
      "num_input_tokens_seen": 8959495,
      "step": 417,
      "time_per_iteration": 2.6392664909362793
    },
    {
      "auxiliary_loss_clip": 0.01294189,
      "auxiliary_loss_mlp": 0.01047843,
      "balance_loss_clip": 1.08543086,
      "balance_loss_mlp": 1.03430068,
      "epoch": 0.05026152828714002,
      "flos": 17705576620800.0,
      "grad_norm": 1.99972775542304,
      "language_loss": 0.97208148,
      "learning_rate": 3.995719969013709e-06,
      "loss": 0.99550188,
      "num_input_tokens_seen": 8976675,
      "step": 418,
      "time_per_iteration": 2.487362861633301
    },
    {
      "auxiliary_loss_clip": 0.01258053,
      "auxiliary_loss_mlp": 0.01050189,
      "balance_loss_clip": 1.08111763,
      "balance_loss_mlp": 1.03622365,
      "epoch": 0.05038177117777912,
      "flos": 19133277477120.0,
      "grad_norm": 2.468051138273594,
      "language_loss": 0.86171067,
      "learning_rate": 3.995668883071655e-06,
      "loss": 0.88479304,
      "num_input_tokens_seen": 8992900,
      "step": 419,
      "time_per_iteration": 2.5523312091827393
    },
    {
      "auxiliary_loss_clip": 0.0129759,
      "auxiliary_loss_mlp": 0.01049753,
      "balance_loss_clip": 1.08567035,
      "balance_loss_mlp": 1.0362525,
      "epoch": 0.050502014068418206,
      "flos": 20667704618880.0,
      "grad_norm": 2.1454004437651313,
      "language_loss": 0.90749693,
      "learning_rate": 3.995617494388219e-06,
      "loss": 0.93097043,
      "num_input_tokens_seen": 9011020,
      "step": 420,
      "time_per_iteration": 2.5103988647460938
    },
    {
      "auxiliary_loss_clip": 0.01255647,
      "auxiliary_loss_mlp": 0.01047747,
      "balance_loss_clip": 1.07592797,
      "balance_loss_mlp": 1.03388321,
      "epoch": 0.050622256959057296,
      "flos": 21361103740800.0,
      "grad_norm": 2.0217634439995242,
      "language_loss": 0.80180848,
      "learning_rate": 3.995565802971196e-06,
      "loss": 0.82484245,
      "num_input_tokens_seen": 9030995,
      "step": 421,
      "time_per_iteration": 2.554809808731079
    },
    {
      "auxiliary_loss_clip": 0.01253219,
      "auxiliary_loss_mlp": 0.01053273,
      "balance_loss_clip": 1.07750309,
      "balance_loss_mlp": 1.04055333,
      "epoch": 0.050742499849696386,
      "flos": 27673588909440.0,
      "grad_norm": 1.8357476880734556,
      "language_loss": 0.67166471,
      "learning_rate": 3.995513808828427e-06,
      "loss": 0.69472969,
      "num_input_tokens_seen": 9053790,
      "step": 422,
      "time_per_iteration": 2.6102421283721924
    },
    {
      "auxiliary_loss_clip": 0.01255398,
      "auxiliary_loss_mlp": 0.01054033,
      "balance_loss_clip": 1.07738996,
      "balance_loss_mlp": 1.04152226,
      "epoch": 0.050862742740335476,
      "flos": 19865999013120.0,
      "grad_norm": 2.1071532624206544,
      "language_loss": 0.76526892,
      "learning_rate": 3.9954615119678e-06,
      "loss": 0.78836322,
      "num_input_tokens_seen": 9072345,
      "step": 423,
      "time_per_iteration": 2.6151645183563232
    },
    {
      "auxiliary_loss_clip": 0.0126732,
      "auxiliary_loss_mlp": 0.01051632,
      "balance_loss_clip": 1.07964826,
      "balance_loss_mlp": 1.03835785,
      "epoch": 0.050982985630974566,
      "flos": 22085098272000.0,
      "grad_norm": 2.359323369248374,
      "language_loss": 0.80617499,
      "learning_rate": 3.995408912397248e-06,
      "loss": 0.82936454,
      "num_input_tokens_seen": 9090240,
      "step": 424,
      "time_per_iteration": 2.617138624191284
    },
    {
      "auxiliary_loss_clip": 0.01263592,
      "auxiliary_loss_mlp": 0.01051227,
      "balance_loss_clip": 1.08294737,
      "balance_loss_mlp": 1.03799462,
      "epoch": 0.05110322852161366,
      "flos": 20740962407040.0,
      "grad_norm": 2.6523715599048336,
      "language_loss": 0.93048763,
      "learning_rate": 3.99535601012475e-06,
      "loss": 0.95363581,
      "num_input_tokens_seen": 9105570,
      "step": 425,
      "time_per_iteration": 2.555788278579712
    },
    {
      "auxiliary_loss_clip": 0.01239661,
      "auxiliary_loss_mlp": 0.00765854,
      "balance_loss_clip": 1.0765276,
      "balance_loss_mlp": 1.00021994,
      "epoch": 0.05122347141225275,
      "flos": 28547295327360.0,
      "grad_norm": 1.573369411613478,
      "language_loss": 0.75455809,
      "learning_rate": 3.995302805158333e-06,
      "loss": 0.77461326,
      "num_input_tokens_seen": 9128225,
      "step": 426,
      "time_per_iteration": 2.6614022254943848
    },
    {
      "auxiliary_loss_clip": 0.01245788,
      "auxiliary_loss_mlp": 0.01049733,
      "balance_loss_clip": 1.07432151,
      "balance_loss_mlp": 1.03483212,
      "epoch": 0.05134371430289184,
      "flos": 19722679747200.0,
      "grad_norm": 1.9382619732740944,
      "language_loss": 0.83390868,
      "learning_rate": 3.9952492975060665e-06,
      "loss": 0.85686386,
      "num_input_tokens_seen": 9148295,
      "step": 427,
      "time_per_iteration": 2.577453851699829
    },
    {
      "auxiliary_loss_clip": 0.01269587,
      "auxiliary_loss_mlp": 0.01052422,
      "balance_loss_clip": 1.0774678,
      "balance_loss_mlp": 1.03863537,
      "epoch": 0.05146395719353093,
      "flos": 34458945649920.0,
      "grad_norm": 3.360417968498193,
      "language_loss": 0.85334933,
      "learning_rate": 3.995195487176067e-06,
      "loss": 0.87656939,
      "num_input_tokens_seen": 9168525,
      "step": 428,
      "time_per_iteration": 2.6565427780151367
    },
    {
      "auxiliary_loss_clip": 0.01289953,
      "auxiliary_loss_mlp": 0.01051526,
      "balance_loss_clip": 1.08221388,
      "balance_loss_mlp": 1.0379777,
      "epoch": 0.05158420008417002,
      "flos": 21760286561280.0,
      "grad_norm": 2.267059024345018,
      "language_loss": 0.8572619,
      "learning_rate": 3.995141374176499e-06,
      "loss": 0.88067663,
      "num_input_tokens_seen": 9186920,
      "step": 429,
      "time_per_iteration": 2.5055947303771973
    },
    {
      "auxiliary_loss_clip": 0.01134975,
      "auxiliary_loss_mlp": 0.00756409,
      "balance_loss_clip": 1.0451827,
      "balance_loss_mlp": 1.00035572,
      "epoch": 0.05170444297480911,
      "flos": 72553956226560.0,
      "grad_norm": 0.87072985540237,
      "language_loss": 0.6306479,
      "learning_rate": 3.995086958515572e-06,
      "loss": 0.64956164,
      "num_input_tokens_seen": 9244940,
      "step": 430,
      "time_per_iteration": 3.192450523376465
    },
    {
      "auxiliary_loss_clip": 0.01178492,
      "auxiliary_loss_mlp": 0.00756453,
      "balance_loss_clip": 1.05127549,
      "balance_loss_mlp": 1.00037003,
      "epoch": 0.05182468586544821,
      "flos": 62416159326720.0,
      "grad_norm": 0.861409078045119,
      "language_loss": 0.59941179,
      "learning_rate": 3.995032240201538e-06,
      "loss": 0.61876124,
      "num_input_tokens_seen": 9307335,
      "step": 431,
      "time_per_iteration": 3.071185827255249
    },
    {
      "auxiliary_loss_clip": 0.01156768,
      "auxiliary_loss_mlp": 0.01005101,
      "balance_loss_clip": 1.04923344,
      "balance_loss_mlp": 1.00061917,
      "epoch": 0.0519449287560873,
      "flos": 41225989432320.0,
      "grad_norm": 0.9376186999624959,
      "language_loss": 0.63126254,
      "learning_rate": 3.9949772192427e-06,
      "loss": 0.65288126,
      "num_input_tokens_seen": 9353960,
      "step": 432,
      "time_per_iteration": 2.881181478500366
    },
    {
      "auxiliary_loss_clip": 0.01254989,
      "auxiliary_loss_mlp": 0.01056224,
      "balance_loss_clip": 1.07559323,
      "balance_loss_mlp": 1.04291999,
      "epoch": 0.05206517164672639,
      "flos": 17494530261120.0,
      "grad_norm": 2.3182655093361073,
      "language_loss": 0.79703385,
      "learning_rate": 3.994921895647405e-06,
      "loss": 0.82014596,
      "num_input_tokens_seen": 9372130,
      "step": 433,
      "time_per_iteration": 2.6137983798980713
    },
    {
      "auxiliary_loss_clip": 0.01174519,
      "auxiliary_loss_mlp": 0.0100441,
      "balance_loss_clip": 1.04849982,
      "balance_loss_mlp": 0.9999277,
      "epoch": 0.05218541453736548,
      "flos": 64002762973440.0,
      "grad_norm": 0.8386474568642345,
      "language_loss": 0.55351651,
      "learning_rate": 3.994866269424043e-06,
      "loss": 0.57530582,
      "num_input_tokens_seen": 9428500,
      "step": 434,
      "time_per_iteration": 3.0588133335113525
    },
    {
      "auxiliary_loss_clip": 0.01200905,
      "auxiliary_loss_mlp": 0.01046583,
      "balance_loss_clip": 1.062482,
      "balance_loss_mlp": 1.03335643,
      "epoch": 0.05230565742800457,
      "flos": 19317319787520.0,
      "grad_norm": 2.522417044646948,
      "language_loss": 0.78459275,
      "learning_rate": 3.9948103405810545e-06,
      "loss": 0.80706763,
      "num_input_tokens_seen": 9447450,
      "step": 435,
      "time_per_iteration": 3.4613616466522217
    },
    {
      "auxiliary_loss_clip": 0.01222064,
      "auxiliary_loss_mlp": 0.01047497,
      "balance_loss_clip": 1.07028818,
      "balance_loss_mlp": 1.03465199,
      "epoch": 0.05242590031864366,
      "flos": 25298636538240.0,
      "grad_norm": 2.085014167704342,
      "language_loss": 0.85912788,
      "learning_rate": 3.994754109126923e-06,
      "loss": 0.88182354,
      "num_input_tokens_seen": 9468945,
      "step": 436,
      "time_per_iteration": 3.416391372680664
    },
    {
      "auxiliary_loss_clip": 0.01194591,
      "auxiliary_loss_mlp": 0.01044413,
      "balance_loss_clip": 1.06957769,
      "balance_loss_mlp": 1.03187799,
      "epoch": 0.052546143209282754,
      "flos": 26211629456640.0,
      "grad_norm": 1.7412550006792054,
      "language_loss": 0.93521237,
      "learning_rate": 3.994697575070181e-06,
      "loss": 0.95760238,
      "num_input_tokens_seen": 9488405,
      "step": 437,
      "time_per_iteration": 2.6974403858184814
    },
    {
      "auxiliary_loss_clip": 0.01252336,
      "auxiliary_loss_mlp": 0.01053546,
      "balance_loss_clip": 1.07903302,
      "balance_loss_mlp": 1.04083276,
      "epoch": 0.052666386099921844,
      "flos": 22158140578560.0,
      "grad_norm": 1.7437034635853288,
      "language_loss": 0.91521209,
      "learning_rate": 3.994640738419402e-06,
      "loss": 0.93827087,
      "num_input_tokens_seen": 9507780,
      "step": 438,
      "time_per_iteration": 3.303598165512085
    },
    {
      "auxiliary_loss_clip": 0.01271946,
      "auxiliary_loss_mlp": 0.01050872,
      "balance_loss_clip": 1.07973862,
      "balance_loss_mlp": 1.03834939,
      "epoch": 0.052786628990560934,
      "flos": 23881817502720.0,
      "grad_norm": 1.9401678109865004,
      "language_loss": 0.80728865,
      "learning_rate": 3.9945835991832075e-06,
      "loss": 0.83051687,
      "num_input_tokens_seen": 9529665,
      "step": 439,
      "time_per_iteration": 3.4677796363830566
    },
    {
      "auxiliary_loss_clip": 0.01289536,
      "auxiliary_loss_mlp": 0.01061502,
      "balance_loss_clip": 1.08461857,
      "balance_loss_mlp": 1.04893744,
      "epoch": 0.052906871881200024,
      "flos": 24605021934720.0,
      "grad_norm": 2.128685379124622,
      "language_loss": 0.93042034,
      "learning_rate": 3.994526157370268e-06,
      "loss": 0.95393074,
      "num_input_tokens_seen": 9548280,
      "step": 440,
      "time_per_iteration": 2.601649284362793
    },
    {
      "auxiliary_loss_clip": 0.01144662,
      "auxiliary_loss_mlp": 0.01006134,
      "balance_loss_clip": 1.03857684,
      "balance_loss_mlp": 1.00172281,
      "epoch": 0.053027114771839114,
      "flos": 56461631143680.0,
      "grad_norm": 0.8952910628839714,
      "language_loss": 0.59311569,
      "learning_rate": 3.994468412989296e-06,
      "loss": 0.61462361,
      "num_input_tokens_seen": 9609690,
      "step": 441,
      "time_per_iteration": 3.3135902881622314
    },
    {
      "auxiliary_loss_clip": 0.01225044,
      "auxiliary_loss_mlp": 0.0106072,
      "balance_loss_clip": 1.07109904,
      "balance_loss_mlp": 1.04780972,
      "epoch": 0.053147357662478203,
      "flos": 17311098481920.0,
      "grad_norm": 1.8989996564038183,
      "language_loss": 0.92544341,
      "learning_rate": 3.994410366049052e-06,
      "loss": 0.94830108,
      "num_input_tokens_seen": 9627550,
      "step": 442,
      "time_per_iteration": 2.706190347671509
    },
    {
      "auxiliary_loss_clip": 0.01271602,
      "auxiliary_loss_mlp": 0.01050672,
      "balance_loss_clip": 1.08151484,
      "balance_loss_mlp": 1.03729713,
      "epoch": 0.0532676005531173,
      "flos": 17164977955200.0,
      "grad_norm": 3.4508367864686607,
      "language_loss": 0.83190441,
      "learning_rate": 3.994352016558341e-06,
      "loss": 0.85512722,
      "num_input_tokens_seen": 9644855,
      "step": 443,
      "time_per_iteration": 2.5963544845581055
    },
    {
      "auxiliary_loss_clip": 0.01267008,
      "auxiliary_loss_mlp": 0.01043487,
      "balance_loss_clip": 1.08181465,
      "balance_loss_mlp": 1.03077924,
      "epoch": 0.05338784344375639,
      "flos": 27819960831360.0,
      "grad_norm": 1.8030842004347758,
      "language_loss": 0.73726535,
      "learning_rate": 3.994293364526014e-06,
      "loss": 0.76037025,
      "num_input_tokens_seen": 9665740,
      "step": 444,
      "time_per_iteration": 2.635636568069458
    },
    {
      "auxiliary_loss_clip": 0.01244395,
      "auxiliary_loss_mlp": 0.01059504,
      "balance_loss_clip": 1.07720017,
      "balance_loss_mlp": 1.04525292,
      "epoch": 0.05350808633439548,
      "flos": 21507691144320.0,
      "grad_norm": 1.8900497194750072,
      "language_loss": 0.849015,
      "learning_rate": 3.99423440996097e-06,
      "loss": 0.87205398,
      "num_input_tokens_seen": 9685280,
      "step": 445,
      "time_per_iteration": 2.613539457321167
    },
    {
      "auxiliary_loss_clip": 0.01260712,
      "auxiliary_loss_mlp": 0.01049994,
      "balance_loss_clip": 1.08255184,
      "balance_loss_mlp": 1.03690481,
      "epoch": 0.05362832922503457,
      "flos": 20084299920000.0,
      "grad_norm": 2.8864332708279488,
      "language_loss": 0.81618953,
      "learning_rate": 3.994175152872152e-06,
      "loss": 0.83929658,
      "num_input_tokens_seen": 9704365,
      "step": 446,
      "time_per_iteration": 2.5638628005981445
    },
    {
      "auxiliary_loss_clip": 0.01272591,
      "auxiliary_loss_mlp": 0.01049621,
      "balance_loss_clip": 1.08070183,
      "balance_loss_mlp": 1.03755105,
      "epoch": 0.05374857211567366,
      "flos": 26137222433280.0,
      "grad_norm": 2.571150385035351,
      "language_loss": 0.78572059,
      "learning_rate": 3.994115593268548e-06,
      "loss": 0.80894279,
      "num_input_tokens_seen": 9724145,
      "step": 447,
      "time_per_iteration": 2.602689743041992
    },
    {
      "auxiliary_loss_clip": 0.01287673,
      "auxiliary_loss_mlp": 0.01052261,
      "balance_loss_clip": 1.08456528,
      "balance_loss_mlp": 1.03927898,
      "epoch": 0.05386881500631275,
      "flos": 27486817165440.0,
      "grad_norm": 3.1059486198903636,
      "language_loss": 0.8194207,
      "learning_rate": 3.994055731159195e-06,
      "loss": 0.84281999,
      "num_input_tokens_seen": 9741615,
      "step": 448,
      "time_per_iteration": 2.5432629585266113
    },
    {
      "auxiliary_loss_clip": 0.01277357,
      "auxiliary_loss_mlp": 0.01056202,
      "balance_loss_clip": 1.08640933,
      "balance_loss_mlp": 1.04434037,
      "epoch": 0.053989057896951846,
      "flos": 23585087249280.0,
      "grad_norm": 1.8270769778445632,
      "language_loss": 0.86827898,
      "learning_rate": 3.993995566553172e-06,
      "loss": 0.89161456,
      "num_input_tokens_seen": 9760580,
      "step": 449,
      "time_per_iteration": 2.571065902709961
    },
    {
      "auxiliary_loss_clip": 0.01231399,
      "auxiliary_loss_mlp": 0.01054468,
      "balance_loss_clip": 1.07058597,
      "balance_loss_mlp": 1.04112864,
      "epoch": 0.054109300787590936,
      "flos": 25228862369280.0,
      "grad_norm": 1.7350831621146896,
      "language_loss": 0.77089834,
      "learning_rate": 3.993935099459607e-06,
      "loss": 0.79375702,
      "num_input_tokens_seen": 9782195,
      "step": 450,
      "time_per_iteration": 2.6469762325286865
    },
    {
      "auxiliary_loss_clip": 0.01281178,
      "auxiliary_loss_mlp": 0.01049769,
      "balance_loss_clip": 1.08079302,
      "balance_loss_mlp": 1.038903,
      "epoch": 0.054229543678230026,
      "flos": 23841525421440.0,
      "grad_norm": 1.9774020252577473,
      "language_loss": 0.73769426,
      "learning_rate": 3.993874329887673e-06,
      "loss": 0.76100379,
      "num_input_tokens_seen": 9800850,
      "step": 451,
      "time_per_iteration": 2.528331756591797
    },
    {
      "auxiliary_loss_clip": 0.01268189,
      "auxiliary_loss_mlp": 0.0104603,
      "balance_loss_clip": 1.08129859,
      "balance_loss_mlp": 1.03326893,
      "epoch": 0.054349786568869116,
      "flos": 16320933192960.0,
      "grad_norm": 2.848805563495819,
      "language_loss": 0.86948776,
      "learning_rate": 3.993813257846589e-06,
      "loss": 0.89262998,
      "num_input_tokens_seen": 9817605,
      "step": 452,
      "time_per_iteration": 2.5223114490509033
    },
    {
      "auxiliary_loss_clip": 0.01271749,
      "auxiliary_loss_mlp": 0.01059343,
      "balance_loss_clip": 1.08173156,
      "balance_loss_mlp": 1.04540706,
      "epoch": 0.054470029459508205,
      "flos": 18660729127680.0,
      "grad_norm": 2.4571028221530606,
      "language_loss": 0.92884243,
      "learning_rate": 3.993751883345619e-06,
      "loss": 0.95215338,
      "num_input_tokens_seen": 9835965,
      "step": 453,
      "time_per_iteration": 2.569429874420166
    },
    {
      "auxiliary_loss_clip": 0.01253113,
      "auxiliary_loss_mlp": 0.0105317,
      "balance_loss_clip": 1.07999086,
      "balance_loss_mlp": 1.03958035,
      "epoch": 0.054590272350147295,
      "flos": 17785298856960.0,
      "grad_norm": 2.9615395213201423,
      "language_loss": 0.87885469,
      "learning_rate": 3.993690206394073e-06,
      "loss": 0.90191758,
      "num_input_tokens_seen": 9852265,
      "step": 454,
      "time_per_iteration": 2.6793081760406494
    },
    {
      "auxiliary_loss_clip": 0.01259241,
      "auxiliary_loss_mlp": 0.01053685,
      "balance_loss_clip": 1.07823539,
      "balance_loss_mlp": 1.04088151,
      "epoch": 0.054710515240786385,
      "flos": 17785945301760.0,
      "grad_norm": 2.3829721169322804,
      "language_loss": 0.87610936,
      "learning_rate": 3.993628227001307e-06,
      "loss": 0.89923859,
      "num_input_tokens_seen": 9870465,
      "step": 455,
      "time_per_iteration": 2.600213050842285
    },
    {
      "auxiliary_loss_clip": 0.01253476,
      "auxiliary_loss_mlp": 0.01060696,
      "balance_loss_clip": 1.07685494,
      "balance_loss_mlp": 1.04775548,
      "epoch": 0.05483075813142548,
      "flos": 48210900180480.0,
      "grad_norm": 1.86371478785799,
      "language_loss": 0.71201313,
      "learning_rate": 3.993565945176726e-06,
      "loss": 0.73515487,
      "num_input_tokens_seen": 9891490,
      "step": 456,
      "time_per_iteration": 2.8394179344177246
    },
    {
      "auxiliary_loss_clip": 0.01252527,
      "auxiliary_loss_mlp": 0.01053686,
      "balance_loss_clip": 1.07886767,
      "balance_loss_mlp": 1.04117525,
      "epoch": 0.05495100102206457,
      "flos": 19682244011520.0,
      "grad_norm": 1.8961694396020916,
      "language_loss": 0.84159821,
      "learning_rate": 3.993503360929776e-06,
      "loss": 0.86466032,
      "num_input_tokens_seen": 9910375,
      "step": 457,
      "time_per_iteration": 2.623880386352539
    },
    {
      "auxiliary_loss_clip": 0.01186274,
      "auxiliary_loss_mlp": 0.01055995,
      "balance_loss_clip": 1.06665397,
      "balance_loss_mlp": 1.04293013,
      "epoch": 0.05507124391270366,
      "flos": 26360048453760.0,
      "grad_norm": 1.6163516234840345,
      "language_loss": 0.80881095,
      "learning_rate": 3.99344047426995e-06,
      "loss": 0.83123362,
      "num_input_tokens_seen": 9931635,
      "step": 458,
      "time_per_iteration": 2.973371744155884
    },
    {
      "auxiliary_loss_clip": 0.01223668,
      "auxiliary_loss_mlp": 0.01055138,
      "balance_loss_clip": 1.07347775,
      "balance_loss_mlp": 1.04133368,
      "epoch": 0.05519148680334275,
      "flos": 22601314581120.0,
      "grad_norm": 2.422910552736015,
      "language_loss": 0.93424994,
      "learning_rate": 3.993377285206789e-06,
      "loss": 0.95703799,
      "num_input_tokens_seen": 9951420,
      "step": 459,
      "time_per_iteration": 2.9041428565979004
    },
    {
      "auxiliary_loss_clip": 0.01208098,
      "auxiliary_loss_mlp": 0.01054909,
      "balance_loss_clip": 1.06985867,
      "balance_loss_mlp": 1.04205275,
      "epoch": 0.05531172969398184,
      "flos": 40552519380480.0,
      "grad_norm": 1.6399430347489103,
      "language_loss": 0.86271828,
      "learning_rate": 3.99331379374988e-06,
      "loss": 0.88534838,
      "num_input_tokens_seen": 9975025,
      "step": 460,
      "time_per_iteration": 2.806896448135376
    },
    {
      "auxiliary_loss_clip": 0.01250109,
      "auxiliary_loss_mlp": 0.01047104,
      "balance_loss_clip": 1.07105684,
      "balance_loss_mlp": 1.03405058,
      "epoch": 0.05543197258462093,
      "flos": 23477894087040.0,
      "grad_norm": 1.9169020249850914,
      "language_loss": 0.79927778,
      "learning_rate": 3.993249999908852e-06,
      "loss": 0.82224989,
      "num_input_tokens_seen": 9995175,
      "step": 461,
      "time_per_iteration": 3.909759759902954
    },
    {
      "auxiliary_loss_clip": 0.0128572,
      "auxiliary_loss_mlp": 0.01056352,
      "balance_loss_clip": 1.08106542,
      "balance_loss_mlp": 1.04258919,
      "epoch": 0.05555221547526003,
      "flos": 18624603024000.0,
      "grad_norm": 2.0178269687371406,
      "language_loss": 0.86910272,
      "learning_rate": 3.993185903693384e-06,
      "loss": 0.89252341,
      "num_input_tokens_seen": 10011975,
      "step": 462,
      "time_per_iteration": 3.3313541412353516
    },
    {
      "auxiliary_loss_clip": 0.01256365,
      "auxiliary_loss_mlp": 0.01044911,
      "balance_loss_clip": 1.07848072,
      "balance_loss_mlp": 1.03282356,
      "epoch": 0.05567245836589912,
      "flos": 23587098410880.0,
      "grad_norm": 1.9623062631651915,
      "language_loss": 0.82131141,
      "learning_rate": 3.9931215051131995e-06,
      "loss": 0.84432411,
      "num_input_tokens_seen": 10032620,
      "step": 463,
      "time_per_iteration": 2.6627554893493652
    },
    {
      "auxiliary_loss_clip": 0.01252951,
      "auxiliary_loss_mlp": 0.01055579,
      "balance_loss_clip": 1.07330322,
      "balance_loss_mlp": 1.04343176,
      "epoch": 0.05579270125653821,
      "flos": 27746667129600.0,
      "grad_norm": 1.7975228642570558,
      "language_loss": 0.80013227,
      "learning_rate": 3.993056804178068e-06,
      "loss": 0.82321751,
      "num_input_tokens_seen": 10054165,
      "step": 464,
      "time_per_iteration": 3.447030544281006
    },
    {
      "auxiliary_loss_clip": 0.0121318,
      "auxiliary_loss_mlp": 0.01045566,
      "balance_loss_clip": 1.07312107,
      "balance_loss_mlp": 1.03231001,
      "epoch": 0.0559129441471773,
      "flos": 27014161075200.0,
      "grad_norm": 2.0867259080534852,
      "language_loss": 0.84481251,
      "learning_rate": 3.992991800897803e-06,
      "loss": 0.86739999,
      "num_input_tokens_seen": 10073970,
      "step": 465,
      "time_per_iteration": 3.5320496559143066
    },
    {
      "auxiliary_loss_clip": 0.0128294,
      "auxiliary_loss_mlp": 0.01051739,
      "balance_loss_clip": 1.08213663,
      "balance_loss_mlp": 1.03907943,
      "epoch": 0.05603318703781639,
      "flos": 15229787794560.0,
      "grad_norm": 2.153599102477539,
      "language_loss": 0.90160614,
      "learning_rate": 3.9929264952822665e-06,
      "loss": 0.92495286,
      "num_input_tokens_seen": 10091505,
      "step": 466,
      "time_per_iteration": 2.5776610374450684
    },
    {
      "auxiliary_loss_clip": 0.01272353,
      "auxiliary_loss_mlp": 0.01043554,
      "balance_loss_clip": 1.07913721,
      "balance_loss_mlp": 1.03077495,
      "epoch": 0.05615342992845548,
      "flos": 22266482976000.0,
      "grad_norm": 1.820271584543583,
      "language_loss": 0.8828811,
      "learning_rate": 3.992860887341366e-06,
      "loss": 0.90604019,
      "num_input_tokens_seen": 10109675,
      "step": 467,
      "time_per_iteration": 2.6077022552490234
    },
    {
      "auxiliary_loss_clip": 0.01217005,
      "auxiliary_loss_mlp": 0.01049196,
      "balance_loss_clip": 1.07021856,
      "balance_loss_mlp": 1.03680468,
      "epoch": 0.056273672819094574,
      "flos": 23584979508480.0,
      "grad_norm": 2.306178896194831,
      "language_loss": 0.81092757,
      "learning_rate": 3.992794977085052e-06,
      "loss": 0.83358961,
      "num_input_tokens_seen": 10127675,
      "step": 468,
      "time_per_iteration": 2.6807377338409424
    },
    {
      "auxiliary_loss_clip": 0.01231583,
      "auxiliary_loss_mlp": 0.01053009,
      "balance_loss_clip": 1.07422876,
      "balance_loss_mlp": 1.04012871,
      "epoch": 0.056393915709733664,
      "flos": 19858708552320.0,
      "grad_norm": 2.463538662185252,
      "language_loss": 0.84968269,
      "learning_rate": 3.992728764523326e-06,
      "loss": 0.87252855,
      "num_input_tokens_seen": 10146620,
      "step": 469,
      "time_per_iteration": 2.665590524673462
    },
    {
      "auxiliary_loss_clip": 0.01252376,
      "auxiliary_loss_mlp": 0.01049901,
      "balance_loss_clip": 1.08221054,
      "balance_loss_mlp": 1.03685331,
      "epoch": 0.05651415860037275,
      "flos": 22163779013760.0,
      "grad_norm": 1.6155816969014418,
      "language_loss": 0.80784452,
      "learning_rate": 3.99266224966623e-06,
      "loss": 0.83086729,
      "num_input_tokens_seen": 10167535,
      "step": 470,
      "time_per_iteration": 2.683539628982544
    },
    {
      "auxiliary_loss_clip": 0.01238851,
      "auxiliary_loss_mlp": 0.01041877,
      "balance_loss_clip": 1.07476354,
      "balance_loss_mlp": 1.03061771,
      "epoch": 0.05663440149101184,
      "flos": 19463548055040.0,
      "grad_norm": 2.257288582091297,
      "language_loss": 0.88032812,
      "learning_rate": 3.992595432523855e-06,
      "loss": 0.90313542,
      "num_input_tokens_seen": 10184825,
      "step": 471,
      "time_per_iteration": 2.641402244567871
    },
    {
      "auxiliary_loss_clip": 0.01224688,
      "auxiliary_loss_mlp": 0.01045207,
      "balance_loss_clip": 1.07480681,
      "balance_loss_mlp": 1.03353691,
      "epoch": 0.05675464438165093,
      "flos": 22670226823680.0,
      "grad_norm": 1.8079599410593228,
      "language_loss": 0.85897762,
      "learning_rate": 3.992528313106338e-06,
      "loss": 0.88167655,
      "num_input_tokens_seen": 10203025,
      "step": 472,
      "time_per_iteration": 2.6415812969207764
    },
    {
      "auxiliary_loss_clip": 0.01286143,
      "auxiliary_loss_mlp": 0.00765343,
      "balance_loss_clip": 1.08535147,
      "balance_loss_mlp": 1.0001328,
      "epoch": 0.05687488727229002,
      "flos": 16901177495040.0,
      "grad_norm": 2.1730951562135177,
      "language_loss": 0.81838316,
      "learning_rate": 3.9924608914238595e-06,
      "loss": 0.83889794,
      "num_input_tokens_seen": 10218020,
      "step": 473,
      "time_per_iteration": 2.565941095352173
    },
    {
      "auxiliary_loss_clip": 0.01270613,
      "auxiliary_loss_mlp": 0.01044465,
      "balance_loss_clip": 1.08378506,
      "balance_loss_mlp": 1.03157282,
      "epoch": 0.05699513016292912,
      "flos": 29168980945920.0,
      "grad_norm": 2.493490821766446,
      "language_loss": 0.83596319,
      "learning_rate": 3.992393167486648e-06,
      "loss": 0.85911393,
      "num_input_tokens_seen": 10237170,
      "step": 474,
      "time_per_iteration": 2.664119005203247
    },
    {
      "auxiliary_loss_clip": 0.01284622,
      "auxiliary_loss_mlp": 0.01049945,
      "balance_loss_clip": 1.08247864,
      "balance_loss_mlp": 1.03691554,
      "epoch": 0.05711537305356821,
      "flos": 18916197632640.0,
      "grad_norm": 2.253327412820963,
      "language_loss": 0.80934286,
      "learning_rate": 3.992325141304977e-06,
      "loss": 0.83268845,
      "num_input_tokens_seen": 10255125,
      "step": 475,
      "time_per_iteration": 2.580497980117798
    },
    {
      "auxiliary_loss_clip": 0.0122116,
      "auxiliary_loss_mlp": 0.01045124,
      "balance_loss_clip": 1.0743897,
      "balance_loss_mlp": 1.03332829,
      "epoch": 0.0572356159442073,
      "flos": 26758979879040.0,
      "grad_norm": 2.1486149398180414,
      "language_loss": 0.86641967,
      "learning_rate": 3.992256812889166e-06,
      "loss": 0.88908249,
      "num_input_tokens_seen": 10271230,
      "step": 476,
      "time_per_iteration": 2.711813449859619
    },
    {
      "auxiliary_loss_clip": 0.01285824,
      "auxiliary_loss_mlp": 0.01044094,
      "balance_loss_clip": 1.08711183,
      "balance_loss_mlp": 1.03077853,
      "epoch": 0.05735585883484639,
      "flos": 35116146840960.0,
      "grad_norm": 2.3166712591256178,
      "language_loss": 0.76771188,
      "learning_rate": 3.992188182249582e-06,
      "loss": 0.7910111,
      "num_input_tokens_seen": 10293125,
      "step": 477,
      "time_per_iteration": 2.701622724533081
    },
    {
      "auxiliary_loss_clip": 0.01250598,
      "auxiliary_loss_mlp": 0.01045489,
      "balance_loss_clip": 1.08040357,
      "balance_loss_mlp": 1.03230453,
      "epoch": 0.05747610172548548,
      "flos": 18734381965440.0,
      "grad_norm": 1.9997645238857515,
      "language_loss": 0.90440452,
      "learning_rate": 3.992119249396633e-06,
      "loss": 0.92736536,
      "num_input_tokens_seen": 10311810,
      "step": 478,
      "time_per_iteration": 2.627744674682617
    },
    {
      "auxiliary_loss_clip": 0.01244504,
      "auxiliary_loss_mlp": 0.00765191,
      "balance_loss_clip": 1.07536936,
      "balance_loss_mlp": 1.00013268,
      "epoch": 0.05759634461612457,
      "flos": 27964752554880.0,
      "grad_norm": 1.7838257933945507,
      "language_loss": 0.82092017,
      "learning_rate": 3.992050014340778e-06,
      "loss": 0.84101713,
      "num_input_tokens_seen": 10332165,
      "step": 479,
      "time_per_iteration": 2.7037198543548584
    },
    {
      "auxiliary_loss_clip": 0.01151865,
      "auxiliary_loss_mlp": 0.01019331,
      "balance_loss_clip": 1.04844546,
      "balance_loss_mlp": 1.01556361,
      "epoch": 0.057716587506763666,
      "flos": 69292009405440.0,
      "grad_norm": 1.0389851911997585,
      "language_loss": 0.55077559,
      "learning_rate": 3.99198047709252e-06,
      "loss": 0.57248759,
      "num_input_tokens_seen": 10393685,
      "step": 480,
      "time_per_iteration": 3.2599875926971436
    },
    {
      "auxiliary_loss_clip": 0.01225517,
      "auxiliary_loss_mlp": 0.01044346,
      "balance_loss_clip": 1.06961679,
      "balance_loss_mlp": 1.03118563,
      "epoch": 0.057836830397402755,
      "flos": 25009196745600.0,
      "grad_norm": 2.0398306006536586,
      "language_loss": 0.78891838,
      "learning_rate": 3.991910637662408e-06,
      "loss": 0.81161702,
      "num_input_tokens_seen": 10413975,
      "step": 481,
      "time_per_iteration": 2.6791133880615234
    },
    {
      "auxiliary_loss_clip": 0.01277702,
      "auxiliary_loss_mlp": 0.0104815,
      "balance_loss_clip": 1.08081758,
      "balance_loss_mlp": 1.03579962,
      "epoch": 0.057957073288041845,
      "flos": 25593894334080.0,
      "grad_norm": 2.9022775182528546,
      "language_loss": 0.80757415,
      "learning_rate": 3.9918404960610355e-06,
      "loss": 0.83083266,
      "num_input_tokens_seen": 10433005,
      "step": 482,
      "time_per_iteration": 2.589202404022217
    },
    {
      "auxiliary_loss_clip": 0.01271507,
      "auxiliary_loss_mlp": 0.01064802,
      "balance_loss_clip": 1.0812248,
      "balance_loss_mlp": 1.0522548,
      "epoch": 0.058077316178680935,
      "flos": 20777411733120.0,
      "grad_norm": 2.3474990242863836,
      "language_loss": 0.77201498,
      "learning_rate": 3.991770052299043e-06,
      "loss": 0.79537809,
      "num_input_tokens_seen": 10451235,
      "step": 483,
      "time_per_iteration": 2.6040334701538086
    },
    {
      "auxiliary_loss_clip": 0.012504,
      "auxiliary_loss_mlp": 0.01044498,
      "balance_loss_clip": 1.07479393,
      "balance_loss_mlp": 1.03234434,
      "epoch": 0.058197559069320025,
      "flos": 18916484941440.0,
      "grad_norm": 2.366093495335668,
      "language_loss": 0.87785608,
      "learning_rate": 3.991699306387118e-06,
      "loss": 0.900805,
      "num_input_tokens_seen": 10469705,
      "step": 484,
      "time_per_iteration": 2.639477491378784
    },
    {
      "auxiliary_loss_clip": 0.01264085,
      "auxiliary_loss_mlp": 0.01056878,
      "balance_loss_clip": 1.07866752,
      "balance_loss_mlp": 1.04350877,
      "epoch": 0.058317801959959115,
      "flos": 24863327614080.0,
      "grad_norm": 1.8776906326022489,
      "language_loss": 0.78029037,
      "learning_rate": 3.991628258335991e-06,
      "loss": 0.80349994,
      "num_input_tokens_seen": 10491910,
      "step": 485,
      "time_per_iteration": 2.771372079849243
    },
    {
      "auxiliary_loss_clip": 0.01222634,
      "auxiliary_loss_mlp": 0.01048229,
      "balance_loss_clip": 1.07086694,
      "balance_loss_mlp": 1.03460288,
      "epoch": 0.05843804485059821,
      "flos": 23257977068160.0,
      "grad_norm": 3.196519336640929,
      "language_loss": 0.87219238,
      "learning_rate": 3.991556908156442e-06,
      "loss": 0.89490104,
      "num_input_tokens_seen": 10508435,
      "step": 486,
      "time_per_iteration": 2.753873586654663
    },
    {
      "auxiliary_loss_clip": 0.0125008,
      "auxiliary_loss_mlp": 0.01049136,
      "balance_loss_clip": 1.07383764,
      "balance_loss_mlp": 1.03635049,
      "epoch": 0.0585582877412373,
      "flos": 23150532510720.0,
      "grad_norm": 1.8187273938789579,
      "language_loss": 0.87655455,
      "learning_rate": 3.9914852558592914e-06,
      "loss": 0.89954674,
      "num_input_tokens_seen": 10529485,
      "step": 487,
      "time_per_iteration": 3.490030288696289
    },
    {
      "auxiliary_loss_clip": 0.01262742,
      "auxiliary_loss_mlp": 0.01042054,
      "balance_loss_clip": 1.07959008,
      "balance_loss_mlp": 1.0294838,
      "epoch": 0.05867853063187639,
      "flos": 23506406507520.0,
      "grad_norm": 3.0969166512267554,
      "language_loss": 0.80824935,
      "learning_rate": 3.991413301455413e-06,
      "loss": 0.83129728,
      "num_input_tokens_seen": 10545935,
      "step": 488,
      "time_per_iteration": 3.335170030593872
    },
    {
      "auxiliary_loss_clip": 0.01235497,
      "auxiliary_loss_mlp": 0.01054452,
      "balance_loss_clip": 1.07361495,
      "balance_loss_mlp": 1.04227471,
      "epoch": 0.05879877352251548,
      "flos": 29495803818240.0,
      "grad_norm": 2.240678389649679,
      "language_loss": 0.77789956,
      "learning_rate": 3.991341044955719e-06,
      "loss": 0.80079901,
      "num_input_tokens_seen": 10565690,
      "step": 489,
      "time_per_iteration": 3.4928698539733887
    },
    {
      "auxiliary_loss_clip": 0.0126155,
      "auxiliary_loss_mlp": 0.00765787,
      "balance_loss_clip": 1.07562983,
      "balance_loss_mlp": 1.00016189,
      "epoch": 0.05891901641315457,
      "flos": 20157485880960.0,
      "grad_norm": 1.9682611208958198,
      "language_loss": 0.81622517,
      "learning_rate": 3.991268486371172e-06,
      "loss": 0.8364985,
      "num_input_tokens_seen": 10584245,
      "step": 490,
      "time_per_iteration": 2.8060574531555176
    },
    {
      "auxiliary_loss_clip": 0.01244284,
      "auxiliary_loss_mlp": 0.01059061,
      "balance_loss_clip": 1.07197094,
      "balance_loss_mlp": 1.0460794,
      "epoch": 0.05903925930379366,
      "flos": 24644200694400.0,
      "grad_norm": 2.141381900601277,
      "language_loss": 0.88035762,
      "learning_rate": 3.991195625712779e-06,
      "loss": 0.90339112,
      "num_input_tokens_seen": 10601210,
      "step": 491,
      "time_per_iteration": 3.451523780822754
    },
    {
      "auxiliary_loss_clip": 0.01279524,
      "auxiliary_loss_mlp": 0.01037099,
      "balance_loss_clip": 1.08127785,
      "balance_loss_mlp": 1.02448702,
      "epoch": 0.05915950219443276,
      "flos": 21250391045760.0,
      "grad_norm": 2.4487033730797525,
      "language_loss": 0.81763554,
      "learning_rate": 3.991122462991592e-06,
      "loss": 0.84080178,
      "num_input_tokens_seen": 10620730,
      "step": 492,
      "time_per_iteration": 2.55501127243042
    },
    {
      "auxiliary_loss_clip": 0.01279631,
      "auxiliary_loss_mlp": 0.01053818,
      "balance_loss_clip": 1.07846665,
      "balance_loss_mlp": 1.03994179,
      "epoch": 0.05927974508507185,
      "flos": 9902727319680.0,
      "grad_norm": 4.224674981410079,
      "language_loss": 0.81611991,
      "learning_rate": 3.991048998218712e-06,
      "loss": 0.83945441,
      "num_input_tokens_seen": 10634035,
      "step": 493,
      "time_per_iteration": 2.5507798194885254
    },
    {
      "auxiliary_loss_clip": 0.01262976,
      "auxiliary_loss_mlp": 0.01042799,
      "balance_loss_clip": 1.07591891,
      "balance_loss_mlp": 1.02976382,
      "epoch": 0.05939998797571094,
      "flos": 18259499232000.0,
      "grad_norm": 2.270025212133049,
      "language_loss": 0.7691642,
      "learning_rate": 3.990975231405281e-06,
      "loss": 0.7922219,
      "num_input_tokens_seen": 10652485,
      "step": 494,
      "time_per_iteration": 2.825798749923706
    },
    {
      "auxiliary_loss_clip": 0.01262594,
      "auxiliary_loss_mlp": 0.01057792,
      "balance_loss_clip": 1.07764196,
      "balance_loss_mlp": 1.04494143,
      "epoch": 0.05952023086635003,
      "flos": 28256598558720.0,
      "grad_norm": 1.8822811824409889,
      "language_loss": 0.78998637,
      "learning_rate": 3.990901162562491e-06,
      "loss": 0.81319028,
      "num_input_tokens_seen": 10673175,
      "step": 495,
      "time_per_iteration": 2.8355653285980225
    },
    {
      "auxiliary_loss_clip": 0.01227636,
      "auxiliary_loss_mlp": 0.00765813,
      "balance_loss_clip": 1.06904292,
      "balance_loss_mlp": 1.00011921,
      "epoch": 0.05964047375698912,
      "flos": 14902498045440.0,
      "grad_norm": 1.9067899420400232,
      "language_loss": 0.90687156,
      "learning_rate": 3.9908267917015765e-06,
      "loss": 0.92680609,
      "num_input_tokens_seen": 10691235,
      "step": 496,
      "time_per_iteration": 2.834641695022583
    },
    {
      "auxiliary_loss_clip": 0.01254186,
      "auxiliary_loss_mlp": 0.01049893,
      "balance_loss_clip": 1.07632351,
      "balance_loss_mlp": 1.03648162,
      "epoch": 0.059760716647628206,
      "flos": 23185581206400.0,
      "grad_norm": 1.9529658582542784,
      "language_loss": 0.93222392,
      "learning_rate": 3.990752118833821e-06,
      "loss": 0.95526475,
      "num_input_tokens_seen": 10708675,
      "step": 497,
      "time_per_iteration": 2.652101755142212
    },
    {
      "auxiliary_loss_clip": 0.01278935,
      "auxiliary_loss_mlp": 0.01047184,
      "balance_loss_clip": 1.08081293,
      "balance_loss_mlp": 1.03498316,
      "epoch": 0.0598809595382673,
      "flos": 22746968231040.0,
      "grad_norm": 1.7711190029344144,
      "language_loss": 0.77715874,
      "learning_rate": 3.990677143970553e-06,
      "loss": 0.80041993,
      "num_input_tokens_seen": 10729485,
      "step": 498,
      "time_per_iteration": 2.663672924041748
    },
    {
      "auxiliary_loss_clip": 0.01229819,
      "auxiliary_loss_mlp": 0.01049363,
      "balance_loss_clip": 1.07697511,
      "balance_loss_mlp": 1.0370245,
      "epoch": 0.06000120242890639,
      "flos": 22127221946880.0,
      "grad_norm": 2.364353016712349,
      "language_loss": 0.81232512,
      "learning_rate": 3.990601867123144e-06,
      "loss": 0.83511698,
      "num_input_tokens_seen": 10749210,
      "step": 499,
      "time_per_iteration": 2.663703680038452
    },
    {
      "auxiliary_loss_clip": 0.01211839,
      "auxiliary_loss_mlp": 0.01053164,
      "balance_loss_clip": 1.06825471,
      "balance_loss_mlp": 1.0413506,
      "epoch": 0.06012144531954548,
      "flos": 19171773878400.0,
      "grad_norm": 2.931374988443663,
      "language_loss": 0.85082281,
      "learning_rate": 3.990526288303014e-06,
      "loss": 0.87347287,
      "num_input_tokens_seen": 10768000,
      "step": 500,
      "time_per_iteration": 2.7065019607543945
    },
    {
      "auxiliary_loss_clip": 0.01246428,
      "auxiliary_loss_mlp": 0.00764705,
      "balance_loss_clip": 1.07767248,
      "balance_loss_mlp": 1.0001061,
      "epoch": 0.06024168821018457,
      "flos": 22783345729920.0,
      "grad_norm": 2.555422856451407,
      "language_loss": 0.90650702,
      "learning_rate": 3.9904504075216295e-06,
      "loss": 0.92661834,
      "num_input_tokens_seen": 10788760,
      "step": 501,
      "time_per_iteration": 2.6695640087127686
    },
    {
      "auxiliary_loss_clip": 0.01224168,
      "auxiliary_loss_mlp": 0.01047045,
      "balance_loss_clip": 1.06664872,
      "balance_loss_mlp": 1.03402758,
      "epoch": 0.06036193110082366,
      "flos": 18770687637120.0,
      "grad_norm": 2.3387759037397773,
      "language_loss": 0.93729424,
      "learning_rate": 3.990374224790501e-06,
      "loss": 0.9600063,
      "num_input_tokens_seen": 10806965,
      "step": 502,
      "time_per_iteration": 2.6540727615356445
    },
    {
      "auxiliary_loss_clip": 0.01248186,
      "auxiliary_loss_mlp": 0.01044985,
      "balance_loss_clip": 1.07831144,
      "balance_loss_mlp": 1.03227139,
      "epoch": 0.06048217399146275,
      "flos": 17201570935680.0,
      "grad_norm": 2.3223539321526174,
      "language_loss": 0.70795619,
      "learning_rate": 3.990297740121185e-06,
      "loss": 0.73088789,
      "num_input_tokens_seen": 10824900,
      "step": 503,
      "time_per_iteration": 2.6196703910827637
    },
    {
      "auxiliary_loss_clip": 0.01264024,
      "auxiliary_loss_mlp": 0.00765414,
      "balance_loss_clip": 1.07805657,
      "balance_loss_mlp": 1.00009596,
      "epoch": 0.06060241688210185,
      "flos": 24024131187840.0,
      "grad_norm": 1.909524278749976,
      "language_loss": 0.77913809,
      "learning_rate": 3.990220953525284e-06,
      "loss": 0.7994324,
      "num_input_tokens_seen": 10842010,
      "step": 504,
      "time_per_iteration": 2.6030960083007812
    },
    {
      "auxiliary_loss_clip": 0.01234511,
      "auxiliary_loss_mlp": 0.010468,
      "balance_loss_clip": 1.07193315,
      "balance_loss_mlp": 1.03350866,
      "epoch": 0.06072265977274094,
      "flos": 14611190745600.0,
      "grad_norm": 2.4236272251043163,
      "language_loss": 0.74305475,
      "learning_rate": 3.9901438650144465e-06,
      "loss": 0.76586783,
      "num_input_tokens_seen": 10858260,
      "step": 505,
      "time_per_iteration": 2.602219581604004
    },
    {
      "auxiliary_loss_clip": 0.01255432,
      "auxiliary_loss_mlp": 0.0104689,
      "balance_loss_clip": 1.0771842,
      "balance_loss_mlp": 1.03490341,
      "epoch": 0.06084290266338003,
      "flos": 20558284813440.0,
      "grad_norm": 3.0150272096919,
      "language_loss": 0.92135012,
      "learning_rate": 3.990066474600367e-06,
      "loss": 0.94437337,
      "num_input_tokens_seen": 10876230,
      "step": 506,
      "time_per_iteration": 2.5963540077209473
    },
    {
      "auxiliary_loss_clip": 0.01250544,
      "auxiliary_loss_mlp": 0.01045916,
      "balance_loss_clip": 1.07336879,
      "balance_loss_mlp": 1.0338347,
      "epoch": 0.06096314555401912,
      "flos": 22309217182080.0,
      "grad_norm": 1.9702131881666787,
      "language_loss": 0.68245375,
      "learning_rate": 3.989988782294786e-06,
      "loss": 0.70541835,
      "num_input_tokens_seen": 10896320,
      "step": 507,
      "time_per_iteration": 2.5814578533172607
    },
    {
      "auxiliary_loss_clip": 0.01213388,
      "auxiliary_loss_mlp": 0.01044135,
      "balance_loss_clip": 1.06772304,
      "balance_loss_mlp": 1.03258967,
      "epoch": 0.06108338844465821,
      "flos": 19131374056320.0,
      "grad_norm": 1.6843598113591054,
      "language_loss": 0.95091623,
      "learning_rate": 3.989910788109489e-06,
      "loss": 0.97349143,
      "num_input_tokens_seen": 10912970,
      "step": 508,
      "time_per_iteration": 2.646916627883911
    },
    {
      "auxiliary_loss_clip": 0.01223241,
      "auxiliary_loss_mlp": 0.01045812,
      "balance_loss_clip": 1.07105041,
      "balance_loss_mlp": 1.03419542,
      "epoch": 0.0612036313352973,
      "flos": 33584018169600.0,
      "grad_norm": 2.324021716955585,
      "language_loss": 0.74746776,
      "learning_rate": 3.989832492056307e-06,
      "loss": 0.77015817,
      "num_input_tokens_seen": 10933995,
      "step": 509,
      "time_per_iteration": 2.7751195430755615
    },
    {
      "auxiliary_loss_clip": 0.0126113,
      "auxiliary_loss_mlp": 0.01065266,
      "balance_loss_clip": 1.07912171,
      "balance_loss_mlp": 1.05252266,
      "epoch": 0.06132387422593639,
      "flos": 27490552179840.0,
      "grad_norm": 2.11969783564211,
      "language_loss": 0.8090592,
      "learning_rate": 3.989753894147119e-06,
      "loss": 0.83232313,
      "num_input_tokens_seen": 10954120,
      "step": 510,
      "time_per_iteration": 2.631540536880493
    },
    {
      "auxiliary_loss_clip": 0.01259764,
      "auxiliary_loss_mlp": 0.01048297,
      "balance_loss_clip": 1.0827086,
      "balance_loss_mlp": 1.03825998,
      "epoch": 0.061444117116575485,
      "flos": 25885057979520.0,
      "grad_norm": 2.127522715318158,
      "language_loss": 0.80073464,
      "learning_rate": 3.989674994393846e-06,
      "loss": 0.82381523,
      "num_input_tokens_seen": 10973595,
      "step": 511,
      "time_per_iteration": 2.633462905883789
    },
    {
      "auxiliary_loss_clip": 0.01259983,
      "auxiliary_loss_mlp": 0.0104612,
      "balance_loss_clip": 1.07956398,
      "balance_loss_mlp": 1.0342406,
      "epoch": 0.061564360007214575,
      "flos": 28512031150080.0,
      "grad_norm": 2.0540382646430686,
      "language_loss": 0.93818754,
      "learning_rate": 3.98959579280846e-06,
      "loss": 0.96124864,
      "num_input_tokens_seen": 10991995,
      "step": 512,
      "time_per_iteration": 2.6894302368164062
    },
    {
      "auxiliary_loss_clip": 0.01190529,
      "auxiliary_loss_mlp": 0.01057076,
      "balance_loss_clip": 1.06907701,
      "balance_loss_mlp": 1.0452323,
      "epoch": 0.061684602897853665,
      "flos": 12094355652480.0,
      "grad_norm": 2.043367070532318,
      "language_loss": 0.82960051,
      "learning_rate": 3.989516289402973e-06,
      "loss": 0.85207653,
      "num_input_tokens_seen": 11007625,
      "step": 513,
      "time_per_iteration": 3.5100958347320557
    },
    {
      "auxiliary_loss_clip": 0.01171159,
      "auxiliary_loss_mlp": 0.01043948,
      "balance_loss_clip": 1.059955,
      "balance_loss_mlp": 1.03236687,
      "epoch": 0.061804845788492754,
      "flos": 19532639865600.0,
      "grad_norm": 2.3344179504444633,
      "language_loss": 0.80111456,
      "learning_rate": 3.989436484189447e-06,
      "loss": 0.82326567,
      "num_input_tokens_seen": 11025570,
      "step": 514,
      "time_per_iteration": 3.447075843811035
    },
    {
      "auxiliary_loss_clip": 0.01260435,
      "auxiliary_loss_mlp": 0.01047756,
      "balance_loss_clip": 1.074422,
      "balance_loss_mlp": 1.0357275,
      "epoch": 0.061925088679131844,
      "flos": 15341111020800.0,
      "grad_norm": 2.6921143612041765,
      "language_loss": 0.8103888,
      "learning_rate": 3.9893563771799885e-06,
      "loss": 0.8334707,
      "num_input_tokens_seen": 11042045,
      "step": 515,
      "time_per_iteration": 3.3282310962677
    },
    {
      "auxiliary_loss_clip": 0.0127667,
      "auxiliary_loss_mlp": 0.01050917,
      "balance_loss_clip": 1.07835281,
      "balance_loss_mlp": 1.03896046,
      "epoch": 0.062045331569770934,
      "flos": 25919927107200.0,
      "grad_norm": 2.189404071044197,
      "language_loss": 0.86477757,
      "learning_rate": 3.989275968386749e-06,
      "loss": 0.88805342,
      "num_input_tokens_seen": 11059955,
      "step": 516,
      "time_per_iteration": 2.606621742248535
    },
    {
      "auxiliary_loss_clip": 0.01235541,
      "auxiliary_loss_mlp": 0.01047229,
      "balance_loss_clip": 1.06994545,
      "balance_loss_mlp": 1.03583241,
      "epoch": 0.06216557446041003,
      "flos": 28110621686400.0,
      "grad_norm": 1.9546431551730092,
      "language_loss": 0.76878774,
      "learning_rate": 3.989195257821926e-06,
      "loss": 0.79161537,
      "num_input_tokens_seen": 11078440,
      "step": 517,
      "time_per_iteration": 3.4888803958892822
    },
    {
      "auxiliary_loss_clip": 0.0124169,
      "auxiliary_loss_mlp": 0.0105274,
      "balance_loss_clip": 1.07802033,
      "balance_loss_mlp": 1.04061019,
      "epoch": 0.06228581735104912,
      "flos": 23478181395840.0,
      "grad_norm": 2.026318371043068,
      "language_loss": 0.84411228,
      "learning_rate": 3.989114245497765e-06,
      "loss": 0.86705661,
      "num_input_tokens_seen": 11098240,
      "step": 518,
      "time_per_iteration": 2.64931583404541
    },
    {
      "auxiliary_loss_clip": 0.01261106,
      "auxiliary_loss_mlp": 0.01043902,
      "balance_loss_clip": 1.07438207,
      "balance_loss_mlp": 1.03343558,
      "epoch": 0.06240606024168821,
      "flos": 15195205975680.0,
      "grad_norm": 2.125786540813677,
      "language_loss": 0.95091236,
      "learning_rate": 3.989032931426554e-06,
      "loss": 0.97396243,
      "num_input_tokens_seen": 11115395,
      "step": 519,
      "time_per_iteration": 2.5838935375213623
    },
    {
      "auxiliary_loss_clip": 0.01244845,
      "auxiliary_loss_mlp": 0.01043473,
      "balance_loss_clip": 1.07580841,
      "balance_loss_mlp": 1.03225553,
      "epoch": 0.06252630313232731,
      "flos": 20631829910400.0,
      "grad_norm": 2.42934819584401,
      "language_loss": 0.86949825,
      "learning_rate": 3.9889513156206295e-06,
      "loss": 0.89238149,
      "num_input_tokens_seen": 11134835,
      "step": 520,
      "time_per_iteration": 2.652130365371704
    },
    {
      "auxiliary_loss_clip": 0.01231963,
      "auxiliary_loss_mlp": 0.01058614,
      "balance_loss_clip": 1.07343757,
      "balance_loss_mlp": 1.04541779,
      "epoch": 0.06264654602296639,
      "flos": 20778058177920.0,
      "grad_norm": 2.8371954327795255,
      "language_loss": 0.73773623,
      "learning_rate": 3.988869398092371e-06,
      "loss": 0.76064205,
      "num_input_tokens_seen": 11154745,
      "step": 521,
      "time_per_iteration": 2.6562561988830566
    },
    {
      "auxiliary_loss_clip": 0.01240809,
      "auxiliary_loss_mlp": 0.01043903,
      "balance_loss_clip": 1.07411027,
      "balance_loss_mlp": 1.03139758,
      "epoch": 0.06276678891360549,
      "flos": 29605798241280.0,
      "grad_norm": 2.628429363213342,
      "language_loss": 0.78717422,
      "learning_rate": 3.988787178854206e-06,
      "loss": 0.81002134,
      "num_input_tokens_seen": 11174280,
      "step": 522,
      "time_per_iteration": 2.68205189704895
    },
    {
      "auxiliary_loss_clip": 0.01278509,
      "auxiliary_loss_mlp": 0.01048228,
      "balance_loss_clip": 1.08134794,
      "balance_loss_mlp": 1.03672409,
      "epoch": 0.06288703180424457,
      "flos": 22126288193280.0,
      "grad_norm": 2.2042800810250136,
      "language_loss": 0.87602359,
      "learning_rate": 3.988704657918608e-06,
      "loss": 0.89929098,
      "num_input_tokens_seen": 11193340,
      "step": 523,
      "time_per_iteration": 2.616750717163086
    },
    {
      "auxiliary_loss_clip": 0.01264149,
      "auxiliary_loss_mlp": 0.0104335,
      "balance_loss_clip": 1.08097339,
      "balance_loss_mlp": 1.03157771,
      "epoch": 0.06300727469488367,
      "flos": 14976689587200.0,
      "grad_norm": 3.0882908112088288,
      "language_loss": 0.79932868,
      "learning_rate": 3.988621835298094e-06,
      "loss": 0.82240367,
      "num_input_tokens_seen": 11210555,
      "step": 524,
      "time_per_iteration": 2.6330788135528564
    },
    {
      "auxiliary_loss_clip": 0.01273342,
      "auxiliary_loss_mlp": 0.01051038,
      "balance_loss_clip": 1.08046305,
      "balance_loss_mlp": 1.03970098,
      "epoch": 0.06312751758552275,
      "flos": 24535391420160.0,
      "grad_norm": 2.162556529675408,
      "language_loss": 0.9157061,
      "learning_rate": 3.988538711005229e-06,
      "loss": 0.93894994,
      "num_input_tokens_seen": 11230010,
      "step": 525,
      "time_per_iteration": 2.6286098957061768
    },
    {
      "auxiliary_loss_clip": 0.01251471,
      "auxiliary_loss_mlp": 0.01047489,
      "balance_loss_clip": 1.07588553,
      "balance_loss_mlp": 1.03624809,
      "epoch": 0.06324776047616185,
      "flos": 21507008785920.0,
      "grad_norm": 2.087883055641979,
      "language_loss": 0.88036543,
      "learning_rate": 3.988455285052622e-06,
      "loss": 0.903355,
      "num_input_tokens_seen": 11246190,
      "step": 526,
      "time_per_iteration": 2.630317449569702
    },
    {
      "auxiliary_loss_clip": 0.01256617,
      "auxiliary_loss_mlp": 0.01042018,
      "balance_loss_clip": 1.07741177,
      "balance_loss_mlp": 1.02954292,
      "epoch": 0.06336800336680094,
      "flos": 21688034353920.0,
      "grad_norm": 2.169447297624444,
      "language_loss": 0.8366589,
      "learning_rate": 3.98837155745293e-06,
      "loss": 0.85964519,
      "num_input_tokens_seen": 11264230,
      "step": 527,
      "time_per_iteration": 2.5888521671295166
    },
    {
      "auxiliary_loss_clip": 0.01258339,
      "auxiliary_loss_mlp": 0.0104221,
      "balance_loss_clip": 1.07819629,
      "balance_loss_mlp": 1.03028297,
      "epoch": 0.06348824625744003,
      "flos": 19500895221120.0,
      "grad_norm": 2.5797051560958226,
      "language_loss": 0.75784987,
      "learning_rate": 3.988287528218854e-06,
      "loss": 0.7808553,
      "num_input_tokens_seen": 11283015,
      "step": 528,
      "time_per_iteration": 2.609233856201172
    },
    {
      "auxiliary_loss_clip": 0.0125655,
      "auxiliary_loss_mlp": 0.01054859,
      "balance_loss_clip": 1.07893288,
      "balance_loss_mlp": 1.0440284,
      "epoch": 0.06360848914807912,
      "flos": 15481233976320.0,
      "grad_norm": 2.1687781662981367,
      "language_loss": 0.90363562,
      "learning_rate": 3.98820319736314e-06,
      "loss": 0.92674971,
      "num_input_tokens_seen": 11299630,
      "step": 529,
      "time_per_iteration": 2.5724587440490723
    },
    {
      "auxiliary_loss_clip": 0.01229638,
      "auxiliary_loss_mlp": 0.01041206,
      "balance_loss_clip": 1.07270598,
      "balance_loss_mlp": 1.02949929,
      "epoch": 0.0637287320387182,
      "flos": 20593369422720.0,
      "grad_norm": 1.907999202349445,
      "language_loss": 0.85219598,
      "learning_rate": 3.988118564898582e-06,
      "loss": 0.87490445,
      "num_input_tokens_seen": 11319170,
      "step": 530,
      "time_per_iteration": 2.691847085952759
    },
    {
      "auxiliary_loss_clip": 0.01219407,
      "auxiliary_loss_mlp": 0.00764728,
      "balance_loss_clip": 1.07325852,
      "balance_loss_mlp": 1.00017428,
      "epoch": 0.0638489749293573,
      "flos": 17412222245760.0,
      "grad_norm": 2.200254877225638,
      "language_loss": 0.89075184,
      "learning_rate": 3.988033630838019e-06,
      "loss": 0.91059315,
      "num_input_tokens_seen": 11333210,
      "step": 531,
      "time_per_iteration": 2.6956379413604736
    },
    {
      "auxiliary_loss_clip": 0.01261038,
      "auxiliary_loss_mlp": 0.01057138,
      "balance_loss_clip": 1.07915831,
      "balance_loss_mlp": 1.04488337,
      "epoch": 0.0639692178199964,
      "flos": 23807661874560.0,
      "grad_norm": 1.9362046985089154,
      "language_loss": 0.88025516,
      "learning_rate": 3.987948395194334e-06,
      "loss": 0.9034369,
      "num_input_tokens_seen": 11355590,
      "step": 532,
      "time_per_iteration": 2.668919563293457
    },
    {
      "auxiliary_loss_clip": 0.01253883,
      "auxiliary_loss_mlp": 0.01046992,
      "balance_loss_clip": 1.07495499,
      "balance_loss_mlp": 1.03581595,
      "epoch": 0.06408946071063548,
      "flos": 18477225521280.0,
      "grad_norm": 2.0093428933547646,
      "language_loss": 0.76737905,
      "learning_rate": 3.987862857980458e-06,
      "loss": 0.79038787,
      "num_input_tokens_seen": 11371535,
      "step": 533,
      "time_per_iteration": 2.6454381942749023
    },
    {
      "auxiliary_loss_clip": 0.01231488,
      "auxiliary_loss_mlp": 0.01041555,
      "balance_loss_clip": 1.07368457,
      "balance_loss_mlp": 1.02990234,
      "epoch": 0.06420970360127458,
      "flos": 27162220936320.0,
      "grad_norm": 2.5432495496211383,
      "language_loss": 0.76837718,
      "learning_rate": 3.987777019209368e-06,
      "loss": 0.79110765,
      "num_input_tokens_seen": 11392050,
      "step": 534,
      "time_per_iteration": 2.7288451194763184
    },
    {
      "auxiliary_loss_clip": 0.01272522,
      "auxiliary_loss_mlp": 0.01059962,
      "balance_loss_clip": 1.07847691,
      "balance_loss_mlp": 1.04931068,
      "epoch": 0.06432994649191366,
      "flos": 23659673840640.0,
      "grad_norm": 1.7103509691446848,
      "language_loss": 0.80824757,
      "learning_rate": 3.987690878894084e-06,
      "loss": 0.83157241,
      "num_input_tokens_seen": 11411765,
      "step": 535,
      "time_per_iteration": 2.615147113800049
    },
    {
      "auxiliary_loss_clip": 0.01244894,
      "auxiliary_loss_mlp": 0.01044388,
      "balance_loss_clip": 1.07421041,
      "balance_loss_mlp": 1.03303981,
      "epoch": 0.06445018938255276,
      "flos": 23403953940480.0,
      "grad_norm": 2.247405817676521,
      "language_loss": 0.84957033,
      "learning_rate": 3.987604437047673e-06,
      "loss": 0.87246317,
      "num_input_tokens_seen": 11431565,
      "step": 536,
      "time_per_iteration": 2.6302084922790527
    },
    {
      "auxiliary_loss_clip": 0.01252186,
      "auxiliary_loss_mlp": 0.01056235,
      "balance_loss_clip": 1.0735687,
      "balance_loss_mlp": 1.04409933,
      "epoch": 0.06457043227319184,
      "flos": 19646692525440.0,
      "grad_norm": 2.028803833334278,
      "language_loss": 0.77318633,
      "learning_rate": 3.987517693683251e-06,
      "loss": 0.79627049,
      "num_input_tokens_seen": 11450140,
      "step": 537,
      "time_per_iteration": 2.6171834468841553
    },
    {
      "auxiliary_loss_clip": 0.01235869,
      "auxiliary_loss_mlp": 0.01055873,
      "balance_loss_clip": 1.07624435,
      "balance_loss_mlp": 1.04348755,
      "epoch": 0.06469067516383094,
      "flos": 16978744915200.0,
      "grad_norm": 2.5694532576729787,
      "language_loss": 0.95938122,
      "learning_rate": 3.9874306488139745e-06,
      "loss": 0.98229861,
      "num_input_tokens_seen": 11465400,
      "step": 538,
      "time_per_iteration": 3.455869197845459
    },
    {
      "auxiliary_loss_clip": 0.01223202,
      "auxiliary_loss_mlp": 0.0104354,
      "balance_loss_clip": 1.07426751,
      "balance_loss_mlp": 1.03230453,
      "epoch": 0.06481091805447003,
      "flos": 23296401642240.0,
      "grad_norm": 1.9402330626601905,
      "language_loss": 0.87993586,
      "learning_rate": 3.987343302453049e-06,
      "loss": 0.90260327,
      "num_input_tokens_seen": 11486675,
      "step": 539,
      "time_per_iteration": 2.7129294872283936
    },
    {
      "auxiliary_loss_clip": 0.01235147,
      "auxiliary_loss_mlp": 0.01043343,
      "balance_loss_clip": 1.07344604,
      "balance_loss_mlp": 1.03191686,
      "epoch": 0.06493116094510912,
      "flos": 29172356824320.0,
      "grad_norm": 1.6105975875237397,
      "language_loss": 0.82369244,
      "learning_rate": 3.987255654613724e-06,
      "loss": 0.84647739,
      "num_input_tokens_seen": 11510440,
      "step": 540,
      "time_per_iteration": 3.4826035499572754
    },
    {
      "auxiliary_loss_clip": 0.01219884,
      "auxiliary_loss_mlp": 0.01047173,
      "balance_loss_clip": 1.06847107,
      "balance_loss_mlp": 1.03591418,
      "epoch": 0.06505140383574821,
      "flos": 19865065259520.0,
      "grad_norm": 3.0762599994689683,
      "language_loss": 0.70741105,
      "learning_rate": 3.987167705309296e-06,
      "loss": 0.73008168,
      "num_input_tokens_seen": 11529715,
      "step": 541,
      "time_per_iteration": 2.676022529602051
    },
    {
      "auxiliary_loss_clip": 0.01252561,
      "auxiliary_loss_mlp": 0.00763634,
      "balance_loss_clip": 1.07284391,
      "balance_loss_mlp": 1.0001626,
      "epoch": 0.0651716467263873,
      "flos": 17924703540480.0,
      "grad_norm": 2.113963994377332,
      "language_loss": 0.95046639,
      "learning_rate": 3.987079454553108e-06,
      "loss": 0.97062832,
      "num_input_tokens_seen": 11547665,
      "step": 542,
      "time_per_iteration": 3.4150912761688232
    },
    {
      "auxiliary_loss_clip": 0.0122367,
      "auxiliary_loss_mlp": 0.01051054,
      "balance_loss_clip": 1.07699931,
      "balance_loss_mlp": 1.03880548,
      "epoch": 0.0652918896170264,
      "flos": 20842840356480.0,
      "grad_norm": 1.8116096917957822,
      "language_loss": 0.91254866,
      "learning_rate": 3.986990902358546e-06,
      "loss": 0.93529588,
      "num_input_tokens_seen": 11564605,
      "step": 543,
      "time_per_iteration": 2.6770262718200684
    },
    {
      "auxiliary_loss_clip": 0.0125858,
      "auxiliary_loss_mlp": 0.01047128,
      "balance_loss_clip": 1.07661843,
      "balance_loss_mlp": 1.03601193,
      "epoch": 0.06541213250766549,
      "flos": 21872507627520.0,
      "grad_norm": 1.992415342854652,
      "language_loss": 0.9335835,
      "learning_rate": 3.986902048739045e-06,
      "loss": 0.9566406,
      "num_input_tokens_seen": 11584550,
      "step": 544,
      "time_per_iteration": 3.4635348320007324
    },
    {
      "auxiliary_loss_clip": 0.01236475,
      "auxiliary_loss_mlp": 0.01046858,
      "balance_loss_clip": 1.07139945,
      "balance_loss_mlp": 1.03407884,
      "epoch": 0.06553237539830457,
      "flos": 23110743219840.0,
      "grad_norm": 2.7779404446210365,
      "language_loss": 0.79984528,
      "learning_rate": 3.986812893708082e-06,
      "loss": 0.82267857,
      "num_input_tokens_seen": 11600740,
      "step": 545,
      "time_per_iteration": 2.614450454711914
    },
    {
      "auxiliary_loss_clip": 0.01241546,
      "auxiliary_loss_mlp": 0.01040111,
      "balance_loss_clip": 1.07030034,
      "balance_loss_mlp": 1.0275228,
      "epoch": 0.06565261828894367,
      "flos": 17923769786880.0,
      "grad_norm": 5.6602268639259625,
      "language_loss": 0.81279171,
      "learning_rate": 3.9867234372791826e-06,
      "loss": 0.83560836,
      "num_input_tokens_seen": 11618695,
      "step": 546,
      "time_per_iteration": 2.6608729362487793
    },
    {
      "auxiliary_loss_clip": 0.01251632,
      "auxiliary_loss_mlp": 0.0104503,
      "balance_loss_clip": 1.07362247,
      "balance_loss_mlp": 1.03331137,
      "epoch": 0.06577286117958275,
      "flos": 22783058421120.0,
      "grad_norm": 1.5139172224092556,
      "language_loss": 0.87150651,
      "learning_rate": 3.986633679465918e-06,
      "loss": 0.89447308,
      "num_input_tokens_seen": 11638850,
      "step": 547,
      "time_per_iteration": 2.596078872680664
    },
    {
      "auxiliary_loss_clip": 0.01213132,
      "auxiliary_loss_mlp": 0.01048154,
      "balance_loss_clip": 1.07187581,
      "balance_loss_mlp": 1.03727019,
      "epoch": 0.06589310407022185,
      "flos": 23696194993920.0,
      "grad_norm": 2.4132777934797045,
      "language_loss": 0.80538988,
      "learning_rate": 3.986543620281904e-06,
      "loss": 0.82800269,
      "num_input_tokens_seen": 11658500,
      "step": 548,
      "time_per_iteration": 2.6817259788513184
    },
    {
      "auxiliary_loss_clip": 0.01217636,
      "auxiliary_loss_mlp": 0.01043277,
      "balance_loss_clip": 1.06776524,
      "balance_loss_mlp": 1.03266716,
      "epoch": 0.06601334696086093,
      "flos": 26864772410880.0,
      "grad_norm": 1.6736101607819514,
      "language_loss": 0.91200387,
      "learning_rate": 3.986453259740802e-06,
      "loss": 0.93461299,
      "num_input_tokens_seen": 11676670,
      "step": 549,
      "time_per_iteration": 2.6795802116394043
    },
    {
      "auxiliary_loss_clip": 0.01242286,
      "auxiliary_loss_mlp": 0.0105535,
      "balance_loss_clip": 1.07764482,
      "balance_loss_mlp": 1.04350662,
      "epoch": 0.06613358985150003,
      "flos": 12567694101120.0,
      "grad_norm": 3.2093966990138516,
      "language_loss": 0.78900301,
      "learning_rate": 3.986362597856319e-06,
      "loss": 0.81197941,
      "num_input_tokens_seen": 11693170,
      "step": 550,
      "time_per_iteration": 2.6329336166381836
    },
    {
      "auxiliary_loss_clip": 0.0123304,
      "auxiliary_loss_mlp": 0.00765553,
      "balance_loss_clip": 1.0702951,
      "balance_loss_mlp": 1.00016415,
      "epoch": 0.06625383274213913,
      "flos": 18332505624960.0,
      "grad_norm": 2.6527589830274714,
      "language_loss": 0.82140541,
      "learning_rate": 3.986271634642211e-06,
      "loss": 0.84139132,
      "num_input_tokens_seen": 11710150,
      "step": 551,
      "time_per_iteration": 2.6207046508789062
    },
    {
      "auxiliary_loss_clip": 0.01271643,
      "auxiliary_loss_mlp": 0.01045794,
      "balance_loss_clip": 1.07981014,
      "balance_loss_mlp": 1.03420675,
      "epoch": 0.06637407563277821,
      "flos": 15375585098880.0,
      "grad_norm": 2.1735218353436054,
      "language_loss": 0.81802225,
      "learning_rate": 3.986180370112274e-06,
      "loss": 0.84119666,
      "num_input_tokens_seen": 11726670,
      "step": 552,
      "time_per_iteration": 2.536914587020874
    },
    {
      "auxiliary_loss_clip": 0.01255075,
      "auxiliary_loss_mlp": 0.00765468,
      "balance_loss_clip": 1.07515502,
      "balance_loss_mlp": 1.00020313,
      "epoch": 0.0664943185234173,
      "flos": 24025244509440.0,
      "grad_norm": 1.7439477321207166,
      "language_loss": 0.74575949,
      "learning_rate": 3.986088804280354e-06,
      "loss": 0.76596498,
      "num_input_tokens_seen": 11746400,
      "step": 553,
      "time_per_iteration": 2.631284475326538
    },
    {
      "auxiliary_loss_clip": 0.01238389,
      "auxiliary_loss_mlp": 0.01042547,
      "balance_loss_clip": 1.07316065,
      "balance_loss_mlp": 1.03142476,
      "epoch": 0.06661456141405639,
      "flos": 20957503547520.0,
      "grad_norm": 2.6278381746747557,
      "language_loss": 0.94006652,
      "learning_rate": 3.985996937160342e-06,
      "loss": 0.96287584,
      "num_input_tokens_seen": 11765590,
      "step": 554,
      "time_per_iteration": 2.636082172393799
    },
    {
      "auxiliary_loss_clip": 0.01254282,
      "auxiliary_loss_mlp": 0.01043477,
      "balance_loss_clip": 1.08012795,
      "balance_loss_mlp": 1.03277183,
      "epoch": 0.06673480430469549,
      "flos": 52223953322880.0,
      "grad_norm": 1.9934659938001698,
      "language_loss": 0.68913209,
      "learning_rate": 3.985904768766173e-06,
      "loss": 0.71210963,
      "num_input_tokens_seen": 11788365,
      "step": 555,
      "time_per_iteration": 2.831266403198242
    },
    {
      "auxiliary_loss_clip": 0.01225772,
      "auxiliary_loss_mlp": 0.01053713,
      "balance_loss_clip": 1.0713582,
      "balance_loss_mlp": 1.04082084,
      "epoch": 0.06685504719533458,
      "flos": 16217079995520.0,
      "grad_norm": 2.296471053138094,
      "language_loss": 0.76338327,
      "learning_rate": 3.98581229911183e-06,
      "loss": 0.78617817,
      "num_input_tokens_seen": 11807285,
      "step": 556,
      "time_per_iteration": 2.6244442462921143
    },
    {
      "auxiliary_loss_clip": 0.0124989,
      "auxiliary_loss_mlp": 0.010459,
      "balance_loss_clip": 1.07046759,
      "balance_loss_mlp": 1.03424096,
      "epoch": 0.06697529008597367,
      "flos": 22491535639680.0,
      "grad_norm": 1.8626137658438633,
      "language_loss": 0.92026615,
      "learning_rate": 3.985719528211341e-06,
      "loss": 0.94322407,
      "num_input_tokens_seen": 11826655,
      "step": 557,
      "time_per_iteration": 2.5586485862731934
    },
    {
      "auxiliary_loss_clip": 0.01148461,
      "auxiliary_loss_mlp": 0.01007209,
      "balance_loss_clip": 1.04742885,
      "balance_loss_mlp": 1.00272715,
      "epoch": 0.06709553297661276,
      "flos": 62688216936960.0,
      "grad_norm": 0.8434689346951716,
      "language_loss": 0.63035369,
      "learning_rate": 3.985626456078777e-06,
      "loss": 0.65191042,
      "num_input_tokens_seen": 11891310,
      "step": 558,
      "time_per_iteration": 3.2917227745056152
    },
    {
      "auxiliary_loss_clip": 0.01223988,
      "auxiliary_loss_mlp": 0.01064637,
      "balance_loss_clip": 1.07326961,
      "balance_loss_mlp": 1.05341959,
      "epoch": 0.06721577586725185,
      "flos": 11216590997760.0,
      "grad_norm": 2.094885742819683,
      "language_loss": 0.86666155,
      "learning_rate": 3.985533082728259e-06,
      "loss": 0.88954777,
      "num_input_tokens_seen": 11906965,
      "step": 559,
      "time_per_iteration": 2.6066524982452393
    },
    {
      "auxiliary_loss_clip": 0.01265259,
      "auxiliary_loss_mlp": 0.01046213,
      "balance_loss_clip": 1.07319772,
      "balance_loss_mlp": 1.03360105,
      "epoch": 0.06733601875789094,
      "flos": 25922189664000.0,
      "grad_norm": 2.3755254895534,
      "language_loss": 0.74948728,
      "learning_rate": 3.985439408173951e-06,
      "loss": 0.77260208,
      "num_input_tokens_seen": 11927190,
      "step": 560,
      "time_per_iteration": 2.5739145278930664
    },
    {
      "auxiliary_loss_clip": 0.01270558,
      "auxiliary_loss_mlp": 0.01053039,
      "balance_loss_clip": 1.07823896,
      "balance_loss_mlp": 1.04112399,
      "epoch": 0.06745626164853002,
      "flos": 20813645577600.0,
      "grad_norm": 1.8919298430118658,
      "language_loss": 0.70740288,
      "learning_rate": 3.9853454324300634e-06,
      "loss": 0.73063886,
      "num_input_tokens_seen": 11946400,
      "step": 561,
      "time_per_iteration": 2.533182382583618
    },
    {
      "auxiliary_loss_clip": 0.01193327,
      "auxiliary_loss_mlp": 0.0103685,
      "balance_loss_clip": 1.06375563,
      "balance_loss_mlp": 1.02568042,
      "epoch": 0.06757650453916912,
      "flos": 19829262378240.0,
      "grad_norm": 2.006314212278964,
      "language_loss": 0.78276628,
      "learning_rate": 3.985251155510852e-06,
      "loss": 0.80506802,
      "num_input_tokens_seen": 11965430,
      "step": 562,
      "time_per_iteration": 2.691563606262207
    },
    {
      "auxiliary_loss_clip": 0.01204275,
      "auxiliary_loss_mlp": 0.01047113,
      "balance_loss_clip": 1.06972229,
      "balance_loss_mlp": 1.03574049,
      "epoch": 0.06769674742980822,
      "flos": 25739224761600.0,
      "grad_norm": 1.863915646526964,
      "language_loss": 0.80191898,
      "learning_rate": 3.98515657743062e-06,
      "loss": 0.82443291,
      "num_input_tokens_seen": 11984895,
      "step": 563,
      "time_per_iteration": 2.700944423675537
    },
    {
      "auxiliary_loss_clip": 0.012355,
      "auxiliary_loss_mlp": 0.01040289,
      "balance_loss_clip": 1.0714339,
      "balance_loss_mlp": 1.02878547,
      "epoch": 0.0678169903204473,
      "flos": 13074788355840.0,
      "grad_norm": 2.05161011808784,
      "language_loss": 0.77685714,
      "learning_rate": 3.985061698203711e-06,
      "loss": 0.79961509,
      "num_input_tokens_seen": 12002010,
      "step": 564,
      "time_per_iteration": 3.3761446475982666
    },
    {
      "auxiliary_loss_clip": 0.0116595,
      "auxiliary_loss_mlp": 0.01005224,
      "balance_loss_clip": 1.04803491,
      "balance_loss_mlp": 1.00083721,
      "epoch": 0.0679372332110864,
      "flos": 70865830788480.0,
      "grad_norm": 0.8835995455287483,
      "language_loss": 0.63859272,
      "learning_rate": 3.984966517844523e-06,
      "loss": 0.66030455,
      "num_input_tokens_seen": 12057255,
      "step": 565,
      "time_per_iteration": 3.094402551651001
    },
    {
      "auxiliary_loss_clip": 0.01268775,
      "auxiliary_loss_mlp": 0.01052581,
      "balance_loss_clip": 1.0763849,
      "balance_loss_mlp": 1.03927147,
      "epoch": 0.06805747610172548,
      "flos": 28256418990720.0,
      "grad_norm": 2.346608830005219,
      "language_loss": 0.80287027,
      "learning_rate": 3.984871036367492e-06,
      "loss": 0.8260839,
      "num_input_tokens_seen": 12077280,
      "step": 566,
      "time_per_iteration": 3.350151300430298
    },
    {
      "auxiliary_loss_clip": 0.01253555,
      "auxiliary_loss_mlp": 0.00764731,
      "balance_loss_clip": 1.07732797,
      "balance_loss_mlp": 1.0002383,
      "epoch": 0.06817771899236458,
      "flos": 20120533764480.0,
      "grad_norm": 1.8372645751282337,
      "language_loss": 0.8299458,
      "learning_rate": 3.984775253787102e-06,
      "loss": 0.85012859,
      "num_input_tokens_seen": 12095570,
      "step": 567,
      "time_per_iteration": 2.581946611404419
    },
    {
      "auxiliary_loss_clip": 0.01254616,
      "auxiliary_loss_mlp": 0.01049241,
      "balance_loss_clip": 1.07276189,
      "balance_loss_mlp": 1.03718305,
      "epoch": 0.06829796188300366,
      "flos": 17930629284480.0,
      "grad_norm": 2.761085261091875,
      "language_loss": 0.87530982,
      "learning_rate": 3.984679170117885e-06,
      "loss": 0.89834845,
      "num_input_tokens_seen": 12111775,
      "step": 568,
      "time_per_iteration": 3.3261606693267822
    },
    {
      "auxiliary_loss_clip": 0.01248153,
      "auxiliary_loss_mlp": 0.01042395,
      "balance_loss_clip": 1.07280684,
      "balance_loss_mlp": 1.03056371,
      "epoch": 0.06841820477364276,
      "flos": 14501627285760.0,
      "grad_norm": 2.382805183823801,
      "language_loss": 0.78667068,
      "learning_rate": 3.984582785374415e-06,
      "loss": 0.80957621,
      "num_input_tokens_seen": 12129215,
      "step": 569,
      "time_per_iteration": 2.527209520339966
    },
    {
      "auxiliary_loss_clip": 0.01239002,
      "auxiliary_loss_mlp": 0.00764958,
      "balance_loss_clip": 1.0745995,
      "balance_loss_mlp": 1.0001905,
      "epoch": 0.06853844766428185,
      "flos": 21938474954880.0,
      "grad_norm": 1.8555510296166642,
      "language_loss": 0.80676079,
      "learning_rate": 3.9844860995713155e-06,
      "loss": 0.82680035,
      "num_input_tokens_seen": 12148755,
      "step": 570,
      "time_per_iteration": 3.444244146347046
    },
    {
      "auxiliary_loss_clip": 0.01252332,
      "auxiliary_loss_mlp": 0.01042293,
      "balance_loss_clip": 1.07786918,
      "balance_loss_mlp": 1.03080094,
      "epoch": 0.06865869055492094,
      "flos": 16800628348800.0,
      "grad_norm": 3.6142838903423478,
      "language_loss": 0.8301686,
      "learning_rate": 3.9843891127232524e-06,
      "loss": 0.85311478,
      "num_input_tokens_seen": 12166290,
      "step": 571,
      "time_per_iteration": 2.507911205291748
    },
    {
      "auxiliary_loss_clip": 0.01195819,
      "auxiliary_loss_mlp": 0.01047849,
      "balance_loss_clip": 1.06633854,
      "balance_loss_mlp": 1.0361253,
      "epoch": 0.06877893344556003,
      "flos": 19937281553280.0,
      "grad_norm": 2.8282058937793004,
      "language_loss": 0.66688108,
      "learning_rate": 3.984291824844938e-06,
      "loss": 0.6893177,
      "num_input_tokens_seen": 12181385,
      "step": 572,
      "time_per_iteration": 2.650733709335327
    },
    {
      "auxiliary_loss_clip": 0.01266473,
      "auxiliary_loss_mlp": 0.01042681,
      "balance_loss_clip": 1.07532477,
      "balance_loss_mlp": 1.02953243,
      "epoch": 0.06889917633619912,
      "flos": 23039388852480.0,
      "grad_norm": 2.355002766446531,
      "language_loss": 0.85251236,
      "learning_rate": 3.984194235951132e-06,
      "loss": 0.8756038,
      "num_input_tokens_seen": 12197530,
      "step": 573,
      "time_per_iteration": 2.5389151573181152
    },
    {
      "auxiliary_loss_clip": 0.01271923,
      "auxiliary_loss_mlp": 0.01054624,
      "balance_loss_clip": 1.0799644,
      "balance_loss_mlp": 1.0432272,
      "epoch": 0.06901941922683821,
      "flos": 20960556203520.0,
      "grad_norm": 4.43937720080714,
      "language_loss": 0.8419416,
      "learning_rate": 3.9840963460566375e-06,
      "loss": 0.86520708,
      "num_input_tokens_seen": 12216310,
      "step": 574,
      "time_per_iteration": 2.6067237854003906
    },
    {
      "auxiliary_loss_clip": 0.01169661,
      "auxiliary_loss_mlp": 0.01037272,
      "balance_loss_clip": 1.06198335,
      "balance_loss_mlp": 1.02562571,
      "epoch": 0.06913966211747731,
      "flos": 24821850384000.0,
      "grad_norm": 3.311179596209405,
      "language_loss": 0.894072,
      "learning_rate": 3.983998155176305e-06,
      "loss": 0.91614133,
      "num_input_tokens_seen": 12236670,
      "step": 575,
      "time_per_iteration": 2.763029098510742
    },
    {
      "auxiliary_loss_clip": 0.01160508,
      "auxiliary_loss_mlp": 0.010049,
      "balance_loss_clip": 1.04397118,
      "balance_loss_mlp": 1.00072801,
      "epoch": 0.06925990500811639,
      "flos": 58367446957440.0,
      "grad_norm": 0.830623703357473,
      "language_loss": 0.57063115,
      "learning_rate": 3.9838996633250305e-06,
      "loss": 0.59228522,
      "num_input_tokens_seen": 12297185,
      "step": 576,
      "time_per_iteration": 3.122347116470337
    },
    {
      "auxiliary_loss_clip": 0.01249599,
      "auxiliary_loss_mlp": 0.01044346,
      "balance_loss_clip": 1.07066917,
      "balance_loss_mlp": 1.0330683,
      "epoch": 0.06938014789875549,
      "flos": 12749940731520.0,
      "grad_norm": 2.3058349790249943,
      "language_loss": 0.88155687,
      "learning_rate": 3.983800870517753e-06,
      "loss": 0.90449637,
      "num_input_tokens_seen": 12313975,
      "step": 577,
      "time_per_iteration": 2.5862770080566406
    },
    {
      "auxiliary_loss_clip": 0.01254505,
      "auxiliary_loss_mlp": 0.01044456,
      "balance_loss_clip": 1.08156228,
      "balance_loss_mlp": 1.03395391,
      "epoch": 0.06950039078939457,
      "flos": 22820226019200.0,
      "grad_norm": 3.824997725836329,
      "language_loss": 0.77813506,
      "learning_rate": 3.983701776769463e-06,
      "loss": 0.80112469,
      "num_input_tokens_seen": 12331385,
      "step": 578,
      "time_per_iteration": 2.5845932960510254
    },
    {
      "auxiliary_loss_clip": 0.01249057,
      "auxiliary_loss_mlp": 0.01044742,
      "balance_loss_clip": 1.07783639,
      "balance_loss_mlp": 1.03361964,
      "epoch": 0.06962063368003367,
      "flos": 21941348042880.0,
      "grad_norm": 1.8842813170030117,
      "language_loss": 0.85491693,
      "learning_rate": 3.9836023820951885e-06,
      "loss": 0.87785494,
      "num_input_tokens_seen": 12350600,
      "step": 579,
      "time_per_iteration": 2.6369168758392334
    },
    {
      "auxiliary_loss_clip": 0.01209793,
      "auxiliary_loss_mlp": 0.01039378,
      "balance_loss_clip": 1.06545925,
      "balance_loss_mlp": 1.02808857,
      "epoch": 0.06974087657067275,
      "flos": 20706021452160.0,
      "grad_norm": 2.209766851651428,
      "language_loss": 0.68487215,
      "learning_rate": 3.983502686510011e-06,
      "loss": 0.70736384,
      "num_input_tokens_seen": 12371430,
      "step": 580,
      "time_per_iteration": 2.6180615425109863
    },
    {
      "auxiliary_loss_clip": 0.01252036,
      "auxiliary_loss_mlp": 0.00764583,
      "balance_loss_clip": 1.07233715,
      "balance_loss_mlp": 1.00018048,
      "epoch": 0.06986111946131185,
      "flos": 22638230784000.0,
      "grad_norm": 2.0430110505292576,
      "language_loss": 0.73393524,
      "learning_rate": 3.9834026900290525e-06,
      "loss": 0.7541014,
      "num_input_tokens_seen": 12390825,
      "step": 581,
      "time_per_iteration": 2.574289321899414
    },
    {
      "auxiliary_loss_clip": 0.01271397,
      "auxiliary_loss_mlp": 0.01053371,
      "balance_loss_clip": 1.07903743,
      "balance_loss_mlp": 1.04242194,
      "epoch": 0.06998136235195095,
      "flos": 26943453152640.0,
      "grad_norm": 1.8425962360295622,
      "language_loss": 1.00423014,
      "learning_rate": 3.983302392667482e-06,
      "loss": 1.02747786,
      "num_input_tokens_seen": 12411670,
      "step": 582,
      "time_per_iteration": 2.5741097927093506
    },
    {
      "auxiliary_loss_clip": 0.01249497,
      "auxiliary_loss_mlp": 0.01039405,
      "balance_loss_clip": 1.07566059,
      "balance_loss_mlp": 1.0282824,
      "epoch": 0.07010160524259003,
      "flos": 22492505306880.0,
      "grad_norm": 1.697367303578273,
      "language_loss": 0.93736303,
      "learning_rate": 3.983201794440517e-06,
      "loss": 0.96025205,
      "num_input_tokens_seen": 12431245,
      "step": 583,
      "time_per_iteration": 2.582667827606201
    },
    {
      "auxiliary_loss_clip": 0.0122792,
      "auxiliary_loss_mlp": 0.01048745,
      "balance_loss_clip": 1.07451367,
      "balance_loss_mlp": 1.03746176,
      "epoch": 0.07022184813322913,
      "flos": 18332541538560.0,
      "grad_norm": 1.7939622328596636,
      "language_loss": 0.67671353,
      "learning_rate": 3.9831008953634165e-06,
      "loss": 0.69948018,
      "num_input_tokens_seen": 12450535,
      "step": 584,
      "time_per_iteration": 2.5641040802001953
    },
    {
      "auxiliary_loss_clip": 0.01190205,
      "auxiliary_loss_mlp": 0.01037188,
      "balance_loss_clip": 1.06516588,
      "balance_loss_mlp": 1.02477789,
      "epoch": 0.07034209102386821,
      "flos": 24675550289280.0,
      "grad_norm": 2.045445422766,
      "language_loss": 0.81114799,
      "learning_rate": 3.9829996954514864e-06,
      "loss": 0.83342195,
      "num_input_tokens_seen": 12469675,
      "step": 585,
      "time_per_iteration": 2.710773229598999
    },
    {
      "auxiliary_loss_clip": 0.01240791,
      "auxiliary_loss_mlp": 0.0104533,
      "balance_loss_clip": 1.07287335,
      "balance_loss_mlp": 1.03384423,
      "epoch": 0.0704623339145073,
      "flos": 25995878415360.0,
      "grad_norm": 1.8412923705667965,
      "language_loss": 0.84121746,
      "learning_rate": 3.982898194720079e-06,
      "loss": 0.8640787,
      "num_input_tokens_seen": 12490405,
      "step": 586,
      "time_per_iteration": 2.6194872856140137
    },
    {
      "auxiliary_loss_clip": 0.01231159,
      "auxiliary_loss_mlp": 0.00764576,
      "balance_loss_clip": 1.07606363,
      "balance_loss_mlp": 1.00019836,
      "epoch": 0.0705825768051464,
      "flos": 25338318088320.0,
      "grad_norm": 2.1615024574675985,
      "language_loss": 0.82324845,
      "learning_rate": 3.982796393184592e-06,
      "loss": 0.84320581,
      "num_input_tokens_seen": 12509485,
      "step": 587,
      "time_per_iteration": 2.6578071117401123
    },
    {
      "auxiliary_loss_clip": 0.01149022,
      "auxiliary_loss_mlp": 0.01007042,
      "balance_loss_clip": 1.04371333,
      "balance_loss_mlp": 1.00282216,
      "epoch": 0.07070281969578548,
      "flos": 66047552507520.0,
      "grad_norm": 0.7955068865099313,
      "language_loss": 0.62634337,
      "learning_rate": 3.98269429086047e-06,
      "loss": 0.64790404,
      "num_input_tokens_seen": 12567325,
      "step": 588,
      "time_per_iteration": 3.0219156742095947
    },
    {
      "auxiliary_loss_clip": 0.01215297,
      "auxiliary_loss_mlp": 0.01046663,
      "balance_loss_clip": 1.06680799,
      "balance_loss_mlp": 1.03340697,
      "epoch": 0.07082306258642458,
      "flos": 23653568528640.0,
      "grad_norm": 3.0462474769013586,
      "language_loss": 0.86191773,
      "learning_rate": 3.982591887763199e-06,
      "loss": 0.88453728,
      "num_input_tokens_seen": 12584785,
      "step": 589,
      "time_per_iteration": 2.5864641666412354
    },
    {
      "auxiliary_loss_clip": 0.01198788,
      "auxiliary_loss_mlp": 0.01044468,
      "balance_loss_clip": 1.06164217,
      "balance_loss_mlp": 1.03285122,
      "epoch": 0.07094330547706366,
      "flos": 13880049408000.0,
      "grad_norm": 2.1272604177637096,
      "language_loss": 0.81982666,
      "learning_rate": 3.982489183908316e-06,
      "loss": 0.84225923,
      "num_input_tokens_seen": 12601205,
      "step": 590,
      "time_per_iteration": 3.448108196258545
    },
    {
      "auxiliary_loss_clip": 0.01158801,
      "auxiliary_loss_mlp": 0.01042222,
      "balance_loss_clip": 1.05436742,
      "balance_loss_mlp": 1.03102195,
      "epoch": 0.07106354836770276,
      "flos": 24645098534400.0,
      "grad_norm": 1.8030931528634409,
      "language_loss": 0.8456409,
      "learning_rate": 3.982386179311399e-06,
      "loss": 0.86765116,
      "num_input_tokens_seen": 12621725,
      "step": 591,
      "time_per_iteration": 2.7079856395721436
    },
    {
      "auxiliary_loss_clip": 0.01255686,
      "auxiliary_loss_mlp": 0.01050817,
      "balance_loss_clip": 1.0767467,
      "balance_loss_mlp": 1.03776991,
      "epoch": 0.07118379125834184,
      "flos": 16217223649920.0,
      "grad_norm": 35.92314251916883,
      "language_loss": 0.87409043,
      "learning_rate": 3.982282873988075e-06,
      "loss": 0.8971554,
      "num_input_tokens_seen": 12639600,
      "step": 592,
      "time_per_iteration": 3.286536693572998
    },
    {
      "auxiliary_loss_clip": 0.01235046,
      "auxiliary_loss_mlp": 0.01047763,
      "balance_loss_clip": 1.07458925,
      "balance_loss_mlp": 1.03703368,
      "epoch": 0.07130403414898094,
      "flos": 19719986227200.0,
      "grad_norm": 1.643829576791502,
      "language_loss": 0.86781389,
      "learning_rate": 3.982179267954016e-06,
      "loss": 0.89064199,
      "num_input_tokens_seen": 12660030,
      "step": 593,
      "time_per_iteration": 2.6109697818756104
    },
    {
      "auxiliary_loss_clip": 0.01266146,
      "auxiliary_loss_mlp": 0.01039401,
      "balance_loss_clip": 1.07574534,
      "balance_loss_mlp": 1.02668691,
      "epoch": 0.07142427703962004,
      "flos": 21871933009920.0,
      "grad_norm": 2.299075894595495,
      "language_loss": 0.95922935,
      "learning_rate": 3.982075361224937e-06,
      "loss": 0.98228478,
      "num_input_tokens_seen": 12678395,
      "step": 594,
      "time_per_iteration": 3.2897706031799316
    },
    {
      "auxiliary_loss_clip": 0.01246527,
      "auxiliary_loss_mlp": 0.00764198,
      "balance_loss_clip": 1.07465398,
      "balance_loss_mlp": 1.0001483,
      "epoch": 0.07154451993025912,
      "flos": 18296595002880.0,
      "grad_norm": 1.7191436384403052,
      "language_loss": 0.87933683,
      "learning_rate": 3.981971153816602e-06,
      "loss": 0.8994441,
      "num_input_tokens_seen": 12696000,
      "step": 595,
      "time_per_iteration": 2.5463991165161133
    },
    {
      "auxiliary_loss_clip": 0.01269739,
      "auxiliary_loss_mlp": 0.01041305,
      "balance_loss_clip": 1.08211076,
      "balance_loss_mlp": 1.03065419,
      "epoch": 0.07166476282089822,
      "flos": 22160690444160.0,
      "grad_norm": 1.7731046260872771,
      "language_loss": 0.9622277,
      "learning_rate": 3.981866645744819e-06,
      "loss": 0.98533809,
      "num_input_tokens_seen": 12716715,
      "step": 596,
      "time_per_iteration": 3.378173351287842
    },
    {
      "auxiliary_loss_clip": 0.01270366,
      "auxiliary_loss_mlp": 0.00764827,
      "balance_loss_clip": 1.07944894,
      "balance_loss_mlp": 1.0001328,
      "epoch": 0.0717850057115373,
      "flos": 14136343925760.0,
      "grad_norm": 2.2172828328293384,
      "language_loss": 0.81395912,
      "learning_rate": 3.9817618370254416e-06,
      "loss": 0.83431107,
      "num_input_tokens_seen": 12733370,
      "step": 597,
      "time_per_iteration": 2.5280351638793945
    },
    {
      "auxiliary_loss_clip": 0.01267403,
      "auxiliary_loss_mlp": 0.01051578,
      "balance_loss_clip": 1.07789242,
      "balance_loss_mlp": 1.03990781,
      "epoch": 0.0719052486021764,
      "flos": 30917794412160.0,
      "grad_norm": 2.4443547115725464,
      "language_loss": 0.87597692,
      "learning_rate": 3.9816567276743684e-06,
      "loss": 0.8991667,
      "num_input_tokens_seen": 12753235,
      "step": 598,
      "time_per_iteration": 2.6008851528167725
    },
    {
      "auxiliary_loss_clip": 0.0123323,
      "auxiliary_loss_mlp": 0.01046572,
      "balance_loss_clip": 1.07458985,
      "balance_loss_mlp": 1.03496051,
      "epoch": 0.0720254914928155,
      "flos": 21287019939840.0,
      "grad_norm": 1.9011773027737742,
      "language_loss": 0.77859026,
      "learning_rate": 3.9815513177075466e-06,
      "loss": 0.80138826,
      "num_input_tokens_seen": 12772020,
      "step": 599,
      "time_per_iteration": 2.5959866046905518
    },
    {
      "auxiliary_loss_clip": 0.0124648,
      "auxiliary_loss_mlp": 0.01051306,
      "balance_loss_clip": 1.07554352,
      "balance_loss_mlp": 1.04066658,
      "epoch": 0.07214573438345458,
      "flos": 27819170732160.0,
      "grad_norm": 1.9484513951244535,
      "language_loss": 0.70091581,
      "learning_rate": 3.9814456071409646e-06,
      "loss": 0.72389364,
      "num_input_tokens_seen": 12792555,
      "step": 600,
      "time_per_iteration": 2.6005210876464844
    },
    {
      "auxiliary_loss_clip": 0.01205268,
      "auxiliary_loss_mlp": 0.01043478,
      "balance_loss_clip": 1.06790924,
      "balance_loss_mlp": 1.03014469,
      "epoch": 0.07226597727409367,
      "flos": 25483576688640.0,
      "grad_norm": 4.6983648096680986,
      "language_loss": 0.85177201,
      "learning_rate": 3.981339595990659e-06,
      "loss": 0.87425947,
      "num_input_tokens_seen": 12811085,
      "step": 601,
      "time_per_iteration": 2.6676185131073
    },
    {
      "auxiliary_loss_clip": 0.01248107,
      "auxiliary_loss_mlp": 0.01038942,
      "balance_loss_clip": 1.07472694,
      "balance_loss_mlp": 1.02660966,
      "epoch": 0.07238622016473276,
      "flos": 23513840622720.0,
      "grad_norm": 2.1407915145963456,
      "language_loss": 0.80976975,
      "learning_rate": 3.981233284272713e-06,
      "loss": 0.83264029,
      "num_input_tokens_seen": 12830830,
      "step": 602,
      "time_per_iteration": 2.5622141361236572
    },
    {
      "auxiliary_loss_clip": 0.01215528,
      "auxiliary_loss_mlp": 0.01060634,
      "balance_loss_clip": 1.06933403,
      "balance_loss_mlp": 1.04940403,
      "epoch": 0.07250646305537185,
      "flos": 25453519983360.0,
      "grad_norm": 1.5844699914677463,
      "language_loss": 0.90049022,
      "learning_rate": 3.981126672003253e-06,
      "loss": 0.92325181,
      "num_input_tokens_seen": 12853505,
      "step": 603,
      "time_per_iteration": 2.7046027183532715
    },
    {
      "auxiliary_loss_clip": 0.01233605,
      "auxiliary_loss_mlp": 0.01045759,
      "balance_loss_clip": 1.06696832,
      "balance_loss_mlp": 1.03337932,
      "epoch": 0.07262670594601094,
      "flos": 27155038216320.0,
      "grad_norm": 3.3874725745970116,
      "language_loss": 0.78695571,
      "learning_rate": 3.981019759198451e-06,
      "loss": 0.80974936,
      "num_input_tokens_seen": 12872455,
      "step": 604,
      "time_per_iteration": 2.620738983154297
    },
    {
      "auxiliary_loss_clip": 0.01238267,
      "auxiliary_loss_mlp": 0.01039353,
      "balance_loss_clip": 1.07208848,
      "balance_loss_mlp": 1.02843976,
      "epoch": 0.07274694883665003,
      "flos": 26651607148800.0,
      "grad_norm": 2.8468756392004404,
      "language_loss": 0.84112298,
      "learning_rate": 3.980912545874528e-06,
      "loss": 0.86389929,
      "num_input_tokens_seen": 12892620,
      "step": 605,
      "time_per_iteration": 2.610797166824341
    },
    {
      "auxiliary_loss_clip": 0.01240819,
      "auxiliary_loss_mlp": 0.00764837,
      "balance_loss_clip": 1.07046115,
      "balance_loss_mlp": 1.00014758,
      "epoch": 0.07286719172728913,
      "flos": 29862344154240.0,
      "grad_norm": 1.8721481056543563,
      "language_loss": 0.85317045,
      "learning_rate": 3.980805032047746e-06,
      "loss": 0.87322706,
      "num_input_tokens_seen": 12914090,
      "step": 606,
      "time_per_iteration": 2.6141562461853027
    },
    {
      "auxiliary_loss_clip": 0.01233356,
      "auxiliary_loss_mlp": 0.01041257,
      "balance_loss_clip": 1.07255697,
      "balance_loss_mlp": 1.02853107,
      "epoch": 0.07298743461792821,
      "flos": 17382057799680.0,
      "grad_norm": 1.977607385466361,
      "language_loss": 0.80752528,
      "learning_rate": 3.980697217734415e-06,
      "loss": 0.83027136,
      "num_input_tokens_seen": 12931830,
      "step": 607,
      "time_per_iteration": 2.549487590789795
    },
    {
      "auxiliary_loss_clip": 0.0121091,
      "auxiliary_loss_mlp": 0.00764146,
      "balance_loss_clip": 1.07085013,
      "balance_loss_mlp": 1.00014973,
      "epoch": 0.07310767750856731,
      "flos": 19498201701120.0,
      "grad_norm": 1.7908156720305866,
      "language_loss": 0.91592026,
      "learning_rate": 3.980589102950891e-06,
      "loss": 0.93567079,
      "num_input_tokens_seen": 12949995,
      "step": 608,
      "time_per_iteration": 2.640376567840576
    },
    {
      "auxiliary_loss_clip": 0.01232004,
      "auxiliary_loss_mlp": 0.01041755,
      "balance_loss_clip": 1.07379341,
      "balance_loss_mlp": 1.02962565,
      "epoch": 0.07322792039920639,
      "flos": 29168693637120.0,
      "grad_norm": 2.6317970254696452,
      "language_loss": 0.76063287,
      "learning_rate": 3.9804806877135755e-06,
      "loss": 0.78337044,
      "num_input_tokens_seen": 12968040,
      "step": 609,
      "time_per_iteration": 2.6500983238220215
    },
    {
      "auxiliary_loss_clip": 0.0125121,
      "auxiliary_loss_mlp": 0.00764646,
      "balance_loss_clip": 1.07265425,
      "balance_loss_mlp": 1.00012732,
      "epoch": 0.07334816328984549,
      "flos": 23477822259840.0,
      "grad_norm": 2.021657155323173,
      "language_loss": 0.86146146,
      "learning_rate": 3.980371972038915e-06,
      "loss": 0.88162005,
      "num_input_tokens_seen": 12988530,
      "step": 610,
      "time_per_iteration": 2.6194887161254883
    },
    {
      "auxiliary_loss_clip": 0.01270407,
      "auxiliary_loss_mlp": 0.01042944,
      "balance_loss_clip": 1.08243108,
      "balance_loss_mlp": 1.03104722,
      "epoch": 0.07346840618048459,
      "flos": 22962467877120.0,
      "grad_norm": 1.8635347826838835,
      "language_loss": 0.84682035,
      "learning_rate": 3.980262955943399e-06,
      "loss": 0.86995387,
      "num_input_tokens_seen": 13008195,
      "step": 611,
      "time_per_iteration": 2.5182180404663086
    },
    {
      "auxiliary_loss_clip": 0.01226572,
      "auxiliary_loss_mlp": 0.01040451,
      "balance_loss_clip": 1.073084,
      "balance_loss_mlp": 1.02895904,
      "epoch": 0.07358864907112367,
      "flos": 17673903803520.0,
      "grad_norm": 2.441455376178146,
      "language_loss": 0.86436659,
      "learning_rate": 3.980153639443569e-06,
      "loss": 0.8870368,
      "num_input_tokens_seen": 13024180,
      "step": 612,
      "time_per_iteration": 2.5855839252471924
    },
    {
      "auxiliary_loss_clip": 0.0124472,
      "auxiliary_loss_mlp": 0.01037037,
      "balance_loss_clip": 1.07335734,
      "balance_loss_mlp": 1.0258671,
      "epoch": 0.07370889196176277,
      "flos": 24097029840000.0,
      "grad_norm": 2.131494234672017,
      "language_loss": 0.79873705,
      "learning_rate": 3.980044022556005e-06,
      "loss": 0.82155466,
      "num_input_tokens_seen": 13043865,
      "step": 613,
      "time_per_iteration": 2.6133341789245605
    },
    {
      "auxiliary_loss_clip": 0.01250266,
      "auxiliary_loss_mlp": 0.01045482,
      "balance_loss_clip": 1.07451773,
      "balance_loss_mlp": 1.03346539,
      "epoch": 0.07382913485240185,
      "flos": 25885919905920.0,
      "grad_norm": 2.1177768348798733,
      "language_loss": 0.72913438,
      "learning_rate": 3.9799341052973375e-06,
      "loss": 0.75209183,
      "num_input_tokens_seen": 13063700,
      "step": 614,
      "time_per_iteration": 2.6194546222686768
    },
    {
      "auxiliary_loss_clip": 0.01237858,
      "auxiliary_loss_mlp": 0.0105344,
      "balance_loss_clip": 1.07642102,
      "balance_loss_mlp": 1.04119086,
      "epoch": 0.07394937774304094,
      "flos": 16873850223360.0,
      "grad_norm": 2.77562861425666,
      "language_loss": 0.75261366,
      "learning_rate": 3.979823887684241e-06,
      "loss": 0.77552664,
      "num_input_tokens_seen": 13082640,
      "step": 615,
      "time_per_iteration": 2.5674898624420166
    },
    {
      "auxiliary_loss_clip": 0.01269917,
      "auxiliary_loss_mlp": 0.01047473,
      "balance_loss_clip": 1.07961082,
      "balance_loss_mlp": 1.03577816,
      "epoch": 0.07406962063368003,
      "flos": 20703471586560.0,
      "grad_norm": 2.915259375920163,
      "language_loss": 0.84845644,
      "learning_rate": 3.979713369733434e-06,
      "loss": 0.87163031,
      "num_input_tokens_seen": 13100505,
      "step": 616,
      "time_per_iteration": 3.3737385272979736
    },
    {
      "auxiliary_loss_clip": 0.01245038,
      "auxiliary_loss_mlp": 0.01040357,
      "balance_loss_clip": 1.07423544,
      "balance_loss_mlp": 1.02841854,
      "epoch": 0.07418986352431912,
      "flos": 21430985650560.0,
      "grad_norm": 1.948726568914937,
      "language_loss": 0.84904587,
      "learning_rate": 3.979602551461683e-06,
      "loss": 0.87189984,
      "num_input_tokens_seen": 13121285,
      "step": 617,
      "time_per_iteration": 2.574411630630493
    },
    {
      "auxiliary_loss_clip": 0.01235467,
      "auxiliary_loss_mlp": 0.01043067,
      "balance_loss_clip": 1.07641602,
      "balance_loss_mlp": 1.03108668,
      "epoch": 0.07431010641495822,
      "flos": 12021133777920.0,
      "grad_norm": 2.406596779872574,
      "language_loss": 0.91457409,
      "learning_rate": 3.979491432885799e-06,
      "loss": 0.93735945,
      "num_input_tokens_seen": 13137550,
      "step": 618,
      "time_per_iteration": 3.3724093437194824
    },
    {
      "auxiliary_loss_clip": 0.01197828,
      "auxiliary_loss_mlp": 0.00763843,
      "balance_loss_clip": 1.06642413,
      "balance_loss_mlp": 1.00006378,
      "epoch": 0.0744303493055973,
      "flos": 20957575374720.0,
      "grad_norm": 1.943455562548442,
      "language_loss": 0.82691789,
      "learning_rate": 3.97938001402264e-06,
      "loss": 0.84653455,
      "num_input_tokens_seen": 13156675,
      "step": 619,
      "time_per_iteration": 2.6533150672912598
    },
    {
      "auxiliary_loss_clip": 0.01212155,
      "auxiliary_loss_mlp": 0.0104591,
      "balance_loss_clip": 1.06935608,
      "balance_loss_mlp": 1.03367352,
      "epoch": 0.0745505921962364,
      "flos": 16253134272000.0,
      "grad_norm": 2.6310430443024995,
      "language_loss": 0.79943204,
      "learning_rate": 3.979268294889105e-06,
      "loss": 0.82201272,
      "num_input_tokens_seen": 13172225,
      "step": 620,
      "time_per_iteration": 3.3003945350646973
    },
    {
      "auxiliary_loss_clip": 0.0126786,
      "auxiliary_loss_mlp": 0.01044713,
      "balance_loss_clip": 1.07977343,
      "balance_loss_mlp": 1.03311968,
      "epoch": 0.07467083508687548,
      "flos": 50944635550080.0,
      "grad_norm": 1.6832701069790912,
      "language_loss": 0.7394312,
      "learning_rate": 3.979156275502143e-06,
      "loss": 0.76255691,
      "num_input_tokens_seen": 13195885,
      "step": 621,
      "time_per_iteration": 2.7930386066436768
    },
    {
      "auxiliary_loss_clip": 0.01217313,
      "auxiliary_loss_mlp": 0.01045321,
      "balance_loss_clip": 1.06979895,
      "balance_loss_mlp": 1.03422308,
      "epoch": 0.07479107797751458,
      "flos": 17529686697600.0,
      "grad_norm": 2.192886830731349,
      "language_loss": 0.91460049,
      "learning_rate": 3.979043955878749e-06,
      "loss": 0.93722683,
      "num_input_tokens_seen": 13213730,
      "step": 622,
      "time_per_iteration": 3.4086458683013916
    },
    {
      "auxiliary_loss_clip": 0.0123177,
      "auxiliary_loss_mlp": 0.01043838,
      "balance_loss_clip": 1.07413077,
      "balance_loss_mlp": 1.03272164,
      "epoch": 0.07491132086815366,
      "flos": 23473943591040.0,
      "grad_norm": 2.0622493458565496,
      "language_loss": 0.83127832,
      "learning_rate": 3.978931336035959e-06,
      "loss": 0.85403436,
      "num_input_tokens_seen": 13232540,
      "step": 623,
      "time_per_iteration": 2.61781644821167
    },
    {
      "auxiliary_loss_clip": 0.01250511,
      "auxiliary_loss_mlp": 0.01052628,
      "balance_loss_clip": 1.07894623,
      "balance_loss_mlp": 1.04073715,
      "epoch": 0.07503156375879276,
      "flos": 20157557708160.0,
      "grad_norm": 2.438260168862922,
      "language_loss": 0.8242076,
      "learning_rate": 3.9788184159908595e-06,
      "loss": 0.84723902,
      "num_input_tokens_seen": 13249670,
      "step": 624,
      "time_per_iteration": 2.5374293327331543
    },
    {
      "auxiliary_loss_clip": 0.01228742,
      "auxiliary_loss_mlp": 0.01047767,
      "balance_loss_clip": 1.07099473,
      "balance_loss_mlp": 1.03744316,
      "epoch": 0.07515180664943186,
      "flos": 15115519653120.0,
      "grad_norm": 9.513964664246583,
      "language_loss": 0.82909536,
      "learning_rate": 3.97870519576058e-06,
      "loss": 0.8518604,
      "num_input_tokens_seen": 13266095,
      "step": 625,
      "time_per_iteration": 2.588409900665283
    },
    {
      "auxiliary_loss_clip": 0.0121461,
      "auxiliary_loss_mlp": 0.00764423,
      "balance_loss_clip": 1.06947434,
      "balance_loss_mlp": 1.00010991,
      "epoch": 0.07527204954007094,
      "flos": 21287702298240.0,
      "grad_norm": 2.4330532550688275,
      "language_loss": 0.81452489,
      "learning_rate": 3.978591675362295e-06,
      "loss": 0.83431524,
      "num_input_tokens_seen": 13284810,
      "step": 626,
      "time_per_iteration": 2.642906665802002
    },
    {
      "auxiliary_loss_clip": 0.01199338,
      "auxiliary_loss_mlp": 0.01044088,
      "balance_loss_clip": 1.07173538,
      "balance_loss_mlp": 1.03231573,
      "epoch": 0.07539229243071004,
      "flos": 21324187537920.0,
      "grad_norm": 1.8016025446176207,
      "language_loss": 0.87751627,
      "learning_rate": 3.978477854813226e-06,
      "loss": 0.8999505,
      "num_input_tokens_seen": 13304150,
      "step": 627,
      "time_per_iteration": 2.636841297149658
    },
    {
      "auxiliary_loss_clip": 0.01247553,
      "auxiliary_loss_mlp": 0.01044285,
      "balance_loss_clip": 1.07349169,
      "balance_loss_mlp": 1.03363967,
      "epoch": 0.07551253532134912,
      "flos": 13042540920960.0,
      "grad_norm": 1.9851511149859102,
      "language_loss": 0.82149374,
      "learning_rate": 3.97836373413064e-06,
      "loss": 0.84441221,
      "num_input_tokens_seen": 13322205,
      "step": 628,
      "time_per_iteration": 2.582611560821533
    },
    {
      "auxiliary_loss_clip": 0.01265033,
      "auxiliary_loss_mlp": 0.01048728,
      "balance_loss_clip": 1.07529688,
      "balance_loss_mlp": 1.0369916,
      "epoch": 0.07563277821198822,
      "flos": 19208761908480.0,
      "grad_norm": 1.948528439334513,
      "language_loss": 0.7498638,
      "learning_rate": 3.978249313331848e-06,
      "loss": 0.77300143,
      "num_input_tokens_seen": 13340435,
      "step": 629,
      "time_per_iteration": 2.5087220668792725
    },
    {
      "auxiliary_loss_clip": 0.01253324,
      "auxiliary_loss_mlp": 0.00764256,
      "balance_loss_clip": 1.07240677,
      "balance_loss_mlp": 1.0000813,
      "epoch": 0.07575302110262731,
      "flos": 19537200892800.0,
      "grad_norm": 5.788076286824976,
      "language_loss": 0.62327188,
      "learning_rate": 3.978134592434208e-06,
      "loss": 0.6434477,
      "num_input_tokens_seen": 13358185,
      "step": 630,
      "time_per_iteration": 2.549100875854492
    },
    {
      "auxiliary_loss_clip": 0.01098796,
      "auxiliary_loss_mlp": 0.01009167,
      "balance_loss_clip": 1.04131293,
      "balance_loss_mlp": 1.00509024,
      "epoch": 0.0758732639932664,
      "flos": 67961808017280.0,
      "grad_norm": 1.0659664905712698,
      "language_loss": 0.59384716,
      "learning_rate": 3.978019571455123e-06,
      "loss": 0.61492682,
      "num_input_tokens_seen": 13410130,
      "step": 631,
      "time_per_iteration": 3.2172605991363525
    },
    {
      "auxiliary_loss_clip": 0.01263272,
      "auxiliary_loss_mlp": 0.01041318,
      "balance_loss_clip": 1.07766306,
      "balance_loss_mlp": 1.03041601,
      "epoch": 0.07599350688390549,
      "flos": 18989204025600.0,
      "grad_norm": 2.534020719606259,
      "language_loss": 0.83666533,
      "learning_rate": 3.977904250412042e-06,
      "loss": 0.85971129,
      "num_input_tokens_seen": 13429085,
      "step": 632,
      "time_per_iteration": 2.521507978439331
    },
    {
      "auxiliary_loss_clip": 0.0123606,
      "auxiliary_loss_mlp": 0.01047614,
      "balance_loss_clip": 1.07108498,
      "balance_loss_mlp": 1.03695107,
      "epoch": 0.07611374977454458,
      "flos": 21069006341760.0,
      "grad_norm": 2.0673782708218473,
      "language_loss": 0.85584974,
      "learning_rate": 3.97778862932246e-06,
      "loss": 0.87868649,
      "num_input_tokens_seen": 13446250,
      "step": 633,
      "time_per_iteration": 2.5648880004882812
    },
    {
      "auxiliary_loss_clip": 0.01129217,
      "auxiliary_loss_mlp": 0.01042748,
      "balance_loss_clip": 1.0504632,
      "balance_loss_mlp": 1.03126812,
      "epoch": 0.07623399266518367,
      "flos": 18514536773760.0,
      "grad_norm": 2.1042966245299923,
      "language_loss": 0.94337523,
      "learning_rate": 3.9776727082039144e-06,
      "loss": 0.96509486,
      "num_input_tokens_seen": 13463220,
      "step": 634,
      "time_per_iteration": 3.1057932376861572
    },
    {
      "auxiliary_loss_clip": 0.01157881,
      "auxiliary_loss_mlp": 0.01003554,
      "balance_loss_clip": 1.0426898,
      "balance_loss_mlp": 0.99954891,
      "epoch": 0.07635423555582276,
      "flos": 44663036077440.0,
      "grad_norm": 0.8047699436480924,
      "language_loss": 0.55511081,
      "learning_rate": 3.977556487073991e-06,
      "loss": 0.57672513,
      "num_input_tokens_seen": 13517775,
      "step": 635,
      "time_per_iteration": 3.328406572341919
    },
    {
      "auxiliary_loss_clip": 0.01220475,
      "auxiliary_loss_mlp": 0.01043024,
      "balance_loss_clip": 1.06516349,
      "balance_loss_mlp": 1.03237891,
      "epoch": 0.07647447844646185,
      "flos": 21761148487680.0,
      "grad_norm": 1.6419205092186242,
      "language_loss": 0.81547046,
      "learning_rate": 3.97743996595032e-06,
      "loss": 0.83810544,
      "num_input_tokens_seen": 13537815,
      "step": 636,
      "time_per_iteration": 2.6144723892211914
    },
    {
      "auxiliary_loss_clip": 0.01265052,
      "auxiliary_loss_mlp": 0.0104693,
      "balance_loss_clip": 1.07586026,
      "balance_loss_mlp": 1.03475857,
      "epoch": 0.07659472133710095,
      "flos": 23806799948160.0,
      "grad_norm": 1.5147758449836068,
      "language_loss": 0.81736451,
      "learning_rate": 3.9773231448505804e-06,
      "loss": 0.84048432,
      "num_input_tokens_seen": 13559605,
      "step": 637,
      "time_per_iteration": 2.5609283447265625
    },
    {
      "auxiliary_loss_clip": 0.01230244,
      "auxiliary_loss_mlp": 0.00764692,
      "balance_loss_clip": 1.07415557,
      "balance_loss_mlp": 1.00004554,
      "epoch": 0.07671496422774003,
      "flos": 21469984842240.0,
      "grad_norm": 10.763873394752201,
      "language_loss": 0.78216302,
      "learning_rate": 3.977206023792491e-06,
      "loss": 0.80211234,
      "num_input_tokens_seen": 13579495,
      "step": 638,
      "time_per_iteration": 2.566572427749634
    },
    {
      "auxiliary_loss_clip": 0.01248069,
      "auxiliary_loss_mlp": 0.01046168,
      "balance_loss_clip": 1.0761708,
      "balance_loss_mlp": 1.0354569,
      "epoch": 0.07683520711837913,
      "flos": 16980971558400.0,
      "grad_norm": 2.0329079762414906,
      "language_loss": 0.81166631,
      "learning_rate": 3.97708860279382e-06,
      "loss": 0.83460867,
      "num_input_tokens_seen": 13597605,
      "step": 639,
      "time_per_iteration": 2.5181567668914795
    },
    {
      "auxiliary_loss_clip": 0.01210637,
      "auxiliary_loss_mlp": 0.01047851,
      "balance_loss_clip": 1.06811142,
      "balance_loss_mlp": 1.03683031,
      "epoch": 0.07695545000901821,
      "flos": 23476744851840.0,
      "grad_norm": 2.607175778071035,
      "language_loss": 0.78069383,
      "learning_rate": 3.97697088187238e-06,
      "loss": 0.80327868,
      "num_input_tokens_seen": 13618120,
      "step": 640,
      "time_per_iteration": 2.6024010181427
    },
    {
      "auxiliary_loss_clip": 0.01227081,
      "auxiliary_loss_mlp": 0.0104762,
      "balance_loss_clip": 1.07290816,
      "balance_loss_mlp": 1.03699255,
      "epoch": 0.07707569289965731,
      "flos": 17634258167040.0,
      "grad_norm": 2.119438085117736,
      "language_loss": 0.91744316,
      "learning_rate": 3.976852861046029e-06,
      "loss": 0.9401902,
      "num_input_tokens_seen": 13634735,
      "step": 641,
      "time_per_iteration": 2.5320682525634766
    },
    {
      "auxiliary_loss_clip": 0.01195335,
      "auxiliary_loss_mlp": 0.01044347,
      "balance_loss_clip": 1.06777835,
      "balance_loss_mlp": 1.03368938,
      "epoch": 0.0771959357902964,
      "flos": 25775674087680.0,
      "grad_norm": 1.7410008581829073,
      "language_loss": 0.80302382,
      "learning_rate": 3.97673454033267e-06,
      "loss": 0.82542062,
      "num_input_tokens_seen": 13656835,
      "step": 642,
      "time_per_iteration": 3.5156381130218506
    },
    {
      "auxiliary_loss_clip": 0.01232599,
      "auxiliary_loss_mlp": 0.01035399,
      "balance_loss_clip": 1.06953156,
      "balance_loss_mlp": 1.02456307,
      "epoch": 0.07731617868093549,
      "flos": 19828651847040.0,
      "grad_norm": 1.9749917261751988,
      "language_loss": 0.82543135,
      "learning_rate": 3.976615919750254e-06,
      "loss": 0.84811139,
      "num_input_tokens_seen": 13674535,
      "step": 643,
      "time_per_iteration": 2.5624449253082275
    },
    {
      "auxiliary_loss_clip": 0.01245732,
      "auxiliary_loss_mlp": 0.01043627,
      "balance_loss_clip": 1.07404852,
      "balance_loss_mlp": 1.03307092,
      "epoch": 0.07743642157157458,
      "flos": 21324654414720.0,
      "grad_norm": 2.0249986741633776,
      "language_loss": 0.86907494,
      "learning_rate": 3.976496999316775e-06,
      "loss": 0.89196849,
      "num_input_tokens_seen": 13693290,
      "step": 644,
      "time_per_iteration": 3.3537442684173584
    },
    {
      "auxiliary_loss_clip": 0.01228062,
      "auxiliary_loss_mlp": 0.01038452,
      "balance_loss_clip": 1.07467937,
      "balance_loss_mlp": 1.02707911,
      "epoch": 0.07755666446221367,
      "flos": 19969133938560.0,
      "grad_norm": 1.9735132389141057,
      "language_loss": 0.84280801,
      "learning_rate": 3.976377779050271e-06,
      "loss": 0.86547315,
      "num_input_tokens_seen": 13711420,
      "step": 645,
      "time_per_iteration": 2.575831413269043
    },
    {
      "auxiliary_loss_clip": 0.01234936,
      "auxiliary_loss_mlp": 0.01059784,
      "balance_loss_clip": 1.07109261,
      "balance_loss_mlp": 1.04786301,
      "epoch": 0.07767690735285276,
      "flos": 23623224514560.0,
      "grad_norm": 2.118699492365852,
      "language_loss": 0.84234107,
      "learning_rate": 3.976258258968831e-06,
      "loss": 0.86528826,
      "num_input_tokens_seen": 13729965,
      "step": 646,
      "time_per_iteration": 3.278416633605957
    },
    {
      "auxiliary_loss_clip": 0.01217054,
      "auxiliary_loss_mlp": 0.01051357,
      "balance_loss_clip": 1.07268691,
      "balance_loss_mlp": 1.04037189,
      "epoch": 0.07779715024349185,
      "flos": 22236246702720.0,
      "grad_norm": 2.0777545475289174,
      "language_loss": 0.74172831,
      "learning_rate": 3.976138439090583e-06,
      "loss": 0.76441234,
      "num_input_tokens_seen": 13748045,
      "step": 647,
      "time_per_iteration": 3.4510860443115234
    },
    {
      "auxiliary_loss_clip": 0.01219911,
      "auxiliary_loss_mlp": 0.01037999,
      "balance_loss_clip": 1.07361889,
      "balance_loss_mlp": 1.02625692,
      "epoch": 0.07791739313413094,
      "flos": 20955097336320.0,
      "grad_norm": 2.3559524373423875,
      "language_loss": 0.85221779,
      "learning_rate": 3.976018319433706e-06,
      "loss": 0.87479687,
      "num_input_tokens_seen": 13765590,
      "step": 648,
      "time_per_iteration": 2.6218042373657227
    },
    {
      "auxiliary_loss_clip": 0.0124986,
      "auxiliary_loss_mlp": 0.0104822,
      "balance_loss_clip": 1.07748771,
      "balance_loss_mlp": 1.03698444,
      "epoch": 0.07803763602477004,
      "flos": 19312327797120.0,
      "grad_norm": 2.5852670402787585,
      "language_loss": 0.91375577,
      "learning_rate": 3.9758979000164205e-06,
      "loss": 0.93673652,
      "num_input_tokens_seen": 13782410,
      "step": 649,
      "time_per_iteration": 2.5280842781066895
    },
    {
      "auxiliary_loss_clip": 0.01216573,
      "auxiliary_loss_mlp": 0.01047688,
      "balance_loss_clip": 1.06968379,
      "balance_loss_mlp": 1.03570163,
      "epoch": 0.07815787891540912,
      "flos": 22710806213760.0,
      "grad_norm": 2.250179591520519,
      "language_loss": 0.72282112,
      "learning_rate": 3.975777180856995e-06,
      "loss": 0.74546373,
      "num_input_tokens_seen": 13801530,
      "step": 650,
      "time_per_iteration": 2.742828130722046
    },
    {
      "auxiliary_loss_clip": 0.012623,
      "auxiliary_loss_mlp": 0.01052915,
      "balance_loss_clip": 1.07408202,
      "balance_loss_mlp": 1.04030252,
      "epoch": 0.07827812180604822,
      "flos": 22711129436160.0,
      "grad_norm": 2.183035153385274,
      "language_loss": 0.86146283,
      "learning_rate": 3.975656161973742e-06,
      "loss": 0.88461494,
      "num_input_tokens_seen": 13820615,
      "step": 651,
      "time_per_iteration": 2.5272269248962402
    },
    {
      "auxiliary_loss_clip": 0.01264425,
      "auxiliary_loss_mlp": 0.0103884,
      "balance_loss_clip": 1.07566297,
      "balance_loss_mlp": 1.02681732,
      "epoch": 0.0783983646966873,
      "flos": 21725597001600.0,
      "grad_norm": 2.484751720943226,
      "language_loss": 0.88969177,
      "learning_rate": 3.9755348433850194e-06,
      "loss": 0.91272449,
      "num_input_tokens_seen": 13835955,
      "step": 652,
      "time_per_iteration": 2.51649808883667
    },
    {
      "auxiliary_loss_clip": 0.01118763,
      "auxiliary_loss_mlp": 0.01028048,
      "balance_loss_clip": 1.03256679,
      "balance_loss_mlp": 1.02432883,
      "epoch": 0.0785186075873264,
      "flos": 60640877537280.0,
      "grad_norm": 0.9816403410167297,
      "language_loss": 0.63708246,
      "learning_rate": 3.975413225109232e-06,
      "loss": 0.65855062,
      "num_input_tokens_seen": 13896505,
      "step": 653,
      "time_per_iteration": 3.2101926803588867
    },
    {
      "auxiliary_loss_clip": 0.01247308,
      "auxiliary_loss_mlp": 0.01046358,
      "balance_loss_clip": 1.07406116,
      "balance_loss_mlp": 1.03492594,
      "epoch": 0.0786388504779655,
      "flos": 23877902920320.0,
      "grad_norm": 5.396901363328418,
      "language_loss": 0.93887901,
      "learning_rate": 3.975291307164829e-06,
      "loss": 0.96181571,
      "num_input_tokens_seen": 13915150,
      "step": 654,
      "time_per_iteration": 2.549163818359375
    },
    {
      "auxiliary_loss_clip": 0.01209529,
      "auxiliary_loss_mlp": 0.01041569,
      "balance_loss_clip": 1.06732082,
      "balance_loss_mlp": 1.03109622,
      "epoch": 0.07875909336860458,
      "flos": 15158684822400.0,
      "grad_norm": 2.5429683545436044,
      "language_loss": 0.84950566,
      "learning_rate": 3.975169089570306e-06,
      "loss": 0.87201667,
      "num_input_tokens_seen": 13933525,
      "step": 655,
      "time_per_iteration": 2.5864341259002686
    },
    {
      "auxiliary_loss_clip": 0.01229129,
      "auxiliary_loss_mlp": 0.01054227,
      "balance_loss_clip": 1.06942475,
      "balance_loss_mlp": 1.0433197,
      "epoch": 0.07887933625924368,
      "flos": 22236857233920.0,
      "grad_norm": 1.9945082297658308,
      "language_loss": 0.91684657,
      "learning_rate": 3.975046572344202e-06,
      "loss": 0.9396801,
      "num_input_tokens_seen": 13949985,
      "step": 656,
      "time_per_iteration": 2.5253217220306396
    },
    {
      "auxiliary_loss_clip": 0.0120916,
      "auxiliary_loss_mlp": 0.01043478,
      "balance_loss_clip": 1.06338,
      "balance_loss_mlp": 1.03276706,
      "epoch": 0.07899957914988276,
      "flos": 20777734955520.0,
      "grad_norm": 1.840019596077486,
      "language_loss": 0.71130049,
      "learning_rate": 3.974923755505103e-06,
      "loss": 0.73382694,
      "num_input_tokens_seen": 13969215,
      "step": 657,
      "time_per_iteration": 2.610306978225708
    },
    {
      "auxiliary_loss_clip": 0.01202773,
      "auxiliary_loss_mlp": 0.01045558,
      "balance_loss_clip": 1.06643629,
      "balance_loss_mlp": 1.03466225,
      "epoch": 0.07911982204052186,
      "flos": 23003047267200.0,
      "grad_norm": 1.6410328173398758,
      "language_loss": 0.9120667,
      "learning_rate": 3.974800639071641e-06,
      "loss": 0.93455005,
      "num_input_tokens_seen": 13989935,
      "step": 658,
      "time_per_iteration": 2.6287972927093506
    },
    {
      "auxiliary_loss_clip": 0.01166353,
      "auxiliary_loss_mlp": 0.00764418,
      "balance_loss_clip": 1.06033051,
      "balance_loss_mlp": 1.0000422,
      "epoch": 0.07924006493116094,
      "flos": 23111389664640.0,
      "grad_norm": 2.3022066213681387,
      "language_loss": 1.00449967,
      "learning_rate": 3.974677223062492e-06,
      "loss": 1.02380741,
      "num_input_tokens_seen": 14007150,
      "step": 659,
      "time_per_iteration": 2.663703203201294
    },
    {
      "auxiliary_loss_clip": 0.01229152,
      "auxiliary_loss_mlp": 0.01047884,
      "balance_loss_clip": 1.07331169,
      "balance_loss_mlp": 1.03629112,
      "epoch": 0.07936030782180004,
      "flos": 16472153450880.0,
      "grad_norm": 2.2380522093004376,
      "language_loss": 0.74331439,
      "learning_rate": 3.974553507496378e-06,
      "loss": 0.76608485,
      "num_input_tokens_seen": 14025725,
      "step": 660,
      "time_per_iteration": 2.540658712387085
    },
    {
      "auxiliary_loss_clip": 0.01217108,
      "auxiliary_loss_mlp": 0.01051161,
      "balance_loss_clip": 1.07054913,
      "balance_loss_mlp": 1.03918672,
      "epoch": 0.07948055071243913,
      "flos": 23733290764800.0,
      "grad_norm": 2.6919412560332487,
      "language_loss": 0.89475942,
      "learning_rate": 3.974429492392068e-06,
      "loss": 0.91744208,
      "num_input_tokens_seen": 14045750,
      "step": 661,
      "time_per_iteration": 2.596468448638916
    },
    {
      "auxiliary_loss_clip": 0.01262585,
      "auxiliary_loss_mlp": 0.00764029,
      "balance_loss_clip": 1.07624722,
      "balance_loss_mlp": 1.00003386,
      "epoch": 0.07960079360307822,
      "flos": 19573326996480.0,
      "grad_norm": 1.8403358785575636,
      "language_loss": 0.90973181,
      "learning_rate": 3.974305177768373e-06,
      "loss": 0.92999792,
      "num_input_tokens_seen": 14063960,
      "step": 662,
      "time_per_iteration": 2.4790172576904297
    },
    {
      "auxiliary_loss_clip": 0.01205186,
      "auxiliary_loss_mlp": 0.01052145,
      "balance_loss_clip": 1.07001829,
      "balance_loss_mlp": 1.04060543,
      "epoch": 0.07972103649371731,
      "flos": 23513409659520.0,
      "grad_norm": 2.304305483605101,
      "language_loss": 0.86608994,
      "learning_rate": 3.974180563644152e-06,
      "loss": 0.88866329,
      "num_input_tokens_seen": 14082525,
      "step": 663,
      "time_per_iteration": 2.611039161682129
    },
    {
      "auxiliary_loss_clip": 0.01232641,
      "auxiliary_loss_mlp": 0.01059702,
      "balance_loss_clip": 1.07253551,
      "balance_loss_mlp": 1.04877639,
      "epoch": 0.0798412793843564,
      "flos": 16726867770240.0,
      "grad_norm": 2.652742128899269,
      "language_loss": 0.89274442,
      "learning_rate": 3.97405565003831e-06,
      "loss": 0.91566783,
      "num_input_tokens_seen": 14098610,
      "step": 664,
      "time_per_iteration": 2.5410053730010986
    },
    {
      "auxiliary_loss_clip": 0.01217088,
      "auxiliary_loss_mlp": 0.01048503,
      "balance_loss_clip": 1.06942248,
      "balance_loss_mlp": 1.03648663,
      "epoch": 0.07996152227499549,
      "flos": 18223337214720.0,
      "grad_norm": 3.11565892683201,
      "language_loss": 0.78300601,
      "learning_rate": 3.973930436969794e-06,
      "loss": 0.80566192,
      "num_input_tokens_seen": 14117065,
      "step": 665,
      "time_per_iteration": 2.5721709728240967
    },
    {
      "auxiliary_loss_clip": 0.01215038,
      "auxiliary_loss_mlp": 0.01045687,
      "balance_loss_clip": 1.06612158,
      "balance_loss_mlp": 1.03390288,
      "epoch": 0.08008176516563459,
      "flos": 20594877793920.0,
      "grad_norm": 1.7785638732380034,
      "language_loss": 0.85457528,
      "learning_rate": 3.973804924457602e-06,
      "loss": 0.87718248,
      "num_input_tokens_seen": 14135145,
      "step": 666,
      "time_per_iteration": 2.5511574745178223
    },
    {
      "auxiliary_loss_clip": 0.01216327,
      "auxiliary_loss_mlp": 0.01055205,
      "balance_loss_clip": 1.06651843,
      "balance_loss_mlp": 1.04362345,
      "epoch": 0.08020200805627367,
      "flos": 31834306863360.0,
      "grad_norm": 1.8682061671208274,
      "language_loss": 0.8556028,
      "learning_rate": 3.973679112520771e-06,
      "loss": 0.87831813,
      "num_input_tokens_seen": 14156860,
      "step": 667,
      "time_per_iteration": 2.659978151321411
    },
    {
      "auxiliary_loss_clip": 0.01204722,
      "auxiliary_loss_mlp": 0.0104045,
      "balance_loss_clip": 1.06583059,
      "balance_loss_mlp": 1.02963793,
      "epoch": 0.08032225094691277,
      "flos": 17783503176960.0,
      "grad_norm": 2.097500647352574,
      "language_loss": 0.98953331,
      "learning_rate": 3.973553001178389e-06,
      "loss": 1.01198506,
      "num_input_tokens_seen": 14174365,
      "step": 668,
      "time_per_iteration": 3.3798298835754395
    },
    {
      "auxiliary_loss_clip": 0.01213259,
      "auxiliary_loss_mlp": 0.01043982,
      "balance_loss_clip": 1.07165504,
      "balance_loss_mlp": 1.03238869,
      "epoch": 0.08044249383755185,
      "flos": 24061693835520.0,
      "grad_norm": 2.006771034728547,
      "language_loss": 0.7574355,
      "learning_rate": 3.973426590449585e-06,
      "loss": 0.78000796,
      "num_input_tokens_seen": 14192320,
      "step": 669,
      "time_per_iteration": 2.624657154083252
    },
    {
      "auxiliary_loss_clip": 0.01195969,
      "auxiliary_loss_mlp": 0.01060293,
      "balance_loss_clip": 1.06792402,
      "balance_loss_mlp": 1.04887271,
      "epoch": 0.08056273672819095,
      "flos": 18223624523520.0,
      "grad_norm": 1.927283123588554,
      "language_loss": 0.75457543,
      "learning_rate": 3.9732998803535364e-06,
      "loss": 0.77713805,
      "num_input_tokens_seen": 14210380,
      "step": 670,
      "time_per_iteration": 2.6177937984466553
    },
    {
      "auxiliary_loss_clip": 0.01261682,
      "auxiliary_loss_mlp": 0.0104671,
      "balance_loss_clip": 1.07504272,
      "balance_loss_mlp": 1.03591013,
      "epoch": 0.08068297961883003,
      "flos": 19676856971520.0,
      "grad_norm": 2.2056199452305605,
      "language_loss": 0.85208964,
      "learning_rate": 3.973172870909465e-06,
      "loss": 0.87517357,
      "num_input_tokens_seen": 14225145,
      "step": 671,
      "time_per_iteration": 3.267508029937744
    },
    {
      "auxiliary_loss_clip": 0.01233209,
      "auxiliary_loss_mlp": 0.01042065,
      "balance_loss_clip": 1.06971145,
      "balance_loss_mlp": 1.03045964,
      "epoch": 0.08080322250946913,
      "flos": 23148736830720.0,
      "grad_norm": 2.8009083776990638,
      "language_loss": 0.80655849,
      "learning_rate": 3.973045562136638e-06,
      "loss": 0.82931119,
      "num_input_tokens_seen": 14241960,
      "step": 672,
      "time_per_iteration": 3.340522050857544
    },
    {
      "auxiliary_loss_clip": 0.01248115,
      "auxiliary_loss_mlp": 0.01042388,
      "balance_loss_clip": 1.07226694,
      "balance_loss_mlp": 1.03157592,
      "epoch": 0.08092346540010822,
      "flos": 21763626526080.0,
      "grad_norm": 2.85389859891496,
      "language_loss": 0.91609263,
      "learning_rate": 3.972917954054368e-06,
      "loss": 0.93899763,
      "num_input_tokens_seen": 14260515,
      "step": 673,
      "time_per_iteration": 3.3753490447998047
    },
    {
      "auxiliary_loss_clip": 0.01225998,
      "auxiliary_loss_mlp": 0.01058112,
      "balance_loss_clip": 1.07287645,
      "balance_loss_mlp": 1.0451597,
      "epoch": 0.08104370829074731,
      "flos": 21032485188480.0,
      "grad_norm": 2.6117459340132902,
      "language_loss": 0.81511486,
      "learning_rate": 3.972790046682013e-06,
      "loss": 0.83795589,
      "num_input_tokens_seen": 14279190,
      "step": 674,
      "time_per_iteration": 2.5614469051361084
    },
    {
      "auxiliary_loss_clip": 0.01216137,
      "auxiliary_loss_mlp": 0.01044367,
      "balance_loss_clip": 1.06608772,
      "balance_loss_mlp": 1.0328747,
      "epoch": 0.0811639511813864,
      "flos": 20083186598400.0,
      "grad_norm": 2.0525096280221815,
      "language_loss": 0.7904824,
      "learning_rate": 3.972661840038977e-06,
      "loss": 0.8130874,
      "num_input_tokens_seen": 14299480,
      "step": 675,
      "time_per_iteration": 2.6291770935058594
    },
    {
      "auxiliary_loss_clip": 0.01244688,
      "auxiliary_loss_mlp": 0.01053148,
      "balance_loss_clip": 1.07299435,
      "balance_loss_mlp": 1.04100657,
      "epoch": 0.08128419407202549,
      "flos": 16836718538880.0,
      "grad_norm": 2.417626437162247,
      "language_loss": 0.83719176,
      "learning_rate": 3.972533334144707e-06,
      "loss": 0.86017013,
      "num_input_tokens_seen": 14316405,
      "step": 676,
      "time_per_iteration": 2.5016086101531982
    },
    {
      "auxiliary_loss_clip": 0.01248695,
      "auxiliary_loss_mlp": 0.01042206,
      "balance_loss_clip": 1.07147455,
      "balance_loss_mlp": 1.02961719,
      "epoch": 0.08140443696266458,
      "flos": 23769273214080.0,
      "grad_norm": 2.0113015524865574,
      "language_loss": 0.78591239,
      "learning_rate": 3.972404529018699e-06,
      "loss": 0.80882138,
      "num_input_tokens_seen": 14336265,
      "step": 677,
      "time_per_iteration": 2.5523087978363037
    },
    {
      "auxiliary_loss_clip": 0.01220624,
      "auxiliary_loss_mlp": 0.01046614,
      "balance_loss_clip": 1.06243527,
      "balance_loss_mlp": 1.03534842,
      "epoch": 0.08152467985330367,
      "flos": 24390132819840.0,
      "grad_norm": 1.73388659445206,
      "language_loss": 0.85619766,
      "learning_rate": 3.972275424680493e-06,
      "loss": 0.87887001,
      "num_input_tokens_seen": 14356375,
      "step": 678,
      "time_per_iteration": 2.5865321159362793
    },
    {
      "auxiliary_loss_clip": 0.01260845,
      "auxiliary_loss_mlp": 0.01045511,
      "balance_loss_clip": 1.07374597,
      "balance_loss_mlp": 1.03440642,
      "epoch": 0.08164492274394276,
      "flos": 19317750750720.0,
      "grad_norm": 2.009376756079574,
      "language_loss": 0.91422939,
      "learning_rate": 3.972146021149673e-06,
      "loss": 0.93729293,
      "num_input_tokens_seen": 14374650,
      "step": 679,
      "time_per_iteration": 2.4957919120788574
    },
    {
      "auxiliary_loss_clip": 0.01211441,
      "auxiliary_loss_mlp": 0.01046124,
      "balance_loss_clip": 1.06708813,
      "balance_loss_mlp": 1.03540659,
      "epoch": 0.08176516563458186,
      "flos": 14830461319680.0,
      "grad_norm": 2.566822069432394,
      "language_loss": 0.78518242,
      "learning_rate": 3.972016318445868e-06,
      "loss": 0.80775803,
      "num_input_tokens_seen": 14392650,
      "step": 680,
      "time_per_iteration": 2.5734832286834717
    },
    {
      "auxiliary_loss_clip": 0.01243731,
      "auxiliary_loss_mlp": 0.01044237,
      "balance_loss_clip": 1.0721519,
      "balance_loss_mlp": 1.03287625,
      "epoch": 0.08188540852522094,
      "flos": 22602320161920.0,
      "grad_norm": 2.02559103890426,
      "language_loss": 0.91878617,
      "learning_rate": 3.971886316588757e-06,
      "loss": 0.94166583,
      "num_input_tokens_seen": 14413155,
      "step": 681,
      "time_per_iteration": 2.537391185760498
    },
    {
      "auxiliary_loss_clip": 0.01199138,
      "auxiliary_loss_mlp": 0.01040944,
      "balance_loss_clip": 1.06776381,
      "balance_loss_mlp": 1.02889204,
      "epoch": 0.08200565141586004,
      "flos": 19463727623040.0,
      "grad_norm": 2.8378798534785616,
      "language_loss": 0.73280638,
      "learning_rate": 3.9717560155980595e-06,
      "loss": 0.75520724,
      "num_input_tokens_seen": 14428805,
      "step": 682,
      "time_per_iteration": 2.5645196437835693
    },
    {
      "auxiliary_loss_clip": 0.01240543,
      "auxiliary_loss_mlp": 0.01046081,
      "balance_loss_clip": 1.06895685,
      "balance_loss_mlp": 1.03513122,
      "epoch": 0.08212589430649912,
      "flos": 20594662312320.0,
      "grad_norm": 6.972362921006647,
      "language_loss": 0.91931212,
      "learning_rate": 3.971625415493542e-06,
      "loss": 0.94217837,
      "num_input_tokens_seen": 14447125,
      "step": 683,
      "time_per_iteration": 2.520059585571289
    },
    {
      "auxiliary_loss_clip": 0.01203308,
      "auxiliary_loss_mlp": 0.0104202,
      "balance_loss_clip": 1.06274617,
      "balance_loss_mlp": 1.03062391,
      "epoch": 0.08224613719713822,
      "flos": 25953611086080.0,
      "grad_norm": 1.920820538584633,
      "language_loss": 0.87649381,
      "learning_rate": 3.971494516295017e-06,
      "loss": 0.89894712,
      "num_input_tokens_seen": 14466575,
      "step": 684,
      "time_per_iteration": 2.6189119815826416
    },
    {
      "auxiliary_loss_clip": 0.01214251,
      "auxiliary_loss_mlp": 0.01036352,
      "balance_loss_clip": 1.06645513,
      "balance_loss_mlp": 1.02564704,
      "epoch": 0.08236638008777732,
      "flos": 23768734510080.0,
      "grad_norm": 1.9946330082411483,
      "language_loss": 0.85407424,
      "learning_rate": 3.971363318022341e-06,
      "loss": 0.87658024,
      "num_input_tokens_seen": 14487915,
      "step": 685,
      "time_per_iteration": 2.6549503803253174
    },
    {
      "auxiliary_loss_clip": 0.01227666,
      "auxiliary_loss_mlp": 0.01042077,
      "balance_loss_clip": 1.06573975,
      "balance_loss_mlp": 1.03070426,
      "epoch": 0.0824866229784164,
      "flos": 38799144887040.0,
      "grad_norm": 1.7906261735343931,
      "language_loss": 0.68415052,
      "learning_rate": 3.971231820695417e-06,
      "loss": 0.70684803,
      "num_input_tokens_seen": 14511530,
      "step": 686,
      "time_per_iteration": 2.7292287349700928
    },
    {
      "auxiliary_loss_clip": 0.01231051,
      "auxiliary_loss_mlp": 0.01042396,
      "balance_loss_clip": 1.07005107,
      "balance_loss_mlp": 1.03173852,
      "epoch": 0.0826068658690555,
      "flos": 23107762391040.0,
      "grad_norm": 2.025721019445671,
      "language_loss": 0.81261134,
      "learning_rate": 3.971100024334193e-06,
      "loss": 0.83534575,
      "num_input_tokens_seen": 14529050,
      "step": 687,
      "time_per_iteration": 2.5721166133880615
    },
    {
      "auxiliary_loss_clip": 0.01194395,
      "auxiliary_loss_mlp": 0.01042696,
      "balance_loss_clip": 1.06104183,
      "balance_loss_mlp": 1.03173518,
      "epoch": 0.08272710875969458,
      "flos": 21136374299520.0,
      "grad_norm": 2.0201295455575488,
      "language_loss": 0.86249757,
      "learning_rate": 3.970967928958663e-06,
      "loss": 0.8848685,
      "num_input_tokens_seen": 14546165,
      "step": 688,
      "time_per_iteration": 2.566673517227173
    },
    {
      "auxiliary_loss_clip": 0.01204227,
      "auxiliary_loss_mlp": 0.01052386,
      "balance_loss_clip": 1.06762934,
      "balance_loss_mlp": 1.04137075,
      "epoch": 0.08284735165033368,
      "flos": 19063000517760.0,
      "grad_norm": 1.5972583763447181,
      "language_loss": 0.83240515,
      "learning_rate": 3.970835534588865e-06,
      "loss": 0.85497129,
      "num_input_tokens_seen": 14563660,
      "step": 689,
      "time_per_iteration": 2.6124744415283203
    },
    {
      "auxiliary_loss_clip": 0.01233614,
      "auxiliary_loss_mlp": 0.01058963,
      "balance_loss_clip": 1.07595003,
      "balance_loss_mlp": 1.04805541,
      "epoch": 0.08296759454097276,
      "flos": 16727442387840.0,
      "grad_norm": 1.8772492461618835,
      "language_loss": 0.85627699,
      "learning_rate": 3.970702841244883e-06,
      "loss": 0.87920272,
      "num_input_tokens_seen": 14581980,
      "step": 690,
      "time_per_iteration": 2.537780523300171
    },
    {
      "auxiliary_loss_clip": 0.01247349,
      "auxiliary_loss_mlp": 0.01049108,
      "balance_loss_clip": 1.07395029,
      "balance_loss_mlp": 1.03638256,
      "epoch": 0.08308783743161186,
      "flos": 18004928567040.0,
      "grad_norm": 1.734435621262683,
      "language_loss": 0.82381076,
      "learning_rate": 3.970569848946847e-06,
      "loss": 0.84677535,
      "num_input_tokens_seen": 14601795,
      "step": 691,
      "time_per_iteration": 2.5620083808898926
    },
    {
      "auxiliary_loss_clip": 0.01225842,
      "auxiliary_loss_mlp": 0.01042153,
      "balance_loss_clip": 1.06828701,
      "balance_loss_mlp": 1.03104305,
      "epoch": 0.08320808032225095,
      "flos": 15079788599040.0,
      "grad_norm": 5.293191812118151,
      "language_loss": 0.82546532,
      "learning_rate": 3.970436557714932e-06,
      "loss": 0.84814525,
      "num_input_tokens_seen": 14618315,
      "step": 692,
      "time_per_iteration": 2.4818572998046875
    },
    {
      "auxiliary_loss_clip": 0.01223005,
      "auxiliary_loss_mlp": 0.01043426,
      "balance_loss_clip": 1.07022727,
      "balance_loss_mlp": 1.0313735,
      "epoch": 0.08332832321289003,
      "flos": 22383085501440.0,
      "grad_norm": 2.0163193963883908,
      "language_loss": 0.86339194,
      "learning_rate": 3.970302967569358e-06,
      "loss": 0.88605624,
      "num_input_tokens_seen": 14636905,
      "step": 693,
      "time_per_iteration": 2.584277868270874
    },
    {
      "auxiliary_loss_clip": 0.01243672,
      "auxiliary_loss_mlp": 0.01041273,
      "balance_loss_clip": 1.07406616,
      "balance_loss_mlp": 1.03042507,
      "epoch": 0.08344856610352913,
      "flos": 24717386655360.0,
      "grad_norm": 2.305276933170146,
      "language_loss": 0.68179417,
      "learning_rate": 3.9701690785303896e-06,
      "loss": 0.70464361,
      "num_input_tokens_seen": 14656100,
      "step": 694,
      "time_per_iteration": 3.3895256519317627
    },
    {
      "auxiliary_loss_clip": 0.01244017,
      "auxiliary_loss_mlp": 0.01038234,
      "balance_loss_clip": 1.06888223,
      "balance_loss_mlp": 1.02693868,
      "epoch": 0.08356880899416821,
      "flos": 25370206387200.0,
      "grad_norm": 2.1474723815963217,
      "language_loss": 0.8856231,
      "learning_rate": 3.970034890618339e-06,
      "loss": 0.90844566,
      "num_input_tokens_seen": 14675790,
      "step": 695,
      "time_per_iteration": 2.5535168647766113
    },
    {
      "auxiliary_loss_clip": 0.01224507,
      "auxiliary_loss_mlp": 0.01042333,
      "balance_loss_clip": 1.06702662,
      "balance_loss_mlp": 1.0322237,
      "epoch": 0.08368905188480731,
      "flos": 24353072962560.0,
      "grad_norm": 22.914590350622273,
      "language_loss": 0.87910128,
      "learning_rate": 3.969900403853562e-06,
      "loss": 0.90176964,
      "num_input_tokens_seen": 14694830,
      "step": 696,
      "time_per_iteration": 3.327944755554199
    },
    {
      "auxiliary_loss_clip": 0.01264535,
      "auxiliary_loss_mlp": 0.01045868,
      "balance_loss_clip": 1.07964706,
      "balance_loss_mlp": 1.0348587,
      "epoch": 0.08380929477544641,
      "flos": 18037319656320.0,
      "grad_norm": 1.582657197330352,
      "language_loss": 0.77917379,
      "learning_rate": 3.96976561825646e-06,
      "loss": 0.8022778,
      "num_input_tokens_seen": 14711920,
      "step": 697,
      "time_per_iteration": 3.2368052005767822
    },
    {
      "auxiliary_loss_clip": 0.01196744,
      "auxiliary_loss_mlp": 0.01044505,
      "balance_loss_clip": 1.06494689,
      "balance_loss_mlp": 1.03292966,
      "epoch": 0.08392953766608549,
      "flos": 26286287875200.0,
      "grad_norm": 3.8433797460103665,
      "language_loss": 0.86889958,
      "learning_rate": 3.969630533847479e-06,
      "loss": 0.89131206,
      "num_input_tokens_seen": 14730880,
      "step": 698,
      "time_per_iteration": 2.660919666290283
    },
    {
      "auxiliary_loss_clip": 0.01240881,
      "auxiliary_loss_mlp": 0.01047536,
      "balance_loss_clip": 1.06835032,
      "balance_loss_mlp": 1.03693259,
      "epoch": 0.08404978055672459,
      "flos": 22492146170880.0,
      "grad_norm": 3.5340369429234304,
      "language_loss": 0.84018809,
      "learning_rate": 3.969495150647113e-06,
      "loss": 0.86307216,
      "num_input_tokens_seen": 14749050,
      "step": 699,
      "time_per_iteration": 2.5112462043762207
    },
    {
      "auxiliary_loss_clip": 0.01207785,
      "auxiliary_loss_mlp": 0.01046834,
      "balance_loss_clip": 1.06958497,
      "balance_loss_mlp": 1.03584874,
      "epoch": 0.08417002344736367,
      "flos": 24826878288000.0,
      "grad_norm": 1.746620662816309,
      "language_loss": 0.76487982,
      "learning_rate": 3.969359468675899e-06,
      "loss": 0.78742599,
      "num_input_tokens_seen": 14769180,
      "step": 700,
      "time_per_iteration": 3.420692205429077
    },
    {
      "auxiliary_loss_clip": 0.01240851,
      "auxiliary_loss_mlp": 0.01037719,
      "balance_loss_clip": 1.07236218,
      "balance_loss_mlp": 1.02741981,
      "epoch": 0.08429026633800277,
      "flos": 16945922862720.0,
      "grad_norm": 1.9754281474535698,
      "language_loss": 0.89508402,
      "learning_rate": 3.969223487954418e-06,
      "loss": 0.91786969,
      "num_input_tokens_seen": 14786640,
      "step": 701,
      "time_per_iteration": 2.515623092651367
    },
    {
      "auxiliary_loss_clip": 0.01196511,
      "auxiliary_loss_mlp": 0.0104933,
      "balance_loss_clip": 1.06763792,
      "balance_loss_mlp": 1.0386076,
      "epoch": 0.08441050922864185,
      "flos": 23841920471040.0,
      "grad_norm": 2.5637877075434625,
      "language_loss": 0.82410628,
      "learning_rate": 3.969087208503301e-06,
      "loss": 0.84656465,
      "num_input_tokens_seen": 14806720,
      "step": 702,
      "time_per_iteration": 2.6498188972473145
    },
    {
      "auxiliary_loss_clip": 0.01190836,
      "auxiliary_loss_mlp": 0.01044117,
      "balance_loss_clip": 1.06499684,
      "balance_loss_mlp": 1.03214216,
      "epoch": 0.08453075211928095,
      "flos": 25520205582720.0,
      "grad_norm": 2.0917302696645246,
      "language_loss": 0.84440154,
      "learning_rate": 3.968950630343219e-06,
      "loss": 0.86675107,
      "num_input_tokens_seen": 14823705,
      "step": 703,
      "time_per_iteration": 2.625582456588745
    },
    {
      "auxiliary_loss_clip": 0.01222835,
      "auxiliary_loss_mlp": 0.01047265,
      "balance_loss_clip": 1.06576729,
      "balance_loss_mlp": 1.03551149,
      "epoch": 0.08465099500992004,
      "flos": 19532496211200.0,
      "grad_norm": 2.2477520888216778,
      "language_loss": 0.93543708,
      "learning_rate": 3.968813753494892e-06,
      "loss": 0.95813811,
      "num_input_tokens_seen": 14841865,
      "step": 704,
      "time_per_iteration": 2.5313544273376465
    },
    {
      "auxiliary_loss_clip": 0.01199034,
      "auxiliary_loss_mlp": 0.00764444,
      "balance_loss_clip": 1.06338692,
      "balance_loss_mlp": 0.99998397,
      "epoch": 0.08477123790055913,
      "flos": 29351299403520.0,
      "grad_norm": 2.0996590777265762,
      "language_loss": 0.75297862,
      "learning_rate": 3.968676577979084e-06,
      "loss": 0.77261341,
      "num_input_tokens_seen": 14861415,
      "step": 705,
      "time_per_iteration": 2.6507866382598877
    },
    {
      "auxiliary_loss_clip": 0.0119077,
      "auxiliary_loss_mlp": 0.01043264,
      "balance_loss_clip": 1.06155872,
      "balance_loss_mlp": 1.03196859,
      "epoch": 0.08489148079119822,
      "flos": 18624495283200.0,
      "grad_norm": 2.0596626757152294,
      "language_loss": 0.78309339,
      "learning_rate": 3.968539103816605e-06,
      "loss": 0.80543375,
      "num_input_tokens_seen": 14879215,
      "step": 706,
      "time_per_iteration": 2.5788023471832275
    },
    {
      "auxiliary_loss_clip": 0.01221809,
      "auxiliary_loss_mlp": 0.007646,
      "balance_loss_clip": 1.06720722,
      "balance_loss_mlp": 1.00001335,
      "epoch": 0.0850117236818373,
      "flos": 23471393725440.0,
      "grad_norm": 1.971108933895679,
      "language_loss": 0.8932091,
      "learning_rate": 3.9684013310283085e-06,
      "loss": 0.91307318,
      "num_input_tokens_seen": 14897900,
      "step": 707,
      "time_per_iteration": 2.6142454147338867
    },
    {
      "auxiliary_loss_clip": 0.01227841,
      "auxiliary_loss_mlp": 0.0105383,
      "balance_loss_clip": 1.07463467,
      "balance_loss_mlp": 1.04274988,
      "epoch": 0.0851319665724764,
      "flos": 40625058896640.0,
      "grad_norm": 1.922811636898232,
      "language_loss": 0.63755035,
      "learning_rate": 3.9682632596350956e-06,
      "loss": 0.66036713,
      "num_input_tokens_seen": 14919065,
      "step": 708,
      "time_per_iteration": 2.7088584899902344
    },
    {
      "auxiliary_loss_clip": 0.01239393,
      "auxiliary_loss_mlp": 0.01035399,
      "balance_loss_clip": 1.0713799,
      "balance_loss_mlp": 1.02483714,
      "epoch": 0.0852522094631155,
      "flos": 15879554870400.0,
      "grad_norm": 1.9172139736044205,
      "language_loss": 0.78341269,
      "learning_rate": 3.968124889657911e-06,
      "loss": 0.80616057,
      "num_input_tokens_seen": 14934165,
      "step": 709,
      "time_per_iteration": 2.4899802207946777
    },
    {
      "auxiliary_loss_clip": 0.01194215,
      "auxiliary_loss_mlp": 0.01047943,
      "balance_loss_clip": 1.06302011,
      "balance_loss_mlp": 1.03677261,
      "epoch": 0.08537245235375458,
      "flos": 14567091822720.0,
      "grad_norm": 2.2804779315666397,
      "language_loss": 0.90672714,
      "learning_rate": 3.967986221117746e-06,
      "loss": 0.92914867,
      "num_input_tokens_seen": 14950105,
      "step": 710,
      "time_per_iteration": 2.5875139236450195
    },
    {
      "auxiliary_loss_clip": 0.01171299,
      "auxiliary_loss_mlp": 0.0104113,
      "balance_loss_clip": 1.06226516,
      "balance_loss_mlp": 1.02911925,
      "epoch": 0.08549269524439368,
      "flos": 26468929555200.0,
      "grad_norm": 2.1460700894606624,
      "language_loss": 0.86490011,
      "learning_rate": 3.967847254035635e-06,
      "loss": 0.8870244,
      "num_input_tokens_seen": 14969490,
      "step": 711,
      "time_per_iteration": 2.7098920345306396
    },
    {
      "auxiliary_loss_clip": 0.01216307,
      "auxiliary_loss_mlp": 0.01046195,
      "balance_loss_clip": 1.06927776,
      "balance_loss_mlp": 1.03444076,
      "epoch": 0.08561293813503276,
      "flos": 13590214565760.0,
      "grad_norm": 2.628019355105924,
      "language_loss": 0.86344844,
      "learning_rate": 3.967707988432661e-06,
      "loss": 0.88607347,
      "num_input_tokens_seen": 14987195,
      "step": 712,
      "time_per_iteration": 2.5610897541046143
    },
    {
      "auxiliary_loss_clip": 0.01253144,
      "auxiliary_loss_mlp": 0.01047841,
      "balance_loss_clip": 1.06913018,
      "balance_loss_mlp": 1.03657532,
      "epoch": 0.08573318102567186,
      "flos": 26943524979840.0,
      "grad_norm": 4.448172811454269,
      "language_loss": 0.87873793,
      "learning_rate": 3.967568424329949e-06,
      "loss": 0.90174782,
      "num_input_tokens_seen": 15007620,
      "step": 713,
      "time_per_iteration": 2.5360701084136963
    },
    {
      "auxiliary_loss_clip": 0.01136642,
      "auxiliary_loss_mlp": 0.01015249,
      "balance_loss_clip": 1.04263878,
      "balance_loss_mlp": 1.01143456,
      "epoch": 0.08585342391631094,
      "flos": 67302739319040.0,
      "grad_norm": 0.8233994813524171,
      "language_loss": 0.55554038,
      "learning_rate": 3.967428561748671e-06,
      "loss": 0.57705933,
      "num_input_tokens_seen": 15075590,
      "step": 714,
      "time_per_iteration": 3.268935203552246
    },
    {
      "auxiliary_loss_clip": 0.0118782,
      "auxiliary_loss_mlp": 0.01047991,
      "balance_loss_clip": 1.06094038,
      "balance_loss_mlp": 1.03619576,
      "epoch": 0.08597366680695004,
      "flos": 22456594684800.0,
      "grad_norm": 1.9155834690868487,
      "language_loss": 0.8773253,
      "learning_rate": 3.967288400710045e-06,
      "loss": 0.89968348,
      "num_input_tokens_seen": 15095055,
      "step": 715,
      "time_per_iteration": 2.6126434803009033
    },
    {
      "auxiliary_loss_clip": 0.01211115,
      "auxiliary_loss_mlp": 0.01046168,
      "balance_loss_clip": 1.07375634,
      "balance_loss_mlp": 1.03530192,
      "epoch": 0.08609390969758914,
      "flos": 23550505430400.0,
      "grad_norm": 1.9398796645487912,
      "language_loss": 0.88494527,
      "learning_rate": 3.9671479412353335e-06,
      "loss": 0.90751803,
      "num_input_tokens_seen": 15113520,
      "step": 716,
      "time_per_iteration": 2.596857786178589
    },
    {
      "auxiliary_loss_clip": 0.01241492,
      "auxiliary_loss_mlp": 0.01046501,
      "balance_loss_clip": 1.07251263,
      "balance_loss_mlp": 1.03509295,
      "epoch": 0.08621415258822822,
      "flos": 25885848078720.0,
      "grad_norm": 2.117115523251957,
      "language_loss": 0.73955894,
      "learning_rate": 3.967007183345843e-06,
      "loss": 0.76243889,
      "num_input_tokens_seen": 15133375,
      "step": 717,
      "time_per_iteration": 2.5691041946411133
    },
    {
      "auxiliary_loss_clip": 0.01236991,
      "auxiliary_loss_mlp": 0.01042304,
      "balance_loss_clip": 1.07244885,
      "balance_loss_mlp": 1.03187311,
      "epoch": 0.08633439547886732,
      "flos": 13589568120960.0,
      "grad_norm": 2.171576951205799,
      "language_loss": 0.89603287,
      "learning_rate": 3.966866127062927e-06,
      "loss": 0.91882581,
      "num_input_tokens_seen": 15150500,
      "step": 718,
      "time_per_iteration": 2.4913113117218018
    },
    {
      "auxiliary_loss_clip": 0.01132519,
      "auxiliary_loss_mlp": 0.01004081,
      "balance_loss_clip": 1.03746223,
      "balance_loss_mlp": 1.00040913,
      "epoch": 0.0864546383695064,
      "flos": 57767342434560.0,
      "grad_norm": 0.8698767782351514,
      "language_loss": 0.62731743,
      "learning_rate": 3.966724772407982e-06,
      "loss": 0.64868343,
      "num_input_tokens_seen": 15208015,
      "step": 719,
      "time_per_iteration": 2.9686620235443115
    },
    {
      "auxiliary_loss_clip": 0.01202048,
      "auxiliary_loss_mlp": 0.01045233,
      "balance_loss_clip": 1.06728303,
      "balance_loss_mlp": 1.03455222,
      "epoch": 0.0865748812601455,
      "flos": 20046952753920.0,
      "grad_norm": 2.4279615561576278,
      "language_loss": 0.8858425,
      "learning_rate": 3.966583119402454e-06,
      "loss": 0.90831536,
      "num_input_tokens_seen": 15224780,
      "step": 720,
      "time_per_iteration": 3.4400806427001953
    },
    {
      "auxiliary_loss_clip": 0.0123849,
      "auxiliary_loss_mlp": 0.00764466,
      "balance_loss_clip": 1.07219291,
      "balance_loss_mlp": 1.00001526,
      "epoch": 0.08669512415078459,
      "flos": 35262446935680.0,
      "grad_norm": 1.7942031630823065,
      "language_loss": 0.82150275,
      "learning_rate": 3.9664411680678305e-06,
      "loss": 0.84153223,
      "num_input_tokens_seen": 15246535,
      "step": 721,
      "time_per_iteration": 2.632458448410034
    },
    {
      "auxiliary_loss_clip": 0.01110107,
      "auxiliary_loss_mlp": 0.01004941,
      "balance_loss_clip": 1.03540564,
      "balance_loss_mlp": 1.00134075,
      "epoch": 0.08681536704142367,
      "flos": 65654870048640.0,
      "grad_norm": 0.8409224373736027,
      "language_loss": 0.61443877,
      "learning_rate": 3.966298918425644e-06,
      "loss": 0.63558924,
      "num_input_tokens_seen": 15304025,
      "step": 722,
      "time_per_iteration": 4.607870578765869
    },
    {
      "auxiliary_loss_clip": 0.01242981,
      "auxiliary_loss_mlp": 0.01048659,
      "balance_loss_clip": 1.06984723,
      "balance_loss_mlp": 1.03755474,
      "epoch": 0.08693560993206277,
      "flos": 34529940881280.0,
      "grad_norm": 2.315905477403946,
      "language_loss": 0.82935256,
      "learning_rate": 3.966156370497476e-06,
      "loss": 0.85226893,
      "num_input_tokens_seen": 15327635,
      "step": 723,
      "time_per_iteration": 2.64089298248291
    },
    {
      "auxiliary_loss_clip": 0.01245954,
      "auxiliary_loss_mlp": 0.01056159,
      "balance_loss_clip": 1.07196927,
      "balance_loss_mlp": 1.04522777,
      "epoch": 0.08705585282270185,
      "flos": 23149419189120.0,
      "grad_norm": 1.9754117695694542,
      "language_loss": 0.8817541,
      "learning_rate": 3.96601352430495e-06,
      "loss": 0.90477526,
      "num_input_tokens_seen": 15347405,
      "step": 724,
      "time_per_iteration": 2.530644178390503
    },
    {
      "auxiliary_loss_clip": 0.01226445,
      "auxiliary_loss_mlp": 0.01053512,
      "balance_loss_clip": 1.07198882,
      "balance_loss_mlp": 1.04270625,
      "epoch": 0.08717609571334095,
      "flos": 29497599498240.0,
      "grad_norm": 1.5711549722037426,
      "language_loss": 0.83087593,
      "learning_rate": 3.965870379869735e-06,
      "loss": 0.85367548,
      "num_input_tokens_seen": 15369450,
      "step": 725,
      "time_per_iteration": 3.46901798248291
    },
    {
      "auxiliary_loss_clip": 0.01241289,
      "auxiliary_loss_mlp": 0.01041565,
      "balance_loss_clip": 1.0689038,
      "balance_loss_mlp": 1.0302937,
      "epoch": 0.08729633860398003,
      "flos": 20667489137280.0,
      "grad_norm": 2.114664527946292,
      "language_loss": 0.86905932,
      "learning_rate": 3.965726937213547e-06,
      "loss": 0.8918879,
      "num_input_tokens_seen": 15388085,
      "step": 726,
      "time_per_iteration": 2.5195484161376953
    },
    {
      "auxiliary_loss_clip": 0.01237242,
      "auxiliary_loss_mlp": 0.01050852,
      "balance_loss_clip": 1.06628406,
      "balance_loss_mlp": 1.03949761,
      "epoch": 0.08741658149461913,
      "flos": 18369493655040.0,
      "grad_norm": 3.948081079557401,
      "language_loss": 0.81172085,
      "learning_rate": 3.965583196358144e-06,
      "loss": 0.83460176,
      "num_input_tokens_seen": 15407120,
      "step": 727,
      "time_per_iteration": 2.514963150024414
    },
    {
      "auxiliary_loss_clip": 0.01257923,
      "auxiliary_loss_mlp": 0.01046486,
      "balance_loss_clip": 1.07250571,
      "balance_loss_mlp": 1.03514314,
      "epoch": 0.08753682438525823,
      "flos": 18729677283840.0,
      "grad_norm": 2.281121079126419,
      "language_loss": 0.74573541,
      "learning_rate": 3.965439157325335e-06,
      "loss": 0.76877952,
      "num_input_tokens_seen": 15424485,
      "step": 728,
      "time_per_iteration": 2.45369553565979
    },
    {
      "auxiliary_loss_clip": 0.01216035,
      "auxiliary_loss_mlp": 0.01042981,
      "balance_loss_clip": 1.06279504,
      "balance_loss_mlp": 1.03151286,
      "epoch": 0.08765706727589731,
      "flos": 27776113303680.0,
      "grad_norm": 1.8219668586111268,
      "language_loss": 0.75805223,
      "learning_rate": 3.965294820136968e-06,
      "loss": 0.78064233,
      "num_input_tokens_seen": 15446285,
      "step": 729,
      "time_per_iteration": 2.6602978706359863
    },
    {
      "auxiliary_loss_clip": 0.01222097,
      "auxiliary_loss_mlp": 0.01054354,
      "balance_loss_clip": 1.07008469,
      "balance_loss_mlp": 1.04184937,
      "epoch": 0.08777731016653641,
      "flos": 24389127239040.0,
      "grad_norm": 1.9740549905113545,
      "language_loss": 0.87062752,
      "learning_rate": 3.965150184814938e-06,
      "loss": 0.89339203,
      "num_input_tokens_seen": 15465770,
      "step": 730,
      "time_per_iteration": 2.5938923358917236
    },
    {
      "auxiliary_loss_clip": 0.01213564,
      "auxiliary_loss_mlp": 0.01045629,
      "balance_loss_clip": 1.06544578,
      "balance_loss_mlp": 1.03388715,
      "epoch": 0.08789755305717549,
      "flos": 21981855605760.0,
      "grad_norm": 2.09990304459907,
      "language_loss": 0.76963872,
      "learning_rate": 3.965005251381189e-06,
      "loss": 0.79223061,
      "num_input_tokens_seen": 15483705,
      "step": 731,
      "time_per_iteration": 2.561878204345703
    },
    {
      "auxiliary_loss_clip": 0.01129354,
      "auxiliary_loss_mlp": 0.01005951,
      "balance_loss_clip": 1.02994061,
      "balance_loss_mlp": 1.00227916,
      "epoch": 0.08801779594781459,
      "flos": 58360120583040.0,
      "grad_norm": 0.8893821239611491,
      "language_loss": 0.64625722,
      "learning_rate": 3.964860019857705e-06,
      "loss": 0.66761029,
      "num_input_tokens_seen": 15548620,
      "step": 732,
      "time_per_iteration": 3.119004964828491
    },
    {
      "auxiliary_loss_clip": 0.01255765,
      "auxiliary_loss_mlp": 0.01041336,
      "balance_loss_clip": 1.07386506,
      "balance_loss_mlp": 1.0305295,
      "epoch": 0.08813803883845367,
      "flos": 23294785530240.0,
      "grad_norm": 1.6981561883948681,
      "language_loss": 0.84292877,
      "learning_rate": 3.964714490266518e-06,
      "loss": 0.8658998,
      "num_input_tokens_seen": 15569265,
      "step": 733,
      "time_per_iteration": 2.493178606033325
    },
    {
      "auxiliary_loss_clip": 0.01122995,
      "auxiliary_loss_mlp": 0.01005682,
      "balance_loss_clip": 1.02938759,
      "balance_loss_mlp": 1.0020107,
      "epoch": 0.08825828172909277,
      "flos": 63424924882560.0,
      "grad_norm": 0.8877097745259482,
      "language_loss": 0.64589357,
      "learning_rate": 3.964568662629706e-06,
      "loss": 0.6671803,
      "num_input_tokens_seen": 15630570,
      "step": 734,
      "time_per_iteration": 3.017296314239502
    },
    {
      "auxiliary_loss_clip": 0.01233277,
      "auxiliary_loss_mlp": 0.0103969,
      "balance_loss_clip": 1.06586373,
      "balance_loss_mlp": 1.02829921,
      "epoch": 0.08837852461973186,
      "flos": 26720986268160.0,
      "grad_norm": 2.0811808846829742,
      "language_loss": 0.84133303,
      "learning_rate": 3.9644225369693895e-06,
      "loss": 0.86406273,
      "num_input_tokens_seen": 15650870,
      "step": 735,
      "time_per_iteration": 2.5731072425842285
    },
    {
      "auxiliary_loss_clip": 0.01255918,
      "auxiliary_loss_mlp": 0.01049824,
      "balance_loss_clip": 1.07482886,
      "balance_loss_mlp": 1.03842151,
      "epoch": 0.08849876751037095,
      "flos": 27265427688960.0,
      "grad_norm": 1.9834170849661659,
      "language_loss": 0.86897874,
      "learning_rate": 3.964276113307735e-06,
      "loss": 0.89203614,
      "num_input_tokens_seen": 15670835,
      "step": 736,
      "time_per_iteration": 2.5332117080688477
    },
    {
      "auxiliary_loss_clip": 0.01207955,
      "auxiliary_loss_mlp": 0.01047424,
      "balance_loss_clip": 1.06933761,
      "balance_loss_mlp": 1.03580117,
      "epoch": 0.08861901040101004,
      "flos": 19828759587840.0,
      "grad_norm": 1.880779118197899,
      "language_loss": 0.80800509,
      "learning_rate": 3.9641293916669574e-06,
      "loss": 0.8305589,
      "num_input_tokens_seen": 15689795,
      "step": 737,
      "time_per_iteration": 2.582185983657837
    },
    {
      "auxiliary_loss_clip": 0.01203548,
      "auxiliary_loss_mlp": 0.01038206,
      "balance_loss_clip": 1.06590128,
      "balance_loss_mlp": 1.02607083,
      "epoch": 0.08873925329164913,
      "flos": 23658704173440.0,
      "grad_norm": 2.131178367497206,
      "language_loss": 0.82845724,
      "learning_rate": 3.9639823720693115e-06,
      "loss": 0.85087484,
      "num_input_tokens_seen": 15711650,
      "step": 738,
      "time_per_iteration": 2.6417157649993896
    },
    {
      "auxiliary_loss_clip": 0.01105564,
      "auxiliary_loss_mlp": 0.0101166,
      "balance_loss_clip": 1.03725469,
      "balance_loss_mlp": 1.00827408,
      "epoch": 0.08885949618228822,
      "flos": 71831541893760.0,
      "grad_norm": 0.8351373750601038,
      "language_loss": 0.60022497,
      "learning_rate": 3.963835054537102e-06,
      "loss": 0.6213972,
      "num_input_tokens_seen": 15780615,
      "step": 739,
      "time_per_iteration": 3.235017776489258
    },
    {
      "auxiliary_loss_clip": 0.01219474,
      "auxiliary_loss_mlp": 0.01049937,
      "balance_loss_clip": 1.06283939,
      "balance_loss_mlp": 1.03970885,
      "epoch": 0.08897973907292732,
      "flos": 22346169298560.0,
      "grad_norm": 2.740952902750381,
      "language_loss": 0.60750973,
      "learning_rate": 3.963687439092676e-06,
      "loss": 0.63020384,
      "num_input_tokens_seen": 15801300,
      "step": 740,
      "time_per_iteration": 2.574176073074341
    },
    {
      "auxiliary_loss_clip": 0.01237672,
      "auxiliary_loss_mlp": 0.01038952,
      "balance_loss_clip": 1.07051992,
      "balance_loss_mlp": 1.02819383,
      "epoch": 0.0890999819635664,
      "flos": 21251827589760.0,
      "grad_norm": 2.00325796831368,
      "language_loss": 0.80350685,
      "learning_rate": 3.963539525758427e-06,
      "loss": 0.82627308,
      "num_input_tokens_seen": 15820860,
      "step": 741,
      "time_per_iteration": 2.535719871520996
    },
    {
      "auxiliary_loss_clip": 0.01226153,
      "auxiliary_loss_mlp": 0.01042005,
      "balance_loss_clip": 1.07086885,
      "balance_loss_mlp": 1.03063226,
      "epoch": 0.0892202248542055,
      "flos": 25370888745600.0,
      "grad_norm": 2.9215691528961774,
      "language_loss": 0.67409343,
      "learning_rate": 3.9633913145567925e-06,
      "loss": 0.69677496,
      "num_input_tokens_seen": 15841350,
      "step": 742,
      "time_per_iteration": 2.5982158184051514
    },
    {
      "auxiliary_loss_clip": 0.01222857,
      "auxiliary_loss_mlp": 0.01040881,
      "balance_loss_clip": 1.06855822,
      "balance_loss_mlp": 1.03054583,
      "epoch": 0.08934046774484458,
      "flos": 24457895827200.0,
      "grad_norm": 1.9140463338723634,
      "language_loss": 0.81376088,
      "learning_rate": 3.9632428055102575e-06,
      "loss": 0.83639824,
      "num_input_tokens_seen": 15861360,
      "step": 743,
      "time_per_iteration": 2.5966713428497314
    },
    {
      "auxiliary_loss_clip": 0.01244286,
      "auxiliary_loss_mlp": 0.0104206,
      "balance_loss_clip": 1.0729003,
      "balance_loss_mlp": 1.03047299,
      "epoch": 0.08946071063548368,
      "flos": 35772773414400.0,
      "grad_norm": 2.2728602138404734,
      "language_loss": 0.66841555,
      "learning_rate": 3.9630939986413495e-06,
      "loss": 0.69127905,
      "num_input_tokens_seen": 15883160,
      "step": 744,
      "time_per_iteration": 2.643235683441162
    },
    {
      "auxiliary_loss_clip": 0.0119124,
      "auxiliary_loss_mlp": 0.01047881,
      "balance_loss_clip": 1.06294763,
      "balance_loss_mlp": 1.03630567,
      "epoch": 0.08958095352612276,
      "flos": 14356584167040.0,
      "grad_norm": 1.7837066340580356,
      "language_loss": 0.78000546,
      "learning_rate": 3.962944893972643e-06,
      "loss": 0.80239666,
      "num_input_tokens_seen": 15901610,
      "step": 745,
      "time_per_iteration": 2.5688962936401367
    },
    {
      "auxiliary_loss_clip": 0.01227588,
      "auxiliary_loss_mlp": 0.01050677,
      "balance_loss_clip": 1.07382917,
      "balance_loss_mlp": 1.03951287,
      "epoch": 0.08970119641676186,
      "flos": 17853277345920.0,
      "grad_norm": 2.383179000791531,
      "language_loss": 0.91139805,
      "learning_rate": 3.962795491526756e-06,
      "loss": 0.93418074,
      "num_input_tokens_seen": 15918770,
      "step": 746,
      "time_per_iteration": 3.3623225688934326
    },
    {
      "auxiliary_loss_clip": 0.01260516,
      "auxiliary_loss_mlp": 0.01041249,
      "balance_loss_clip": 1.07445717,
      "balance_loss_mlp": 1.03057432,
      "epoch": 0.08982143930740095,
      "flos": 20811670329600.0,
      "grad_norm": 2.404111055068438,
      "language_loss": 0.89365947,
      "learning_rate": 3.962645791326354e-06,
      "loss": 0.91667712,
      "num_input_tokens_seen": 15938025,
      "step": 747,
      "time_per_iteration": 2.5380802154541016
    },
    {
      "auxiliary_loss_clip": 0.01237713,
      "auxiliary_loss_mlp": 0.0104221,
      "balance_loss_clip": 1.07236433,
      "balance_loss_mlp": 1.03179085,
      "epoch": 0.08994168219804004,
      "flos": 24097712198400.0,
      "grad_norm": 2.6051877897296913,
      "language_loss": 0.83633399,
      "learning_rate": 3.962495793394146e-06,
      "loss": 0.85913324,
      "num_input_tokens_seen": 15957215,
      "step": 748,
      "time_per_iteration": 3.239537477493286
    },
    {
      "auxiliary_loss_clip": 0.01133871,
      "auxiliary_loss_mlp": 0.01007408,
      "balance_loss_clip": 1.02692497,
      "balance_loss_mlp": 1.00368905,
      "epoch": 0.09006192508867913,
      "flos": 57188893812480.0,
      "grad_norm": 0.7420300803272664,
      "language_loss": 0.61215812,
      "learning_rate": 3.9623454977528864e-06,
      "loss": 0.63357091,
      "num_input_tokens_seen": 16015870,
      "step": 749,
      "time_per_iteration": 3.6701695919036865
    },
    {
      "auxiliary_loss_clip": 0.01212802,
      "auxiliary_loss_mlp": 0.01045977,
      "balance_loss_clip": 1.06826282,
      "balance_loss_mlp": 1.03382349,
      "epoch": 0.09018216797931822,
      "flos": 20487505063680.0,
      "grad_norm": 1.970777033342424,
      "language_loss": 0.85174465,
      "learning_rate": 3.962194904425375e-06,
      "loss": 0.87433243,
      "num_input_tokens_seen": 16036500,
      "step": 750,
      "time_per_iteration": 2.6185920238494873
    },
    {
      "auxiliary_loss_clip": 0.01236023,
      "auxiliary_loss_mlp": 0.01041354,
      "balance_loss_clip": 1.07013035,
      "balance_loss_mlp": 1.02967775,
      "epoch": 0.09030241086995731,
      "flos": 22638123043200.0,
      "grad_norm": 1.8780348303788152,
      "language_loss": 0.67734766,
      "learning_rate": 3.9620440134344566e-06,
      "loss": 0.7001214,
      "num_input_tokens_seen": 16054655,
      "step": 751,
      "time_per_iteration": 3.323148250579834
    },
    {
      "auxiliary_loss_clip": 0.01207105,
      "auxiliary_loss_mlp": 0.01039729,
      "balance_loss_clip": 1.06600189,
      "balance_loss_mlp": 1.02753353,
      "epoch": 0.09042265376059641,
      "flos": 21871502046720.0,
      "grad_norm": 2.0852204154834713,
      "language_loss": 0.82407027,
      "learning_rate": 3.9618928248030215e-06,
      "loss": 0.84653854,
      "num_input_tokens_seen": 16074165,
      "step": 752,
      "time_per_iteration": 2.5707337856292725
    },
    {
      "auxiliary_loss_clip": 0.01236855,
      "auxiliary_loss_mlp": 0.01049881,
      "balance_loss_clip": 1.07000899,
      "balance_loss_mlp": 1.03864515,
      "epoch": 0.0905428966512355,
      "flos": 24316192673280.0,
      "grad_norm": 2.5077444509283473,
      "language_loss": 0.82879514,
      "learning_rate": 3.961741338554005e-06,
      "loss": 0.85166252,
      "num_input_tokens_seen": 16092505,
      "step": 753,
      "time_per_iteration": 2.522052526473999
    },
    {
      "auxiliary_loss_clip": 0.01229213,
      "auxiliary_loss_mlp": 0.01050286,
      "balance_loss_clip": 1.07034528,
      "balance_loss_mlp": 1.03882408,
      "epoch": 0.09066313954187459,
      "flos": 35845061535360.0,
      "grad_norm": 2.929239733972447,
      "language_loss": 0.75438964,
      "learning_rate": 3.9615895547103865e-06,
      "loss": 0.77718472,
      "num_input_tokens_seen": 16116150,
      "step": 754,
      "time_per_iteration": 2.6564793586730957
    },
    {
      "auxiliary_loss_clip": 0.01224218,
      "auxiliary_loss_mlp": 0.01041675,
      "balance_loss_clip": 1.07002878,
      "balance_loss_mlp": 1.03126192,
      "epoch": 0.09078338243251367,
      "flos": 29168729550720.0,
      "grad_norm": 1.864324490066283,
      "language_loss": 0.7791456,
      "learning_rate": 3.961437473295193e-06,
      "loss": 0.80180454,
      "num_input_tokens_seen": 16136295,
      "step": 755,
      "time_per_iteration": 2.601822853088379
    },
    {
      "auxiliary_loss_clip": 0.01179527,
      "auxiliary_loss_mlp": 0.01044438,
      "balance_loss_clip": 1.05827498,
      "balance_loss_mlp": 1.03314924,
      "epoch": 0.09090362532315277,
      "flos": 21907699977600.0,
      "grad_norm": 3.308524560375813,
      "language_loss": 0.72648042,
      "learning_rate": 3.961285094331495e-06,
      "loss": 0.74872005,
      "num_input_tokens_seen": 16154210,
      "step": 756,
      "time_per_iteration": 2.610873222351074
    },
    {
      "auxiliary_loss_clip": 0.01258728,
      "auxiliary_loss_mlp": 0.01044727,
      "balance_loss_clip": 1.07609963,
      "balance_loss_mlp": 1.0341351,
      "epoch": 0.09102386821379185,
      "flos": 27344503480320.0,
      "grad_norm": 1.7085585643146453,
      "language_loss": 0.85624099,
      "learning_rate": 3.961132417842406e-06,
      "loss": 0.8792755,
      "num_input_tokens_seen": 16173995,
      "step": 757,
      "time_per_iteration": 2.5288617610931396
    },
    {
      "auxiliary_loss_clip": 0.01232215,
      "auxiliary_loss_mlp": 0.01044873,
      "balance_loss_clip": 1.07107306,
      "balance_loss_mlp": 1.03375673,
      "epoch": 0.09114411110443095,
      "flos": 20813501923200.0,
      "grad_norm": 2.93800398921089,
      "language_loss": 0.75228864,
      "learning_rate": 3.960979443851089e-06,
      "loss": 0.77505958,
      "num_input_tokens_seen": 16191020,
      "step": 758,
      "time_per_iteration": 2.519134283065796
    },
    {
      "auxiliary_loss_clip": 0.01217618,
      "auxiliary_loss_mlp": 0.01042819,
      "balance_loss_clip": 1.06623197,
      "balance_loss_mlp": 1.03153551,
      "epoch": 0.09126435399507005,
      "flos": 26145949438080.0,
      "grad_norm": 1.6459148665720467,
      "language_loss": 0.78855079,
      "learning_rate": 3.96082617238075e-06,
      "loss": 0.81115514,
      "num_input_tokens_seen": 16213645,
      "step": 759,
      "time_per_iteration": 2.6353423595428467
    },
    {
      "auxiliary_loss_clip": 0.01221737,
      "auxiliary_loss_mlp": 0.01044496,
      "balance_loss_clip": 1.06830132,
      "balance_loss_mlp": 1.03405905,
      "epoch": 0.09138459688570913,
      "flos": 24388911757440.0,
      "grad_norm": 10.141536389560802,
      "language_loss": 0.79798836,
      "learning_rate": 3.960672603454639e-06,
      "loss": 0.8206507,
      "num_input_tokens_seen": 16233625,
      "step": 760,
      "time_per_iteration": 2.5745768547058105
    },
    {
      "auxiliary_loss_clip": 0.01232443,
      "auxiliary_loss_mlp": 0.01040755,
      "balance_loss_clip": 1.06903803,
      "balance_loss_mlp": 1.03016305,
      "epoch": 0.09150483977634823,
      "flos": 21032664756480.0,
      "grad_norm": 3.2292339249235185,
      "language_loss": 0.77025688,
      "learning_rate": 3.960518737096054e-06,
      "loss": 0.7929889,
      "num_input_tokens_seen": 16253255,
      "step": 761,
      "time_per_iteration": 2.510680913925171
    },
    {
      "auxiliary_loss_clip": 0.01240402,
      "auxiliary_loss_mlp": 0.01039463,
      "balance_loss_clip": 1.07233405,
      "balance_loss_mlp": 1.0286628,
      "epoch": 0.09162508266698731,
      "flos": 22856998567680.0,
      "grad_norm": 2.905919305148167,
      "language_loss": 0.7304734,
      "learning_rate": 3.960364573328334e-06,
      "loss": 0.753272,
      "num_input_tokens_seen": 16272580,
      "step": 762,
      "time_per_iteration": 2.5412545204162598
    },
    {
      "auxiliary_loss_clip": 0.01206134,
      "auxiliary_loss_mlp": 0.01040318,
      "balance_loss_clip": 1.06518328,
      "balance_loss_mlp": 1.02964902,
      "epoch": 0.0917453255576264,
      "flos": 21724411852800.0,
      "grad_norm": 1.8615635050885389,
      "language_loss": 0.88563764,
      "learning_rate": 3.9602101121748675e-06,
      "loss": 0.90810215,
      "num_input_tokens_seen": 16293075,
      "step": 763,
      "time_per_iteration": 2.6115684509277344
    },
    {
      "auxiliary_loss_clip": 0.01220669,
      "auxiliary_loss_mlp": 0.01038575,
      "balance_loss_clip": 1.06872499,
      "balance_loss_mlp": 1.02853131,
      "epoch": 0.0918655684482655,
      "flos": 14609215497600.0,
      "grad_norm": 2.008012258462435,
      "language_loss": 0.72656739,
      "learning_rate": 3.960055353659085e-06,
      "loss": 0.74915987,
      "num_input_tokens_seen": 16310185,
      "step": 764,
      "time_per_iteration": 2.5066142082214355
    },
    {
      "auxiliary_loss_clip": 0.01210592,
      "auxiliary_loss_mlp": 0.0104438,
      "balance_loss_clip": 1.06844819,
      "balance_loss_mlp": 1.03379476,
      "epoch": 0.09198581133890459,
      "flos": 23435016226560.0,
      "grad_norm": 1.7948369977375693,
      "language_loss": 0.83571273,
      "learning_rate": 3.959900297804465e-06,
      "loss": 0.85826242,
      "num_input_tokens_seen": 16330355,
      "step": 765,
      "time_per_iteration": 2.594778060913086
    },
    {
      "auxiliary_loss_clip": 0.01205877,
      "auxiliary_loss_mlp": 0.01047644,
      "balance_loss_clip": 1.06420135,
      "balance_loss_mlp": 1.03689098,
      "epoch": 0.09210605422954368,
      "flos": 16795887753600.0,
      "grad_norm": 1.9593770955542835,
      "language_loss": 0.77450544,
      "learning_rate": 3.9597449446345276e-06,
      "loss": 0.79704064,
      "num_input_tokens_seen": 16347600,
      "step": 766,
      "time_per_iteration": 2.5255768299102783
    },
    {
      "auxiliary_loss_clip": 0.0121129,
      "auxiliary_loss_mlp": 0.01041075,
      "balance_loss_clip": 1.0655762,
      "balance_loss_mlp": 1.03047132,
      "epoch": 0.09222629712018277,
      "flos": 22674249146880.0,
      "grad_norm": 2.277951232389627,
      "language_loss": 0.82826042,
      "learning_rate": 3.95958929417284e-06,
      "loss": 0.85078406,
      "num_input_tokens_seen": 16365755,
      "step": 767,
      "time_per_iteration": 2.554725170135498
    },
    {
      "auxiliary_loss_clip": 0.0112295,
      "auxiliary_loss_mlp": 0.01003989,
      "balance_loss_clip": 1.02645481,
      "balance_loss_mlp": 0.99996012,
      "epoch": 0.09234654001082186,
      "flos": 69976756327680.0,
      "grad_norm": 0.7881218553321822,
      "language_loss": 0.58797956,
      "learning_rate": 3.9594333464430145e-06,
      "loss": 0.60924894,
      "num_input_tokens_seen": 16435245,
      "step": 768,
      "time_per_iteration": 3.2438137531280518
    },
    {
      "auxiliary_loss_clip": 0.01152644,
      "auxiliary_loss_mlp": 0.01036034,
      "balance_loss_clip": 1.0567801,
      "balance_loss_mlp": 1.02625251,
      "epoch": 0.09246678290146094,
      "flos": 20011437181440.0,
      "grad_norm": 1.8815795360444685,
      "language_loss": 0.87739921,
      "learning_rate": 3.959277101468709e-06,
      "loss": 0.89928591,
      "num_input_tokens_seen": 16454795,
      "step": 769,
      "time_per_iteration": 2.6676254272460938
    },
    {
      "auxiliary_loss_clip": 0.01204624,
      "auxiliary_loss_mlp": 0.01041977,
      "balance_loss_clip": 1.06451154,
      "balance_loss_mlp": 1.03114712,
      "epoch": 0.09258702579210004,
      "flos": 17747448900480.0,
      "grad_norm": 4.300801006051527,
      "language_loss": 0.78533185,
      "learning_rate": 3.959120559273624e-06,
      "loss": 0.80779791,
      "num_input_tokens_seen": 16472580,
      "step": 770,
      "time_per_iteration": 2.5300285816192627
    },
    {
      "auxiliary_loss_clip": 0.01205416,
      "auxiliary_loss_mlp": 0.01042755,
      "balance_loss_clip": 1.06745899,
      "balance_loss_mlp": 1.0328306,
      "epoch": 0.09270726868273914,
      "flos": 20886544229760.0,
      "grad_norm": 2.2510859508489998,
      "language_loss": 0.83523977,
      "learning_rate": 3.958963719881509e-06,
      "loss": 0.85772145,
      "num_input_tokens_seen": 16490670,
      "step": 771,
      "time_per_iteration": 2.5514917373657227
    },
    {
      "auxiliary_loss_clip": 0.01236097,
      "auxiliary_loss_mlp": 0.01046992,
      "balance_loss_clip": 1.07132244,
      "balance_loss_mlp": 1.03644836,
      "epoch": 0.09282751157337822,
      "flos": 17015697031680.0,
      "grad_norm": 1.8922393242082698,
      "language_loss": 0.93970394,
      "learning_rate": 3.958806583316154e-06,
      "loss": 0.96253479,
      "num_input_tokens_seen": 16508640,
      "step": 772,
      "time_per_iteration": 3.243525981903076
    },
    {
      "auxiliary_loss_clip": 0.01253902,
      "auxiliary_loss_mlp": 0.01039477,
      "balance_loss_clip": 1.07453108,
      "balance_loss_mlp": 1.02870607,
      "epoch": 0.09294775446401732,
      "flos": 32523647748480.0,
      "grad_norm": 1.8091859732670064,
      "language_loss": 0.78661937,
      "learning_rate": 3.9586491496013985e-06,
      "loss": 0.80955315,
      "num_input_tokens_seen": 16531035,
      "step": 773,
      "time_per_iteration": 2.571927309036255
    },
    {
      "auxiliary_loss_clip": 0.01241072,
      "auxiliary_loss_mlp": 0.01045262,
      "balance_loss_clip": 1.07316315,
      "balance_loss_mlp": 1.03405607,
      "epoch": 0.0930679973546564,
      "flos": 18259750627200.0,
      "grad_norm": 2.5128235515169535,
      "language_loss": 0.83176446,
      "learning_rate": 3.958491418761124e-06,
      "loss": 0.85462779,
      "num_input_tokens_seen": 16548605,
      "step": 774,
      "time_per_iteration": 3.197734832763672
    },
    {
      "auxiliary_loss_clip": 0.01220755,
      "auxiliary_loss_mlp": 0.01046075,
      "balance_loss_clip": 1.06334484,
      "balance_loss_mlp": 1.03562033,
      "epoch": 0.0931882402452955,
      "flos": 21099745405440.0,
      "grad_norm": 2.2136310057020534,
      "language_loss": 0.72889328,
      "learning_rate": 3.958333390819258e-06,
      "loss": 0.75156152,
      "num_input_tokens_seen": 16565535,
      "step": 775,
      "time_per_iteration": 3.279985189437866
    },
    {
      "auxiliary_loss_clip": 0.01253735,
      "auxiliary_loss_mlp": 0.01050483,
      "balance_loss_clip": 1.07426572,
      "balance_loss_mlp": 1.04014206,
      "epoch": 0.0933084831359346,
      "flos": 24207275658240.0,
      "grad_norm": 2.2704975768591003,
      "language_loss": 0.80397034,
      "learning_rate": 3.9581750657997754e-06,
      "loss": 0.82701248,
      "num_input_tokens_seen": 16584900,
      "step": 776,
      "time_per_iteration": 2.5226974487304688
    },
    {
      "auxiliary_loss_clip": 0.01217589,
      "auxiliary_loss_mlp": 0.0104137,
      "balance_loss_clip": 1.06566143,
      "balance_loss_mlp": 1.03093302,
      "epoch": 0.09342872602657368,
      "flos": 25480272637440.0,
      "grad_norm": 1.71273368053945,
      "language_loss": 0.89777064,
      "learning_rate": 3.95801644372669e-06,
      "loss": 0.92036021,
      "num_input_tokens_seen": 16604805,
      "step": 777,
      "time_per_iteration": 3.375718116760254
    },
    {
      "auxiliary_loss_clip": 0.01225818,
      "auxiliary_loss_mlp": 0.01036286,
      "balance_loss_clip": 1.06525826,
      "balance_loss_mlp": 1.02547991,
      "epoch": 0.09354896891721277,
      "flos": 23149060053120.0,
      "grad_norm": 2.1268951007902577,
      "language_loss": 0.84312129,
      "learning_rate": 3.957857524624068e-06,
      "loss": 0.86574233,
      "num_input_tokens_seen": 16623685,
      "step": 778,
      "time_per_iteration": 2.5608747005462646
    },
    {
      "auxiliary_loss_clip": 0.01220182,
      "auxiliary_loss_mlp": 0.01039036,
      "balance_loss_clip": 1.06708384,
      "balance_loss_mlp": 1.02873683,
      "epoch": 0.09366921180785186,
      "flos": 24279563779200.0,
      "grad_norm": 1.8082726127464892,
      "language_loss": 0.89375043,
      "learning_rate": 3.957698308516016e-06,
      "loss": 0.91634262,
      "num_input_tokens_seen": 16644985,
      "step": 779,
      "time_per_iteration": 2.5842576026916504
    },
    {
      "auxiliary_loss_clip": 0.01235963,
      "auxiliary_loss_mlp": 0.0076357,
      "balance_loss_clip": 1.07534337,
      "balance_loss_mlp": 1.00003648,
      "epoch": 0.09378945469849095,
      "flos": 18730036419840.0,
      "grad_norm": 1.8862766899888983,
      "language_loss": 0.82688439,
      "learning_rate": 3.957538795426688e-06,
      "loss": 0.84687972,
      "num_input_tokens_seen": 16662410,
      "step": 780,
      "time_per_iteration": 2.519973039627075
    },
    {
      "auxiliary_loss_clip": 0.01221954,
      "auxiliary_loss_mlp": 0.01051366,
      "balance_loss_clip": 1.06897378,
      "balance_loss_mlp": 1.04016054,
      "epoch": 0.09390969758913004,
      "flos": 23218834222080.0,
      "grad_norm": 2.126045240351721,
      "language_loss": 0.77135015,
      "learning_rate": 3.9573789853802804e-06,
      "loss": 0.79408336,
      "num_input_tokens_seen": 16680885,
      "step": 781,
      "time_per_iteration": 2.566715955734253
    },
    {
      "auxiliary_loss_clip": 0.01220732,
      "auxiliary_loss_mlp": 0.00763744,
      "balance_loss_clip": 1.07116389,
      "balance_loss_mlp": 1.0000267,
      "epoch": 0.09402994047976913,
      "flos": 19646728439040.0,
      "grad_norm": 2.2818162487498217,
      "language_loss": 0.74777758,
      "learning_rate": 3.957218878401037e-06,
      "loss": 0.76762229,
      "num_input_tokens_seen": 16699375,
      "step": 782,
      "time_per_iteration": 2.549079418182373
    },
    {
      "auxiliary_loss_clip": 0.01256979,
      "auxiliary_loss_mlp": 0.01047055,
      "balance_loss_clip": 1.07691932,
      "balance_loss_mlp": 1.03592086,
      "epoch": 0.09415018337040823,
      "flos": 29420463041280.0,
      "grad_norm": 3.0429568301400134,
      "language_loss": 0.8947165,
      "learning_rate": 3.957058474513246e-06,
      "loss": 0.9177568,
      "num_input_tokens_seen": 16719230,
      "step": 783,
      "time_per_iteration": 2.55474853515625
    },
    {
      "auxiliary_loss_clip": 0.01234837,
      "auxiliary_loss_mlp": 0.01036693,
      "balance_loss_clip": 1.07162476,
      "balance_loss_mlp": 1.02605975,
      "epoch": 0.09427042626104731,
      "flos": 24572092141440.0,
      "grad_norm": 1.9127993878733849,
      "language_loss": 0.78336012,
      "learning_rate": 3.956897773741241e-06,
      "loss": 0.80607545,
      "num_input_tokens_seen": 16738220,
      "step": 784,
      "time_per_iteration": 2.5585761070251465
    },
    {
      "auxiliary_loss_clip": 0.01207102,
      "auxiliary_loss_mlp": 0.01043029,
      "balance_loss_clip": 1.06444466,
      "balance_loss_mlp": 1.03208566,
      "epoch": 0.09439066915168641,
      "flos": 26359581576960.0,
      "grad_norm": 1.9038368676099195,
      "language_loss": 0.71648973,
      "learning_rate": 3.956736776109398e-06,
      "loss": 0.73899108,
      "num_input_tokens_seen": 16759395,
      "step": 785,
      "time_per_iteration": 2.5801525115966797
    },
    {
      "auxiliary_loss_clip": 0.01228459,
      "auxiliary_loss_mlp": 0.00764391,
      "balance_loss_clip": 1.06875777,
      "balance_loss_mlp": 1.00002289,
      "epoch": 0.09451091204232549,
      "flos": 19427278296960.0,
      "grad_norm": 1.8991505290698107,
      "language_loss": 0.83676255,
      "learning_rate": 3.956575481642143e-06,
      "loss": 0.85669106,
      "num_input_tokens_seen": 16778285,
      "step": 786,
      "time_per_iteration": 2.5168731212615967
    },
    {
      "auxiliary_loss_clip": 0.01185329,
      "auxiliary_loss_mlp": 0.01048422,
      "balance_loss_clip": 1.05971813,
      "balance_loss_mlp": 1.03713906,
      "epoch": 0.09463115493296459,
      "flos": 25368051571200.0,
      "grad_norm": 2.3696279173477497,
      "language_loss": 0.74984396,
      "learning_rate": 3.956413890363943e-06,
      "loss": 0.77218151,
      "num_input_tokens_seen": 16795265,
      "step": 787,
      "time_per_iteration": 2.650292158126831
    },
    {
      "auxiliary_loss_clip": 0.01232435,
      "auxiliary_loss_mlp": 0.01042487,
      "balance_loss_clip": 1.0694437,
      "balance_loss_mlp": 1.03105497,
      "epoch": 0.09475139782360369,
      "flos": 10123254869760.0,
      "grad_norm": 2.190885896853387,
      "language_loss": 0.81679285,
      "learning_rate": 3.956252002299312e-06,
      "loss": 0.83954203,
      "num_input_tokens_seen": 16811165,
      "step": 788,
      "time_per_iteration": 2.4715182781219482
    },
    {
      "auxiliary_loss_clip": 0.01256254,
      "auxiliary_loss_mlp": 0.01037733,
      "balance_loss_clip": 1.07587671,
      "balance_loss_mlp": 1.02636659,
      "epoch": 0.09487164071424277,
      "flos": 17231088936960.0,
      "grad_norm": 2.241285256299441,
      "language_loss": 0.9113155,
      "learning_rate": 3.956089817472807e-06,
      "loss": 0.93425524,
      "num_input_tokens_seen": 16828470,
      "step": 789,
      "time_per_iteration": 2.472564220428467
    },
    {
      "auxiliary_loss_clip": 0.01220526,
      "auxiliary_loss_mlp": 0.01042568,
      "balance_loss_clip": 1.07113636,
      "balance_loss_mlp": 1.03087389,
      "epoch": 0.09499188360488187,
      "flos": 30849564528000.0,
      "grad_norm": 2.6620955090165017,
      "language_loss": 0.85394228,
      "learning_rate": 3.955927335909032e-06,
      "loss": 0.8765732,
      "num_input_tokens_seen": 16851680,
      "step": 790,
      "time_per_iteration": 2.6445605754852295
    },
    {
      "auxiliary_loss_clip": 0.01190948,
      "auxiliary_loss_mlp": 0.01032002,
      "balance_loss_clip": 1.06897473,
      "balance_loss_mlp": 1.02177989,
      "epoch": 0.09511212649552095,
      "flos": 29351694453120.0,
      "grad_norm": 2.2118340729420627,
      "language_loss": 0.75587887,
      "learning_rate": 3.955764557632634e-06,
      "loss": 0.77810836,
      "num_input_tokens_seen": 16871490,
      "step": 791,
      "time_per_iteration": 2.67232346534729
    },
    {
      "auxiliary_loss_clip": 0.01215823,
      "auxiliary_loss_mlp": 0.01041025,
      "balance_loss_clip": 1.06738222,
      "balance_loss_mlp": 1.03064179,
      "epoch": 0.09523236938616005,
      "flos": 10378687461120.0,
      "grad_norm": 2.2210295547720147,
      "language_loss": 0.94729877,
      "learning_rate": 3.955601482668309e-06,
      "loss": 0.96986729,
      "num_input_tokens_seen": 16889350,
      "step": 792,
      "time_per_iteration": 2.539994478225708
    },
    {
      "auxiliary_loss_clip": 0.01181042,
      "auxiliary_loss_mlp": 0.01040544,
      "balance_loss_clip": 1.05739176,
      "balance_loss_mlp": 1.02895117,
      "epoch": 0.09535261227679913,
      "flos": 19061815368960.0,
      "grad_norm": 1.865288579395569,
      "language_loss": 0.88236117,
      "learning_rate": 3.955438111040794e-06,
      "loss": 0.90457702,
      "num_input_tokens_seen": 16907625,
      "step": 793,
      "time_per_iteration": 2.5963032245635986
    },
    {
      "auxiliary_loss_clip": 0.01184178,
      "auxiliary_loss_mlp": 0.0103867,
      "balance_loss_clip": 1.06101346,
      "balance_loss_mlp": 1.02795255,
      "epoch": 0.09547285516743823,
      "flos": 20922993555840.0,
      "grad_norm": 1.910129269260517,
      "language_loss": 0.80285174,
      "learning_rate": 3.955274442774873e-06,
      "loss": 0.82508016,
      "num_input_tokens_seen": 16926205,
      "step": 794,
      "time_per_iteration": 2.619797706604004
    },
    {
      "auxiliary_loss_clip": 0.01234076,
      "auxiliary_loss_mlp": 0.01043207,
      "balance_loss_clip": 1.0691874,
      "balance_loss_mlp": 1.03223348,
      "epoch": 0.09559309805807732,
      "flos": 30154405639680.0,
      "grad_norm": 2.6095774917638,
      "language_loss": 0.70466709,
      "learning_rate": 3.9551104778953725e-06,
      "loss": 0.72743988,
      "num_input_tokens_seen": 16946500,
      "step": 795,
      "time_per_iteration": 2.5989956855773926
    },
    {
      "auxiliary_loss_clip": 0.01202046,
      "auxiliary_loss_mlp": 0.01037789,
      "balance_loss_clip": 1.06345677,
      "balance_loss_mlp": 1.02704215,
      "epoch": 0.0957133409487164,
      "flos": 21066743784960.0,
      "grad_norm": 2.118252157640684,
      "language_loss": 0.85165131,
      "learning_rate": 3.954946216427167e-06,
      "loss": 0.87404966,
      "num_input_tokens_seen": 16966960,
      "step": 796,
      "time_per_iteration": 2.58160400390625
    },
    {
      "auxiliary_loss_clip": 0.01100601,
      "auxiliary_loss_mlp": 0.01007093,
      "balance_loss_clip": 1.03456044,
      "balance_loss_mlp": 1.00294447,
      "epoch": 0.0958335838393555,
      "flos": 71297979315840.0,
      "grad_norm": 0.8764931478142052,
      "language_loss": 0.61545932,
      "learning_rate": 3.954781658395176e-06,
      "loss": 0.63653624,
      "num_input_tokens_seen": 17023215,
      "step": 797,
      "time_per_iteration": 3.122253179550171
    },
    {
      "auxiliary_loss_clip": 0.01228035,
      "auxiliary_loss_mlp": 0.01046699,
      "balance_loss_clip": 1.06754375,
      "balance_loss_mlp": 1.03571939,
      "epoch": 0.09595382672999458,
      "flos": 21872974504320.0,
      "grad_norm": 1.9217903881254186,
      "language_loss": 0.92560965,
      "learning_rate": 3.95461680382436e-06,
      "loss": 0.94835699,
      "num_input_tokens_seen": 17042140,
      "step": 798,
      "time_per_iteration": 3.3956949710845947
    },
    {
      "auxiliary_loss_clip": 0.01241824,
      "auxiliary_loss_mlp": 0.01049435,
      "balance_loss_clip": 1.07592916,
      "balance_loss_mlp": 1.03788948,
      "epoch": 0.09607406962063368,
      "flos": 18695562341760.0,
      "grad_norm": 2.7872625104505433,
      "language_loss": 0.85986185,
      "learning_rate": 3.9544516527397295e-06,
      "loss": 0.88277447,
      "num_input_tokens_seen": 17058490,
      "step": 799,
      "time_per_iteration": 2.528494358062744
    },
    {
      "auxiliary_loss_clip": 0.01203247,
      "auxiliary_loss_mlp": 0.01042529,
      "balance_loss_clip": 1.06584811,
      "balance_loss_mlp": 1.03190184,
      "epoch": 0.09619431251127276,
      "flos": 22568456615040.0,
      "grad_norm": 1.6457009212158078,
      "language_loss": 0.80850255,
      "learning_rate": 3.954286205166338e-06,
      "loss": 0.83096021,
      "num_input_tokens_seen": 17079655,
      "step": 800,
      "time_per_iteration": 3.3440635204315186
    },
    {
      "auxiliary_loss_clip": 0.01241329,
      "auxiliary_loss_mlp": 0.01049898,
      "balance_loss_clip": 1.07410145,
      "balance_loss_mlp": 1.03837669,
      "epoch": 0.09631455540191186,
      "flos": 14246230608000.0,
      "grad_norm": 2.185856652974379,
      "language_loss": 0.8390227,
      "learning_rate": 3.954120461129282e-06,
      "loss": 0.86193496,
      "num_input_tokens_seen": 17097065,
      "step": 801,
      "time_per_iteration": 3.180696725845337
    },
    {
      "auxiliary_loss_clip": 0.01255802,
      "auxiliary_loss_mlp": 0.01042073,
      "balance_loss_clip": 1.07748771,
      "balance_loss_mlp": 1.03102839,
      "epoch": 0.09643479829255096,
      "flos": 20740387789440.0,
      "grad_norm": 1.9157662034450935,
      "language_loss": 0.83700061,
      "learning_rate": 3.953954420653706e-06,
      "loss": 0.85997933,
      "num_input_tokens_seen": 17114090,
      "step": 802,
      "time_per_iteration": 2.499013900756836
    },
    {
      "auxiliary_loss_clip": 0.01238333,
      "auxiliary_loss_mlp": 0.01045403,
      "balance_loss_clip": 1.07250559,
      "balance_loss_mlp": 1.03428078,
      "epoch": 0.09655504118319004,
      "flos": 24420476833920.0,
      "grad_norm": 1.8602508089375915,
      "language_loss": 0.8800419,
      "learning_rate": 3.953788083764798e-06,
      "loss": 0.90287924,
      "num_input_tokens_seen": 17133325,
      "step": 803,
      "time_per_iteration": 3.3925650119781494
    },
    {
      "auxiliary_loss_clip": 0.01189491,
      "auxiliary_loss_mlp": 0.01046537,
      "balance_loss_clip": 1.06865215,
      "balance_loss_mlp": 1.035254,
      "epoch": 0.09667528407382914,
      "flos": 18441961344000.0,
      "grad_norm": 2.1780272973575436,
      "language_loss": 0.92036736,
      "learning_rate": 3.953621450487792e-06,
      "loss": 0.94272763,
      "num_input_tokens_seen": 17151945,
      "step": 804,
      "time_per_iteration": 2.5810434818267822
    },
    {
      "auxiliary_loss_clip": 0.0113568,
      "auxiliary_loss_mlp": 0.0100758,
      "balance_loss_clip": 1.03099465,
      "balance_loss_mlp": 1.00357425,
      "epoch": 0.09679552696446822,
      "flos": 70816455544320.0,
      "grad_norm": 0.8380963686147276,
      "language_loss": 0.61251444,
      "learning_rate": 3.953454520847964e-06,
      "loss": 0.63394701,
      "num_input_tokens_seen": 17216790,
      "step": 805,
      "time_per_iteration": 3.2256829738616943
    },
    {
      "auxiliary_loss_clip": 0.01216512,
      "auxiliary_loss_mlp": 0.01040129,
      "balance_loss_clip": 1.06671476,
      "balance_loss_mlp": 1.02760649,
      "epoch": 0.09691576985510732,
      "flos": 21945514020480.0,
      "grad_norm": 1.8947488758438518,
      "language_loss": 0.73259622,
      "learning_rate": 3.9532872948706395e-06,
      "loss": 0.75516266,
      "num_input_tokens_seen": 17236285,
      "step": 806,
      "time_per_iteration": 2.556593418121338
    },
    {
      "auxiliary_loss_clip": 0.01223589,
      "auxiliary_loss_mlp": 0.0104746,
      "balance_loss_clip": 1.06952691,
      "balance_loss_mlp": 1.03556871,
      "epoch": 0.09703601274574641,
      "flos": 17965211103360.0,
      "grad_norm": 2.367058132327071,
      "language_loss": 0.82585919,
      "learning_rate": 3.9531197725811845e-06,
      "loss": 0.84856969,
      "num_input_tokens_seen": 17251670,
      "step": 807,
      "time_per_iteration": 2.523057460784912
    },
    {
      "auxiliary_loss_clip": 0.01254588,
      "auxiliary_loss_mlp": 0.01044931,
      "balance_loss_clip": 1.07643676,
      "balance_loss_mlp": 1.0345118,
      "epoch": 0.0971562556363855,
      "flos": 22162162901760.0,
      "grad_norm": 1.83577378244001,
      "language_loss": 0.87803149,
      "learning_rate": 3.952951954005013e-06,
      "loss": 0.90102673,
      "num_input_tokens_seen": 17271355,
      "step": 808,
      "time_per_iteration": 2.517145872116089
    },
    {
      "auxiliary_loss_clip": 0.01220373,
      "auxiliary_loss_mlp": 0.01035711,
      "balance_loss_clip": 1.06493044,
      "balance_loss_mlp": 1.02553701,
      "epoch": 0.0972764985270246,
      "flos": 25848716394240.0,
      "grad_norm": 1.921277095838438,
      "language_loss": 0.84816921,
      "learning_rate": 3.952783839167584e-06,
      "loss": 0.87073004,
      "num_input_tokens_seen": 17291400,
      "step": 809,
      "time_per_iteration": 2.5988826751708984
    },
    {
      "auxiliary_loss_clip": 0.01234636,
      "auxiliary_loss_mlp": 0.01051067,
      "balance_loss_clip": 1.0695436,
      "balance_loss_mlp": 1.03924155,
      "epoch": 0.09739674141766368,
      "flos": 20339373375360.0,
      "grad_norm": 2.534639199745739,
      "language_loss": 0.74351418,
      "learning_rate": 3.952615428094398e-06,
      "loss": 0.76637125,
      "num_input_tokens_seen": 17310920,
      "step": 810,
      "time_per_iteration": 2.537630796432495
    },
    {
      "auxiliary_loss_clip": 0.0117562,
      "auxiliary_loss_mlp": 0.01050176,
      "balance_loss_clip": 1.05844605,
      "balance_loss_mlp": 1.03914952,
      "epoch": 0.09751698430830277,
      "flos": 15743059188480.0,
      "grad_norm": 1.6978351886828218,
      "language_loss": 0.73558933,
      "learning_rate": 3.952446720811004e-06,
      "loss": 0.75784731,
      "num_input_tokens_seen": 17329245,
      "step": 811,
      "time_per_iteration": 2.597297430038452
    },
    {
      "auxiliary_loss_clip": 0.01092532,
      "auxiliary_loss_mlp": 0.01005437,
      "balance_loss_clip": 1.02486873,
      "balance_loss_mlp": 1.00121737,
      "epoch": 0.09763722719894186,
      "flos": 63716806800000.0,
      "grad_norm": 0.8335617792241169,
      "language_loss": 0.63605231,
      "learning_rate": 3.952277717342995e-06,
      "loss": 0.65703207,
      "num_input_tokens_seen": 17395680,
      "step": 812,
      "time_per_iteration": 3.269383430480957
    },
    {
      "auxiliary_loss_clip": 0.0122565,
      "auxiliary_loss_mlp": 0.01044378,
      "balance_loss_clip": 1.06866503,
      "balance_loss_mlp": 1.0333811,
      "epoch": 0.09775747008958095,
      "flos": 22090916275200.0,
      "grad_norm": 3.0227241200342463,
      "language_loss": 0.85668766,
      "learning_rate": 3.952108417716009e-06,
      "loss": 0.87938792,
      "num_input_tokens_seen": 17415135,
      "step": 813,
      "time_per_iteration": 2.5668752193450928
    },
    {
      "auxiliary_loss_clip": 0.01240569,
      "auxiliary_loss_mlp": 0.0106392,
      "balance_loss_clip": 1.07471824,
      "balance_loss_mlp": 1.05230927,
      "epoch": 0.09787771298022005,
      "flos": 21286050272640.0,
      "grad_norm": 1.8641025883405788,
      "language_loss": 0.85131419,
      "learning_rate": 3.951938821955727e-06,
      "loss": 0.87435913,
      "num_input_tokens_seen": 17434535,
      "step": 814,
      "time_per_iteration": 2.557171583175659
    },
    {
      "auxiliary_loss_clip": 0.01221039,
      "auxiliary_loss_mlp": 0.01041716,
      "balance_loss_clip": 1.06949186,
      "balance_loss_mlp": 1.03011131,
      "epoch": 0.09799795587085913,
      "flos": 22054574689920.0,
      "grad_norm": 1.5485501188567774,
      "language_loss": 0.76628995,
      "learning_rate": 3.9517689300878786e-06,
      "loss": 0.78891754,
      "num_input_tokens_seen": 17454270,
      "step": 815,
      "time_per_iteration": 2.569112777709961
    },
    {
      "auxiliary_loss_clip": 0.01253196,
      "auxiliary_loss_mlp": 0.01048904,
      "balance_loss_clip": 1.07310891,
      "balance_loss_mlp": 1.03734076,
      "epoch": 0.09811819876149823,
      "flos": 22163743100160.0,
      "grad_norm": 1.8074400912840598,
      "language_loss": 0.78495026,
      "learning_rate": 3.951598742138236e-06,
      "loss": 0.80797124,
      "num_input_tokens_seen": 17472995,
      "step": 816,
      "time_per_iteration": 2.506138801574707
    },
    {
      "auxiliary_loss_clip": 0.01219645,
      "auxiliary_loss_mlp": 0.01044966,
      "balance_loss_clip": 1.06345177,
      "balance_loss_mlp": 1.03365922,
      "epoch": 0.09823844165213731,
      "flos": 22231111057920.0,
      "grad_norm": 2.04210868199264,
      "language_loss": 0.80102754,
      "learning_rate": 3.951428258132615e-06,
      "loss": 0.82367361,
      "num_input_tokens_seen": 17491115,
      "step": 817,
      "time_per_iteration": 2.5632827281951904
    },
    {
      "auxiliary_loss_clip": 0.01225501,
      "auxiliary_loss_mlp": 0.010524,
      "balance_loss_clip": 1.07161558,
      "balance_loss_mlp": 1.04062831,
      "epoch": 0.09835868454277641,
      "flos": 22487728798080.0,
      "grad_norm": 1.90896095399811,
      "language_loss": 0.84509647,
      "learning_rate": 3.951257478096879e-06,
      "loss": 0.86787546,
      "num_input_tokens_seen": 17509480,
      "step": 818,
      "time_per_iteration": 2.5612056255340576
    },
    {
      "auxiliary_loss_clip": 0.01224451,
      "auxiliary_loss_mlp": 0.00764592,
      "balance_loss_clip": 1.06964374,
      "balance_loss_mlp": 1.00006032,
      "epoch": 0.0984789274334155,
      "flos": 16362554077440.0,
      "grad_norm": 2.819386655306629,
      "language_loss": 0.68649983,
      "learning_rate": 3.951086402056936e-06,
      "loss": 0.70639026,
      "num_input_tokens_seen": 17524080,
      "step": 819,
      "time_per_iteration": 2.5498387813568115
    },
    {
      "auxiliary_loss_clip": 0.01157734,
      "auxiliary_loss_mlp": 0.00764563,
      "balance_loss_clip": 1.06728888,
      "balance_loss_mlp": 1.00007248,
      "epoch": 0.09859917032405459,
      "flos": 24243545416320.0,
      "grad_norm": 1.6078236189118822,
      "language_loss": 0.83438742,
      "learning_rate": 3.950915030038735e-06,
      "loss": 0.85361034,
      "num_input_tokens_seen": 17543875,
      "step": 820,
      "time_per_iteration": 2.839092254638672
    },
    {
      "auxiliary_loss_clip": 0.0123058,
      "auxiliary_loss_mlp": 0.01043523,
      "balance_loss_clip": 1.06996238,
      "balance_loss_mlp": 1.03266346,
      "epoch": 0.09871941321469369,
      "flos": 17420195064960.0,
      "grad_norm": 2.1801582473989582,
      "language_loss": 0.83534747,
      "learning_rate": 3.9507433620682765e-06,
      "loss": 0.85808855,
      "num_input_tokens_seen": 17560810,
      "step": 821,
      "time_per_iteration": 2.874157190322876
    },
    {
      "auxiliary_loss_clip": 0.01203699,
      "auxiliary_loss_mlp": 0.01049083,
      "balance_loss_clip": 1.06358516,
      "balance_loss_mlp": 1.03771102,
      "epoch": 0.09883965610533277,
      "flos": 28477341590400.0,
      "grad_norm": 1.656322871188082,
      "language_loss": 0.88049924,
      "learning_rate": 3.9505713981716e-06,
      "loss": 0.90302706,
      "num_input_tokens_seen": 17583640,
      "step": 822,
      "time_per_iteration": 2.6769068241119385
    },
    {
      "auxiliary_loss_clip": 0.01220481,
      "auxiliary_loss_mlp": 0.01042954,
      "balance_loss_clip": 1.07116115,
      "balance_loss_mlp": 1.03226042,
      "epoch": 0.09895989899597187,
      "flos": 23693932437120.0,
      "grad_norm": 1.813834112273762,
      "language_loss": 0.81190968,
      "learning_rate": 3.950399138374795e-06,
      "loss": 0.83454406,
      "num_input_tokens_seen": 17602720,
      "step": 823,
      "time_per_iteration": 2.558539867401123
    },
    {
      "auxiliary_loss_clip": 0.01231509,
      "auxiliary_loss_mlp": 0.01044375,
      "balance_loss_clip": 1.06714046,
      "balance_loss_mlp": 1.03309226,
      "epoch": 0.09908014188661095,
      "flos": 24679608526080.0,
      "grad_norm": 1.6827354318034067,
      "language_loss": 0.74400908,
      "learning_rate": 3.95022658270399e-06,
      "loss": 0.76676798,
      "num_input_tokens_seen": 17623085,
      "step": 824,
      "time_per_iteration": 3.3720877170562744
    },
    {
      "auxiliary_loss_clip": 0.01213241,
      "auxiliary_loss_mlp": 0.01041649,
      "balance_loss_clip": 1.06730986,
      "balance_loss_mlp": 1.03057492,
      "epoch": 0.09920038477725004,
      "flos": 14064307200000.0,
      "grad_norm": 1.9034350792718542,
      "language_loss": 0.781335,
      "learning_rate": 3.9500537311853635e-06,
      "loss": 0.80388391,
      "num_input_tokens_seen": 17641040,
      "step": 825,
      "time_per_iteration": 2.51812744140625
    },
    {
      "auxiliary_loss_clip": 0.0123678,
      "auxiliary_loss_mlp": 0.01043087,
      "balance_loss_clip": 1.0682919,
      "balance_loss_mlp": 1.03160167,
      "epoch": 0.09932062766788914,
      "flos": 13407070095360.0,
      "grad_norm": 3.035083570808106,
      "language_loss": 0.83366144,
      "learning_rate": 3.949880583845136e-06,
      "loss": 0.85646009,
      "num_input_tokens_seen": 17659115,
      "step": 826,
      "time_per_iteration": 2.545421838760376
    },
    {
      "auxiliary_loss_clip": 0.01218963,
      "auxiliary_loss_mlp": 0.01046129,
      "balance_loss_clip": 1.06821251,
      "balance_loss_mlp": 1.03497672,
      "epoch": 0.09944087055852822,
      "flos": 19500751566720.0,
      "grad_norm": 1.9705324209820514,
      "language_loss": 0.81547451,
      "learning_rate": 3.949707140709575e-06,
      "loss": 0.83812535,
      "num_input_tokens_seen": 17678845,
      "step": 827,
      "time_per_iteration": 3.319261074066162
    },
    {
      "auxiliary_loss_clip": 0.01234948,
      "auxiliary_loss_mlp": 0.01039722,
      "balance_loss_clip": 1.06725454,
      "balance_loss_mlp": 1.02941024,
      "epoch": 0.09956111344916732,
      "flos": 17749100926080.0,
      "grad_norm": 2.171570872830343,
      "language_loss": 0.83059132,
      "learning_rate": 3.949533401804991e-06,
      "loss": 0.853338,
      "num_input_tokens_seen": 17695750,
      "step": 828,
      "time_per_iteration": 2.4865283966064453
    },
    {
      "auxiliary_loss_clip": 0.01231718,
      "auxiliary_loss_mlp": 0.00764567,
      "balance_loss_clip": 1.07041526,
      "balance_loss_mlp": 1.00004935,
      "epoch": 0.0996813563398064,
      "flos": 17967581400960.0,
      "grad_norm": 1.8813725245447654,
      "language_loss": 0.90611756,
      "learning_rate": 3.949359367157739e-06,
      "loss": 0.92608047,
      "num_input_tokens_seen": 17714445,
      "step": 829,
      "time_per_iteration": 3.3413572311401367
    },
    {
      "auxiliary_loss_clip": 0.01237427,
      "auxiliary_loss_mlp": 0.01043575,
      "balance_loss_clip": 1.06993675,
      "balance_loss_mlp": 1.03221393,
      "epoch": 0.0998015992304455,
      "flos": 17457039440640.0,
      "grad_norm": 2.305719481196296,
      "language_loss": 0.75660098,
      "learning_rate": 3.949185036794222e-06,
      "loss": 0.77941096,
      "num_input_tokens_seen": 17732455,
      "step": 830,
      "time_per_iteration": 2.5150344371795654
    },
    {
      "auxiliary_loss_clip": 0.01248247,
      "auxiliary_loss_mlp": 0.01043081,
      "balance_loss_clip": 1.07239449,
      "balance_loss_mlp": 1.03232801,
      "epoch": 0.0999218421210846,
      "flos": 25888757080320.0,
      "grad_norm": 1.6409624883503955,
      "language_loss": 0.78551722,
      "learning_rate": 3.949010410740884e-06,
      "loss": 0.80843049,
      "num_input_tokens_seen": 17755280,
      "step": 831,
      "time_per_iteration": 2.5663540363311768
    },
    {
      "auxiliary_loss_clip": 0.01212952,
      "auxiliary_loss_mlp": 0.00764552,
      "balance_loss_clip": 1.06834388,
      "balance_loss_mlp": 1.00009584,
      "epoch": 0.10004208501172368,
      "flos": 21215916967680.0,
      "grad_norm": 1.6188494846926345,
      "language_loss": 0.86455977,
      "learning_rate": 3.948835489024216e-06,
      "loss": 0.8843348,
      "num_input_tokens_seen": 17775015,
      "step": 832,
      "time_per_iteration": 2.564117908477783
    },
    {
      "auxiliary_loss_clip": 0.01235673,
      "auxiliary_loss_mlp": 0.01040208,
      "balance_loss_clip": 1.06813419,
      "balance_loss_mlp": 1.02934837,
      "epoch": 0.10016232790236278,
      "flos": 17348409734400.0,
      "grad_norm": 2.2297956637223053,
      "language_loss": 0.90384543,
      "learning_rate": 3.948660271670755e-06,
      "loss": 0.92660421,
      "num_input_tokens_seen": 17792165,
      "step": 833,
      "time_per_iteration": 2.525570869445801
    },
    {
      "auxiliary_loss_clip": 0.01216253,
      "auxiliary_loss_mlp": 0.01045737,
      "balance_loss_clip": 1.06751299,
      "balance_loss_mlp": 1.03510916,
      "epoch": 0.10028257079300186,
      "flos": 25666541591040.0,
      "grad_norm": 2.0975291970970646,
      "language_loss": 0.84480757,
      "learning_rate": 3.948484758707079e-06,
      "loss": 0.86742747,
      "num_input_tokens_seen": 17811765,
      "step": 834,
      "time_per_iteration": 2.5966455936431885
    },
    {
      "auxiliary_loss_clip": 0.01193094,
      "auxiliary_loss_mlp": 0.01042407,
      "balance_loss_clip": 1.06019402,
      "balance_loss_mlp": 1.03067732,
      "epoch": 0.10040281368364096,
      "flos": 25156035544320.0,
      "grad_norm": 2.0389344336043758,
      "language_loss": 0.83610398,
      "learning_rate": 3.948308950159815e-06,
      "loss": 0.858459,
      "num_input_tokens_seen": 17830445,
      "step": 835,
      "time_per_iteration": 2.60986065864563
    },
    {
      "auxiliary_loss_clip": 0.01194724,
      "auxiliary_loss_mlp": 0.01045499,
      "balance_loss_clip": 1.05928612,
      "balance_loss_mlp": 1.03375053,
      "epoch": 0.10052305657428004,
      "flos": 17603303621760.0,
      "grad_norm": 2.469373135588881,
      "language_loss": 0.75759947,
      "learning_rate": 3.9481328460556326e-06,
      "loss": 0.78000164,
      "num_input_tokens_seen": 17847665,
      "step": 836,
      "time_per_iteration": 2.5570383071899414
    },
    {
      "auxiliary_loss_clip": 0.01211864,
      "auxiliary_loss_mlp": 0.01045634,
      "balance_loss_clip": 1.06635928,
      "balance_loss_mlp": 1.03386247,
      "epoch": 0.10064329946491914,
      "flos": 18660154510080.0,
      "grad_norm": 2.380700869328185,
      "language_loss": 0.89809179,
      "learning_rate": 3.9479564464212455e-06,
      "loss": 0.92066669,
      "num_input_tokens_seen": 17866825,
      "step": 837,
      "time_per_iteration": 2.553717613220215
    },
    {
      "auxiliary_loss_clip": 0.01254779,
      "auxiliary_loss_mlp": 0.0104661,
      "balance_loss_clip": 1.07200336,
      "balance_loss_mlp": 1.03477883,
      "epoch": 0.10076354235555823,
      "flos": 17199056983680.0,
      "grad_norm": 2.3042279231750027,
      "language_loss": 0.76076603,
      "learning_rate": 3.947779751283414e-06,
      "loss": 0.78377998,
      "num_input_tokens_seen": 17883995,
      "step": 838,
      "time_per_iteration": 2.474430561065674
    },
    {
      "auxiliary_loss_clip": 0.01234569,
      "auxiliary_loss_mlp": 0.00765068,
      "balance_loss_clip": 1.07267404,
      "balance_loss_mlp": 1.00010502,
      "epoch": 0.10088378524619732,
      "flos": 22962252395520.0,
      "grad_norm": 1.7998301165170738,
      "language_loss": 0.75888979,
      "learning_rate": 3.947602760668944e-06,
      "loss": 0.77888614,
      "num_input_tokens_seen": 17903785,
      "step": 839,
      "time_per_iteration": 2.532369613647461
    },
    {
      "auxiliary_loss_clip": 0.01237049,
      "auxiliary_loss_mlp": 0.01049751,
      "balance_loss_clip": 1.07411981,
      "balance_loss_mlp": 1.03800297,
      "epoch": 0.10100402813683641,
      "flos": 37885828746240.0,
      "grad_norm": 1.789496284111264,
      "language_loss": 0.71289456,
      "learning_rate": 3.947425474604684e-06,
      "loss": 0.7357626,
      "num_input_tokens_seen": 17927720,
      "step": 840,
      "time_per_iteration": 2.6679022312164307
    },
    {
      "auxiliary_loss_clip": 0.01220094,
      "auxiliary_loss_mlp": 0.01050517,
      "balance_loss_clip": 1.06744123,
      "balance_loss_mlp": 1.03904939,
      "epoch": 0.1011242710274755,
      "flos": 21543458112000.0,
      "grad_norm": 1.9690073978246612,
      "language_loss": 0.92514485,
      "learning_rate": 3.947247893117528e-06,
      "loss": 0.947851,
      "num_input_tokens_seen": 17946225,
      "step": 841,
      "time_per_iteration": 2.5698113441467285
    },
    {
      "auxiliary_loss_clip": 0.01231355,
      "auxiliary_loss_mlp": 0.01043307,
      "balance_loss_clip": 1.0684793,
      "balance_loss_mlp": 1.03178513,
      "epoch": 0.10124451391811459,
      "flos": 13621456419840.0,
      "grad_norm": 3.3586695482244173,
      "language_loss": 0.69828212,
      "learning_rate": 3.947070016234413e-06,
      "loss": 0.7210288,
      "num_input_tokens_seen": 17962015,
      "step": 842,
      "time_per_iteration": 2.489269971847534
    },
    {
      "auxiliary_loss_clip": 0.01231674,
      "auxiliary_loss_mlp": 0.01042418,
      "balance_loss_clip": 1.07137299,
      "balance_loss_mlp": 1.03138518,
      "epoch": 0.10136475680875369,
      "flos": 16649228522880.0,
      "grad_norm": 2.3384660085368556,
      "language_loss": 0.74703544,
      "learning_rate": 3.946891843982326e-06,
      "loss": 0.76977634,
      "num_input_tokens_seen": 17979680,
      "step": 843,
      "time_per_iteration": 2.5321874618530273
    },
    {
      "auxiliary_loss_clip": 0.01236888,
      "auxiliary_loss_mlp": 0.01049568,
      "balance_loss_clip": 1.0729481,
      "balance_loss_mlp": 1.03790367,
      "epoch": 0.10148499969939277,
      "flos": 19461034103040.0,
      "grad_norm": 2.0456967242406976,
      "language_loss": 0.74558693,
      "learning_rate": 3.9467133763882935e-06,
      "loss": 0.76845151,
      "num_input_tokens_seen": 17998145,
      "step": 844,
      "time_per_iteration": 2.509788990020752
    },
    {
      "auxiliary_loss_clip": 0.01221596,
      "auxiliary_loss_mlp": 0.01045163,
      "balance_loss_clip": 1.06753206,
      "balance_loss_mlp": 1.03402841,
      "epoch": 0.10160524259003187,
      "flos": 21104988791040.0,
      "grad_norm": 2.1092080766765533,
      "language_loss": 0.86170149,
      "learning_rate": 3.9465346134793905e-06,
      "loss": 0.88436902,
      "num_input_tokens_seen": 18017955,
      "step": 845,
      "time_per_iteration": 2.541602373123169
    },
    {
      "auxiliary_loss_clip": 0.01208352,
      "auxiliary_loss_mlp": 0.01046795,
      "balance_loss_clip": 1.06977129,
      "balance_loss_mlp": 1.03545833,
      "epoch": 0.10172548548067095,
      "flos": 17712687513600.0,
      "grad_norm": 2.7457940612750606,
      "language_loss": 0.79737431,
      "learning_rate": 3.9463555552827335e-06,
      "loss": 0.81992579,
      "num_input_tokens_seen": 18035125,
      "step": 846,
      "time_per_iteration": 2.6576967239379883
    },
    {
      "auxiliary_loss_clip": 0.01218392,
      "auxiliary_loss_mlp": 0.01039557,
      "balance_loss_clip": 1.0660125,
      "balance_loss_mlp": 1.02819633,
      "epoch": 0.10184572837131005,
      "flos": 21104845136640.0,
      "grad_norm": 2.4021835581696473,
      "language_loss": 0.86235654,
      "learning_rate": 3.946176201825487e-06,
      "loss": 0.88493598,
      "num_input_tokens_seen": 18053160,
      "step": 847,
      "time_per_iteration": 2.539459705352783
    },
    {
      "auxiliary_loss_clip": 0.01222907,
      "auxiliary_loss_mlp": 0.01048764,
      "balance_loss_clip": 1.07386994,
      "balance_loss_mlp": 1.03685498,
      "epoch": 0.10196597126194913,
      "flos": 26067591918720.0,
      "grad_norm": 1.8273498981741665,
      "language_loss": 0.83515042,
      "learning_rate": 3.9459965531348575e-06,
      "loss": 0.85786712,
      "num_input_tokens_seen": 18072815,
      "step": 848,
      "time_per_iteration": 2.589956283569336
    },
    {
      "auxiliary_loss_clip": 0.01221897,
      "auxiliary_loss_mlp": 0.0076538,
      "balance_loss_clip": 1.0712949,
      "balance_loss_mlp": 1.00008512,
      "epoch": 0.10208621415258823,
      "flos": 29314634595840.0,
      "grad_norm": 2.7249732804356817,
      "language_loss": 0.85981137,
      "learning_rate": 3.945816609238098e-06,
      "loss": 0.87968409,
      "num_input_tokens_seen": 18092225,
      "step": 849,
      "time_per_iteration": 2.6345951557159424
    },
    {
      "auxiliary_loss_clip": 0.01181141,
      "auxiliary_loss_mlp": 0.01041118,
      "balance_loss_clip": 1.06556726,
      "balance_loss_mlp": 1.02880394,
      "epoch": 0.10220645704322733,
      "flos": 23805794367360.0,
      "grad_norm": 1.7634513737873236,
      "language_loss": 0.85207468,
      "learning_rate": 3.945636370162507e-06,
      "loss": 0.87429726,
      "num_input_tokens_seen": 18112335,
      "step": 850,
      "time_per_iteration": 3.4216551780700684
    },
    {
      "auxiliary_loss_clip": 0.01232757,
      "auxiliary_loss_mlp": 0.01047381,
      "balance_loss_clip": 1.06991184,
      "balance_loss_mlp": 1.03658044,
      "epoch": 0.10232669993386641,
      "flos": 23218546913280.0,
      "grad_norm": 1.7711195479214523,
      "language_loss": 0.79016912,
      "learning_rate": 3.945455835935425e-06,
      "loss": 0.81297052,
      "num_input_tokens_seen": 18131520,
      "step": 851,
      "time_per_iteration": 2.5623202323913574
    },
    {
      "auxiliary_loss_clip": 0.01220644,
      "auxiliary_loss_mlp": 0.01052154,
      "balance_loss_clip": 1.0691905,
      "balance_loss_mlp": 1.04039359,
      "epoch": 0.1024469428245055,
      "flos": 22922929981440.0,
      "grad_norm": 3.1000149717364254,
      "language_loss": 0.75237119,
      "learning_rate": 3.94527500658424e-06,
      "loss": 0.77509916,
      "num_input_tokens_seen": 18149185,
      "step": 852,
      "time_per_iteration": 2.563441753387451
    },
    {
      "auxiliary_loss_clip": 0.0118852,
      "auxiliary_loss_mlp": 0.01045565,
      "balance_loss_clip": 1.06686306,
      "balance_loss_mlp": 1.03462195,
      "epoch": 0.10256718571514459,
      "flos": 31359495957120.0,
      "grad_norm": 1.8450839460594908,
      "language_loss": 0.81124562,
      "learning_rate": 3.945093882136382e-06,
      "loss": 0.83358651,
      "num_input_tokens_seen": 18172960,
      "step": 853,
      "time_per_iteration": 4.168609380722046
    },
    {
      "auxiliary_loss_clip": 0.01219982,
      "auxiliary_loss_mlp": 0.00764201,
      "balance_loss_clip": 1.0710516,
      "balance_loss_mlp": 1.00010061,
      "epoch": 0.10268742860578368,
      "flos": 23474877344640.0,
      "grad_norm": 1.776553789906378,
      "language_loss": 0.84672195,
      "learning_rate": 3.944912462619329e-06,
      "loss": 0.8665638,
      "num_input_tokens_seen": 18191925,
      "step": 854,
      "time_per_iteration": 2.5688798427581787
    },
    {
      "auxiliary_loss_clip": 0.01230176,
      "auxiliary_loss_mlp": 0.01046636,
      "balance_loss_clip": 1.07111144,
      "balance_loss_mlp": 1.03441656,
      "epoch": 0.10280767149642277,
      "flos": 25520313323520.0,
      "grad_norm": 1.9063932291408048,
      "language_loss": 0.8047682,
      "learning_rate": 3.9447307480606025e-06,
      "loss": 0.82753623,
      "num_input_tokens_seen": 18212010,
      "step": 855,
      "time_per_iteration": 3.4602725505828857
    },
    {
      "auxiliary_loss_clip": 0.01211739,
      "auxiliary_loss_mlp": 0.01049356,
      "balance_loss_clip": 1.06651807,
      "balance_loss_mlp": 1.03835845,
      "epoch": 0.10292791438706186,
      "flos": 17347691462400.0,
      "grad_norm": 2.3858885467876187,
      "language_loss": 0.9018116,
      "learning_rate": 3.944548738487767e-06,
      "loss": 0.92442256,
      "num_input_tokens_seen": 18229525,
      "step": 856,
      "time_per_iteration": 2.51202392578125
    },
    {
      "auxiliary_loss_clip": 0.01259678,
      "auxiliary_loss_mlp": 0.01047372,
      "balance_loss_clip": 1.07812047,
      "balance_loss_mlp": 1.03537333,
      "epoch": 0.10304815727770096,
      "flos": 27052693390080.0,
      "grad_norm": 2.1267893489824714,
      "language_loss": 0.90995711,
      "learning_rate": 3.944366433928434e-06,
      "loss": 0.93302757,
      "num_input_tokens_seen": 18249505,
      "step": 857,
      "time_per_iteration": 2.554260492324829
    },
    {
      "auxiliary_loss_clip": 0.01215116,
      "auxiliary_loss_mlp": 0.01039843,
      "balance_loss_clip": 1.06776571,
      "balance_loss_mlp": 1.02997196,
      "epoch": 0.10316840016834004,
      "flos": 22782591544320.0,
      "grad_norm": 1.586471419624451,
      "language_loss": 0.83529651,
      "learning_rate": 3.9441838344102594e-06,
      "loss": 0.85784614,
      "num_input_tokens_seen": 18269230,
      "step": 858,
      "time_per_iteration": 2.5728750228881836
    },
    {
      "auxiliary_loss_clip": 0.01228881,
      "auxiliary_loss_mlp": 0.01043033,
      "balance_loss_clip": 1.07268262,
      "balance_loss_mlp": 1.03199422,
      "epoch": 0.10328864305897914,
      "flos": 20704584908160.0,
      "grad_norm": 2.1778685821263166,
      "language_loss": 0.67390478,
      "learning_rate": 3.944000939960943e-06,
      "loss": 0.69662392,
      "num_input_tokens_seen": 18287955,
      "step": 859,
      "time_per_iteration": 2.5609171390533447
    },
    {
      "auxiliary_loss_clip": 0.01238206,
      "auxiliary_loss_mlp": 0.01044837,
      "balance_loss_clip": 1.07038188,
      "balance_loss_mlp": 1.03484142,
      "epoch": 0.10340888594961822,
      "flos": 28478814048000.0,
      "grad_norm": 1.9092474364842906,
      "language_loss": 0.79969895,
      "learning_rate": 3.943817750608229e-06,
      "loss": 0.82252932,
      "num_input_tokens_seen": 18310505,
      "step": 860,
      "time_per_iteration": 2.5937371253967285
    },
    {
      "auxiliary_loss_clip": 0.01239797,
      "auxiliary_loss_mlp": 0.01039472,
      "balance_loss_clip": 1.07263398,
      "balance_loss_mlp": 1.02843904,
      "epoch": 0.10352912884025732,
      "flos": 13370333460480.0,
      "grad_norm": 2.362996196133597,
      "language_loss": 0.82070845,
      "learning_rate": 3.943634266379908e-06,
      "loss": 0.84350115,
      "num_input_tokens_seen": 18327400,
      "step": 861,
      "time_per_iteration": 2.493540048599243
    },
    {
      "auxiliary_loss_clip": 0.01241326,
      "auxiliary_loss_mlp": 0.01048773,
      "balance_loss_clip": 1.07421124,
      "balance_loss_mlp": 1.03800273,
      "epoch": 0.10364937173089642,
      "flos": 25558558329600.0,
      "grad_norm": 1.7361210071618254,
      "language_loss": 0.84764922,
      "learning_rate": 3.943450487303815e-06,
      "loss": 0.87055027,
      "num_input_tokens_seen": 18347895,
      "step": 862,
      "time_per_iteration": 2.5528788566589355
    },
    {
      "auxiliary_loss_clip": 0.01237558,
      "auxiliary_loss_mlp": 0.01043617,
      "balance_loss_clip": 1.07397807,
      "balance_loss_mlp": 1.0324589,
      "epoch": 0.1037696146215355,
      "flos": 21215486004480.0,
      "grad_norm": 2.855585656488234,
      "language_loss": 0.8536123,
      "learning_rate": 3.943266413407827e-06,
      "loss": 0.87642407,
      "num_input_tokens_seen": 18367170,
      "step": 863,
      "time_per_iteration": 2.522669792175293
    },
    {
      "auxiliary_loss_clip": 0.01244036,
      "auxiliary_loss_mlp": 0.01048749,
      "balance_loss_clip": 1.07534027,
      "balance_loss_mlp": 1.03844345,
      "epoch": 0.1038898575121746,
      "flos": 25807382818560.0,
      "grad_norm": 4.461103437516621,
      "language_loss": 0.84993929,
      "learning_rate": 3.94308204471987e-06,
      "loss": 0.87286711,
      "num_input_tokens_seen": 18386185,
      "step": 864,
      "time_per_iteration": 2.549011707305908
    },
    {
      "auxiliary_loss_clip": 0.01205562,
      "auxiliary_loss_mlp": 0.01043485,
      "balance_loss_clip": 1.06759715,
      "balance_loss_mlp": 1.0333519,
      "epoch": 0.10401010040281368,
      "flos": 19062425900160.0,
      "grad_norm": 2.2152117553880437,
      "language_loss": 0.74848735,
      "learning_rate": 3.942897381267912e-06,
      "loss": 0.77097785,
      "num_input_tokens_seen": 18402550,
      "step": 865,
      "time_per_iteration": 2.5553743839263916
    },
    {
      "auxiliary_loss_clip": 0.01237557,
      "auxiliary_loss_mlp": 0.01040186,
      "balance_loss_clip": 1.07303452,
      "balance_loss_mlp": 1.02787161,
      "epoch": 0.10413034329345278,
      "flos": 16355119962240.0,
      "grad_norm": 2.461889756001147,
      "language_loss": 0.66259277,
      "learning_rate": 3.942712423079965e-06,
      "loss": 0.68537027,
      "num_input_tokens_seen": 18418940,
      "step": 866,
      "time_per_iteration": 2.4869585037231445
    },
    {
      "auxiliary_loss_clip": 0.01184215,
      "auxiliary_loss_mlp": 0.01042543,
      "balance_loss_clip": 1.05818617,
      "balance_loss_mlp": 1.03173053,
      "epoch": 0.10425058618409186,
      "flos": 17236511890560.0,
      "grad_norm": 2.220438157546921,
      "language_loss": 0.90240145,
      "learning_rate": 3.942527170184088e-06,
      "loss": 0.92466897,
      "num_input_tokens_seen": 18435560,
      "step": 867,
      "time_per_iteration": 2.5531749725341797
    },
    {
      "auxiliary_loss_clip": 0.01257947,
      "auxiliary_loss_mlp": 0.01042323,
      "balance_loss_clip": 1.07668388,
      "balance_loss_mlp": 1.03137374,
      "epoch": 0.10437082907473096,
      "flos": 17967365919360.0,
      "grad_norm": 2.5430816714810547,
      "language_loss": 0.77735454,
      "learning_rate": 3.942341622608385e-06,
      "loss": 0.80035722,
      "num_input_tokens_seen": 18452590,
      "step": 868,
      "time_per_iteration": 2.461937189102173
    },
    {
      "auxiliary_loss_clip": 0.01226418,
      "auxiliary_loss_mlp": 0.01050803,
      "balance_loss_clip": 1.07539332,
      "balance_loss_mlp": 1.04006827,
      "epoch": 0.10449107196537005,
      "flos": 36283315374720.0,
      "grad_norm": 1.5690366864774377,
      "language_loss": 0.77826023,
      "learning_rate": 3.942155780381001e-06,
      "loss": 0.80103242,
      "num_input_tokens_seen": 18476325,
      "step": 869,
      "time_per_iteration": 2.6916921138763428
    },
    {
      "auxiliary_loss_clip": 0.01224249,
      "auxiliary_loss_mlp": 0.01054246,
      "balance_loss_clip": 1.071015,
      "balance_loss_mlp": 1.0421406,
      "epoch": 0.10461131485600914,
      "flos": 23802095266560.0,
      "grad_norm": 1.9178861344789284,
      "language_loss": 0.75863862,
      "learning_rate": 3.94196964353013e-06,
      "loss": 0.78142357,
      "num_input_tokens_seen": 18495775,
      "step": 870,
      "time_per_iteration": 2.5819168090820312
    },
    {
      "auxiliary_loss_clip": 0.01218292,
      "auxiliary_loss_mlp": 0.00764063,
      "balance_loss_clip": 1.06760359,
      "balance_loss_mlp": 1.00012302,
      "epoch": 0.10473155774664823,
      "flos": 18405476104320.0,
      "grad_norm": 2.4057577085901953,
      "language_loss": 0.80807567,
      "learning_rate": 3.941783212084008e-06,
      "loss": 0.82789922,
      "num_input_tokens_seen": 18513530,
      "step": 871,
      "time_per_iteration": 2.560265064239502
    },
    {
      "auxiliary_loss_clip": 0.01201534,
      "auxiliary_loss_mlp": 0.01059156,
      "balance_loss_clip": 1.06629157,
      "balance_loss_mlp": 1.04836202,
      "epoch": 0.10485180063728732,
      "flos": 25592637358080.0,
      "grad_norm": 2.687663357540943,
      "language_loss": 0.78517759,
      "learning_rate": 3.941596486070916e-06,
      "loss": 0.80778456,
      "num_input_tokens_seen": 18531575,
      "step": 872,
      "time_per_iteration": 2.6941590309143066
    },
    {
      "auxiliary_loss_clip": 0.01189214,
      "auxiliary_loss_mlp": 0.0104212,
      "balance_loss_clip": 1.07059467,
      "balance_loss_mlp": 1.03034759,
      "epoch": 0.10497204352792641,
      "flos": 27088747666560.0,
      "grad_norm": 4.50179509434207,
      "language_loss": 0.58819473,
      "learning_rate": 3.941409465519182e-06,
      "loss": 0.61050802,
      "num_input_tokens_seen": 18552100,
      "step": 873,
      "time_per_iteration": 2.6703388690948486
    },
    {
      "auxiliary_loss_clip": 0.01227285,
      "auxiliary_loss_mlp": 0.01050409,
      "balance_loss_clip": 1.06888366,
      "balance_loss_mlp": 1.03920925,
      "epoch": 0.10509228641856551,
      "flos": 32858479353600.0,
      "grad_norm": 1.6998899501423081,
      "language_loss": 0.85299563,
      "learning_rate": 3.941222150457176e-06,
      "loss": 0.8757726,
      "num_input_tokens_seen": 18575355,
      "step": 874,
      "time_per_iteration": 2.6571452617645264
    },
    {
      "auxiliary_loss_clip": 0.01241144,
      "auxiliary_loss_mlp": 0.01047844,
      "balance_loss_clip": 1.07096815,
      "balance_loss_mlp": 1.0365665,
      "epoch": 0.10521252930920459,
      "flos": 14319165173760.0,
      "grad_norm": 2.928694923809926,
      "language_loss": 0.71964639,
      "learning_rate": 3.941034540913311e-06,
      "loss": 0.74253625,
      "num_input_tokens_seen": 18592885,
      "step": 875,
      "time_per_iteration": 2.5020735263824463
    },
    {
      "auxiliary_loss_clip": 0.0123636,
      "auxiliary_loss_mlp": 0.00765104,
      "balance_loss_clip": 1.07246995,
      "balance_loss_mlp": 1.00009084,
      "epoch": 0.10533277219984369,
      "flos": 21687028773120.0,
      "grad_norm": 1.6717632620221499,
      "language_loss": 0.82586408,
      "learning_rate": 3.940846636916051e-06,
      "loss": 0.84587872,
      "num_input_tokens_seen": 18612920,
      "step": 876,
      "time_per_iteration": 3.2996864318847656
    },
    {
      "auxiliary_loss_clip": 0.01213933,
      "auxiliary_loss_mlp": 0.01049704,
      "balance_loss_clip": 1.07016122,
      "balance_loss_mlp": 1.03865361,
      "epoch": 0.10545301509048277,
      "flos": 22269787027200.0,
      "grad_norm": 2.10363794919024,
      "language_loss": 0.8630811,
      "learning_rate": 3.940658438493899e-06,
      "loss": 0.88571751,
      "num_input_tokens_seen": 18630765,
      "step": 877,
      "time_per_iteration": 2.541696071624756
    },
    {
      "auxiliary_loss_clip": 0.01249614,
      "auxiliary_loss_mlp": 0.01045978,
      "balance_loss_clip": 1.06987524,
      "balance_loss_mlp": 1.03397381,
      "epoch": 0.10557325798112187,
      "flos": 22199725549440.0,
      "grad_norm": 2.7504534556730604,
      "language_loss": 0.76005197,
      "learning_rate": 3.940469945675405e-06,
      "loss": 0.78300786,
      "num_input_tokens_seen": 18649150,
      "step": 878,
      "time_per_iteration": 2.485853672027588
    },
    {
      "auxiliary_loss_clip": 0.01164104,
      "auxiliary_loss_mlp": 0.01038212,
      "balance_loss_clip": 1.06195509,
      "balance_loss_mlp": 1.0279485,
      "epoch": 0.10569350087176095,
      "flos": 25775889569280.0,
      "grad_norm": 10.8469812207574,
      "language_loss": 0.91346717,
      "learning_rate": 3.940281158489163e-06,
      "loss": 0.93549031,
      "num_input_tokens_seen": 18668380,
      "step": 879,
      "time_per_iteration": 3.4364988803863525
    },
    {
      "auxiliary_loss_clip": 0.01166412,
      "auxiliary_loss_mlp": 0.01044486,
      "balance_loss_clip": 1.05651855,
      "balance_loss_mlp": 1.03283882,
      "epoch": 0.10581374376240005,
      "flos": 17311385790720.0,
      "grad_norm": 1.6466070118151486,
      "language_loss": 0.82759571,
      "learning_rate": 3.940092076963812e-06,
      "loss": 0.84970474,
      "num_input_tokens_seen": 18685875,
      "step": 880,
      "time_per_iteration": 2.651867151260376
    },
    {
      "auxiliary_loss_clip": 0.01215478,
      "auxiliary_loss_mlp": 0.01049756,
      "balance_loss_clip": 1.06338549,
      "balance_loss_mlp": 1.03774536,
      "epoch": 0.10593398665303914,
      "flos": 34349454017280.0,
      "grad_norm": 2.596769395717113,
      "language_loss": 0.78943717,
      "learning_rate": 3.9399027011280355e-06,
      "loss": 0.81208956,
      "num_input_tokens_seen": 18707970,
      "step": 881,
      "time_per_iteration": 3.5031917095184326
    },
    {
      "auxiliary_loss_clip": 0.01223363,
      "auxiliary_loss_mlp": 0.01046229,
      "balance_loss_clip": 1.0750134,
      "balance_loss_mlp": 1.03425407,
      "epoch": 0.10605422954367823,
      "flos": 23257977068160.0,
      "grad_norm": 1.9741042812167036,
      "language_loss": 0.77013195,
      "learning_rate": 3.939713031010561e-06,
      "loss": 0.7928279,
      "num_input_tokens_seen": 18726335,
      "step": 882,
      "time_per_iteration": 2.5636136531829834
    },
    {
      "auxiliary_loss_clip": 0.01203804,
      "auxiliary_loss_mlp": 0.01044784,
      "balance_loss_clip": 1.06870294,
      "balance_loss_mlp": 1.03313148,
      "epoch": 0.10617447243431732,
      "flos": 22820118278400.0,
      "grad_norm": 2.059907967737452,
      "language_loss": 0.77541101,
      "learning_rate": 3.939523066640163e-06,
      "loss": 0.79789686,
      "num_input_tokens_seen": 18745230,
      "step": 883,
      "time_per_iteration": 2.589808225631714
    },
    {
      "auxiliary_loss_clip": 0.01234299,
      "auxiliary_loss_mlp": 0.0103855,
      "balance_loss_clip": 1.07017171,
      "balance_loss_mlp": 1.02770185,
      "epoch": 0.10629471532495641,
      "flos": 24386577373440.0,
      "grad_norm": 1.839702999152693,
      "language_loss": 0.81147206,
      "learning_rate": 3.939332808045657e-06,
      "loss": 0.83420056,
      "num_input_tokens_seen": 18764880,
      "step": 884,
      "time_per_iteration": 2.563209295272827
    },
    {
      "auxiliary_loss_clip": 0.01204016,
      "auxiliary_loss_mlp": 0.01043246,
      "balance_loss_clip": 1.0670042,
      "balance_loss_mlp": 1.03236246,
      "epoch": 0.1064149582155955,
      "flos": 21105491581440.0,
      "grad_norm": 2.3814567433771683,
      "language_loss": 0.84671998,
      "learning_rate": 3.939142255255906e-06,
      "loss": 0.8691926,
      "num_input_tokens_seen": 18785765,
      "step": 885,
      "time_per_iteration": 2.611672878265381
    },
    {
      "auxiliary_loss_clip": 0.01238769,
      "auxiliary_loss_mlp": 0.01041848,
      "balance_loss_clip": 1.07610118,
      "balance_loss_mlp": 1.03083873,
      "epoch": 0.1065352011062346,
      "flos": 20702035042560.0,
      "grad_norm": 1.773633803805175,
      "language_loss": 0.86951971,
      "learning_rate": 3.938951408299817e-06,
      "loss": 0.89232588,
      "num_input_tokens_seen": 18804605,
      "step": 886,
      "time_per_iteration": 2.534438133239746
    },
    {
      "auxiliary_loss_clip": 0.01097505,
      "auxiliary_loss_mlp": 0.01013407,
      "balance_loss_clip": 1.04948783,
      "balance_loss_mlp": 1.01011682,
      "epoch": 0.10665544399687368,
      "flos": 62659632689280.0,
      "grad_norm": 0.8029375908190648,
      "language_loss": 0.54455644,
      "learning_rate": 3.938760267206342e-06,
      "loss": 0.56566554,
      "num_input_tokens_seen": 18866425,
      "step": 887,
      "time_per_iteration": 3.328860282897949
    },
    {
      "auxiliary_loss_clip": 0.01249148,
      "auxiliary_loss_mlp": 0.01046916,
      "balance_loss_clip": 1.0725491,
      "balance_loss_mlp": 1.0360682,
      "epoch": 0.10677568688751278,
      "flos": 26140382830080.0,
      "grad_norm": 2.333280005403275,
      "language_loss": 0.78506947,
      "learning_rate": 3.938568832004475e-06,
      "loss": 0.80803013,
      "num_input_tokens_seen": 18885130,
      "step": 888,
      "time_per_iteration": 2.669353723526001
    },
    {
      "auxiliary_loss_clip": 0.01208806,
      "auxiliary_loss_mlp": 0.01043293,
      "balance_loss_clip": 1.06697381,
      "balance_loss_mlp": 1.03275442,
      "epoch": 0.10689592977815186,
      "flos": 12786533712000.0,
      "grad_norm": 1.9262929940585052,
      "language_loss": 0.75506639,
      "learning_rate": 3.938377102723257e-06,
      "loss": 0.77758735,
      "num_input_tokens_seen": 18902265,
      "step": 889,
      "time_per_iteration": 2.550930976867676
    },
    {
      "auxiliary_loss_clip": 0.01168247,
      "auxiliary_loss_mlp": 0.01046813,
      "balance_loss_clip": 1.05949044,
      "balance_loss_mlp": 1.03487372,
      "epoch": 0.10701617266879096,
      "flos": 22126683242880.0,
      "grad_norm": 2.0025631197032467,
      "language_loss": 0.83228242,
      "learning_rate": 3.938185079391774e-06,
      "loss": 0.85443306,
      "num_input_tokens_seen": 18919310,
      "step": 890,
      "time_per_iteration": 2.643177032470703
    },
    {
      "auxiliary_loss_clip": 0.01248805,
      "auxiliary_loss_mlp": 0.01038635,
      "balance_loss_clip": 1.0724628,
      "balance_loss_mlp": 1.02796006,
      "epoch": 0.10713641555943004,
      "flos": 19745625559680.0,
      "grad_norm": 3.5349919053328205,
      "language_loss": 1.0572294,
      "learning_rate": 3.937992762039157e-06,
      "loss": 1.08010375,
      "num_input_tokens_seen": 18932635,
      "step": 891,
      "time_per_iteration": 2.474825143814087
    },
    {
      "auxiliary_loss_clip": 0.01231377,
      "auxiliary_loss_mlp": 0.01046879,
      "balance_loss_clip": 1.0717119,
      "balance_loss_mlp": 1.03645396,
      "epoch": 0.10725665845006914,
      "flos": 23952992302080.0,
      "grad_norm": 1.6677336179866498,
      "language_loss": 0.80502594,
      "learning_rate": 3.937800150694577e-06,
      "loss": 0.8278085,
      "num_input_tokens_seen": 18953810,
      "step": 892,
      "time_per_iteration": 2.552788019180298
    },
    {
      "auxiliary_loss_clip": 0.01191222,
      "auxiliary_loss_mlp": 0.01053589,
      "balance_loss_clip": 1.06718135,
      "balance_loss_mlp": 1.04247904,
      "epoch": 0.10737690134070824,
      "flos": 18551704371840.0,
      "grad_norm": 2.1736515174268383,
      "language_loss": 0.75921512,
      "learning_rate": 3.937607245387255e-06,
      "loss": 0.7816633,
      "num_input_tokens_seen": 18973175,
      "step": 893,
      "time_per_iteration": 2.647382974624634
    },
    {
      "auxiliary_loss_clip": 0.01222079,
      "auxiliary_loss_mlp": 0.01048056,
      "balance_loss_clip": 1.06548119,
      "balance_loss_mlp": 1.03822136,
      "epoch": 0.10749714423134732,
      "flos": 22707609903360.0,
      "grad_norm": 2.1539651830437845,
      "language_loss": 0.72253096,
      "learning_rate": 3.937414046146455e-06,
      "loss": 0.74523234,
      "num_input_tokens_seen": 18991130,
      "step": 894,
      "time_per_iteration": 2.5574097633361816
    },
    {
      "auxiliary_loss_clip": 0.01250393,
      "auxiliary_loss_mlp": 0.01057096,
      "balance_loss_clip": 1.07312894,
      "balance_loss_mlp": 1.04552054,
      "epoch": 0.10761738712198642,
      "flos": 21106066199040.0,
      "grad_norm": 2.4015089977165314,
      "language_loss": 0.75854194,
      "learning_rate": 3.9372205530014845e-06,
      "loss": 0.78161681,
      "num_input_tokens_seen": 19009610,
      "step": 895,
      "time_per_iteration": 2.5158603191375732
    },
    {
      "auxiliary_loss_clip": 0.01250407,
      "auxiliary_loss_mlp": 0.01047045,
      "balance_loss_clip": 1.07064474,
      "balance_loss_mlp": 1.03666782,
      "epoch": 0.1077376300126255,
      "flos": 23766723348480.0,
      "grad_norm": 1.9520468563148006,
      "language_loss": 0.71447635,
      "learning_rate": 3.937026765981696e-06,
      "loss": 0.7374509,
      "num_input_tokens_seen": 19029680,
      "step": 896,
      "time_per_iteration": 2.5126612186431885
    },
    {
      "auxiliary_loss_clip": 0.01206343,
      "auxiliary_loss_mlp": 0.010473,
      "balance_loss_clip": 1.06934905,
      "balance_loss_mlp": 1.03596282,
      "epoch": 0.1078578729032646,
      "flos": 20919581763840.0,
      "grad_norm": 2.4407745960401113,
      "language_loss": 0.79492223,
      "learning_rate": 3.936832685116488e-06,
      "loss": 0.81745869,
      "num_input_tokens_seen": 19047775,
      "step": 897,
      "time_per_iteration": 2.5977816581726074
    },
    {
      "auxiliary_loss_clip": 0.01248,
      "auxiliary_loss_mlp": 0.01035667,
      "balance_loss_clip": 1.07149601,
      "balance_loss_mlp": 1.02475977,
      "epoch": 0.10797811579390369,
      "flos": 14829886702080.0,
      "grad_norm": 2.028315717356634,
      "language_loss": 0.89903009,
      "learning_rate": 3.936638310435301e-06,
      "loss": 0.92186677,
      "num_input_tokens_seen": 19065640,
      "step": 898,
      "time_per_iteration": 2.4515583515167236
    },
    {
      "auxiliary_loss_clip": 0.01239838,
      "auxiliary_loss_mlp": 0.0103736,
      "balance_loss_clip": 1.07089341,
      "balance_loss_mlp": 1.02591038,
      "epoch": 0.10809835868454278,
      "flos": 19536985411200.0,
      "grad_norm": 1.7471263731748448,
      "language_loss": 0.81238055,
      "learning_rate": 3.936443641967623e-06,
      "loss": 0.83515257,
      "num_input_tokens_seen": 19084470,
      "step": 899,
      "time_per_iteration": 2.5281920433044434
    },
    {
      "auxiliary_loss_clip": 0.01223425,
      "auxiliary_loss_mlp": 0.01043294,
      "balance_loss_clip": 1.07012331,
      "balance_loss_mlp": 1.03245139,
      "epoch": 0.10821860157518187,
      "flos": 18442320480000.0,
      "grad_norm": 2.3373922167157,
      "language_loss": 0.83319074,
      "learning_rate": 3.936248679742983e-06,
      "loss": 0.85585791,
      "num_input_tokens_seen": 19102965,
      "step": 900,
      "time_per_iteration": 2.5299389362335205
    },
    {
      "auxiliary_loss_clip": 0.01098339,
      "auxiliary_loss_mlp": 0.01008545,
      "balance_loss_clip": 1.03250122,
      "balance_loss_mlp": 1.00520742,
      "epoch": 0.10833884446582095,
      "flos": 49359468447360.0,
      "grad_norm": 1.0472724456862286,
      "language_loss": 0.7019968,
      "learning_rate": 3.936053423790959e-06,
      "loss": 0.72306561,
      "num_input_tokens_seen": 19151285,
      "step": 901,
      "time_per_iteration": 2.9408888816833496
    },
    {
      "auxiliary_loss_clip": 0.01246435,
      "auxiliary_loss_mlp": 0.01042367,
      "balance_loss_clip": 1.07119405,
      "balance_loss_mlp": 1.03173327,
      "epoch": 0.10845908735646005,
      "flos": 20411912891520.0,
      "grad_norm": 1.7430881584978353,
      "language_loss": 0.77579689,
      "learning_rate": 3.935857874141168e-06,
      "loss": 0.7986849,
      "num_input_tokens_seen": 19170120,
      "step": 902,
      "time_per_iteration": 3.32848858833313
    },
    {
      "auxiliary_loss_clip": 0.01208899,
      "auxiliary_loss_mlp": 0.01042285,
      "balance_loss_clip": 1.06662869,
      "balance_loss_mlp": 1.03148472,
      "epoch": 0.10857933024709913,
      "flos": 14027750133120.0,
      "grad_norm": 2.2394725777520694,
      "language_loss": 0.83277917,
      "learning_rate": 3.935662030823279e-06,
      "loss": 0.85529101,
      "num_input_tokens_seen": 19186305,
      "step": 903,
      "time_per_iteration": 2.501368522644043
    },
    {
      "auxiliary_loss_clip": 0.01237746,
      "auxiliary_loss_mlp": 0.01039626,
      "balance_loss_clip": 1.07016683,
      "balance_loss_mlp": 1.02827084,
      "epoch": 0.10869957313773823,
      "flos": 13369004657280.0,
      "grad_norm": 2.6532362540217087,
      "language_loss": 0.72506303,
      "learning_rate": 3.935465893866998e-06,
      "loss": 0.74783671,
      "num_input_tokens_seen": 19204530,
      "step": 904,
      "time_per_iteration": 2.5257515907287598
    },
    {
      "auxiliary_loss_clip": 0.01218221,
      "auxiliary_loss_mlp": 0.01042324,
      "balance_loss_clip": 1.07023072,
      "balance_loss_mlp": 1.03169703,
      "epoch": 0.10881981602837733,
      "flos": 25807095509760.0,
      "grad_norm": 1.9888446048696702,
      "language_loss": 0.80032659,
      "learning_rate": 3.935269463302079e-06,
      "loss": 0.82293206,
      "num_input_tokens_seen": 19222735,
      "step": 905,
      "time_per_iteration": 4.054166555404663
    },
    {
      "auxiliary_loss_clip": 0.01239324,
      "auxiliary_loss_mlp": 0.01040748,
      "balance_loss_clip": 1.0718143,
      "balance_loss_mlp": 1.02771807,
      "epoch": 0.10894005891901641,
      "flos": 20777555387520.0,
      "grad_norm": 1.978536199947901,
      "language_loss": 0.76860577,
      "learning_rate": 3.935072739158322e-06,
      "loss": 0.79140645,
      "num_input_tokens_seen": 19242445,
      "step": 906,
      "time_per_iteration": 2.525634527206421
    },
    {
      "auxiliary_loss_clip": 0.01217747,
      "auxiliary_loss_mlp": 0.01044287,
      "balance_loss_clip": 1.06756938,
      "balance_loss_mlp": 1.03344536,
      "epoch": 0.10906030180965551,
      "flos": 26649883296000.0,
      "grad_norm": 1.513345247887908,
      "language_loss": 0.7973218,
      "learning_rate": 3.934875721465569e-06,
      "loss": 0.81994212,
      "num_input_tokens_seen": 19262865,
      "step": 907,
      "time_per_iteration": 3.4201667308807373
    },
    {
      "auxiliary_loss_clip": 0.01215738,
      "auxiliary_loss_mlp": 0.01049837,
      "balance_loss_clip": 1.06522679,
      "balance_loss_mlp": 1.0379281,
      "epoch": 0.10918054470029459,
      "flos": 36534402420480.0,
      "grad_norm": 2.3763718207639606,
      "language_loss": 0.71539545,
      "learning_rate": 3.9346784102537076e-06,
      "loss": 0.73805118,
      "num_input_tokens_seen": 19285000,
      "step": 908,
      "time_per_iteration": 2.657378673553467
    },
    {
      "auxiliary_loss_clip": 0.01243968,
      "auxiliary_loss_mlp": 0.01045357,
      "balance_loss_clip": 1.06807375,
      "balance_loss_mlp": 1.03514671,
      "epoch": 0.10930078759093369,
      "flos": 21762549118080.0,
      "grad_norm": 1.814088302175646,
      "language_loss": 0.78386641,
      "learning_rate": 3.934480805552669e-06,
      "loss": 0.8067596,
      "num_input_tokens_seen": 19306010,
      "step": 909,
      "time_per_iteration": 2.504197597503662
    },
    {
      "auxiliary_loss_clip": 0.01246513,
      "auxiliary_loss_mlp": 0.00763976,
      "balance_loss_clip": 1.07114446,
      "balance_loss_mlp": 1.00010991,
      "epoch": 0.10942103048157277,
      "flos": 22601781457920.0,
      "grad_norm": 2.0383579493322004,
      "language_loss": 0.87991786,
      "learning_rate": 3.93428290739243e-06,
      "loss": 0.90002275,
      "num_input_tokens_seen": 19325380,
      "step": 910,
      "time_per_iteration": 2.487342119216919
    },
    {
      "auxiliary_loss_clip": 0.01217432,
      "auxiliary_loss_mlp": 0.01035906,
      "balance_loss_clip": 1.0677706,
      "balance_loss_mlp": 1.02446222,
      "epoch": 0.10954127337221187,
      "flos": 15045781397760.0,
      "grad_norm": 2.232268071685451,
      "language_loss": 0.79870701,
      "learning_rate": 3.9340847158030125e-06,
      "loss": 0.82124031,
      "num_input_tokens_seen": 19338960,
      "step": 911,
      "time_per_iteration": 2.5157434940338135
    },
    {
      "auxiliary_loss_clip": 0.01234488,
      "auxiliary_loss_mlp": 0.01047605,
      "balance_loss_clip": 1.06924248,
      "balance_loss_mlp": 1.03698325,
      "epoch": 0.10966151626285096,
      "flos": 21650974496640.0,
      "grad_norm": 1.853598038185026,
      "language_loss": 0.75309628,
      "learning_rate": 3.9338862308144814e-06,
      "loss": 0.77591711,
      "num_input_tokens_seen": 19357780,
      "step": 912,
      "time_per_iteration": 2.502610683441162
    },
    {
      "auxiliary_loss_clip": 0.01248307,
      "auxiliary_loss_mlp": 0.0104701,
      "balance_loss_clip": 1.07165384,
      "balance_loss_mlp": 1.03662121,
      "epoch": 0.10978175915349005,
      "flos": 20121359777280.0,
      "grad_norm": 1.6280666877085248,
      "language_loss": 0.84357464,
      "learning_rate": 3.933687452456946e-06,
      "loss": 0.86652786,
      "num_input_tokens_seen": 19377680,
      "step": 913,
      "time_per_iteration": 2.4641408920288086
    },
    {
      "auxiliary_loss_clip": 0.0119383,
      "auxiliary_loss_mlp": 0.01037327,
      "balance_loss_clip": 1.0591464,
      "balance_loss_mlp": 1.02501249,
      "epoch": 0.10990200204412914,
      "flos": 20412667077120.0,
      "grad_norm": 4.03532649811657,
      "language_loss": 0.86149353,
      "learning_rate": 3.933488380760562e-06,
      "loss": 0.88380516,
      "num_input_tokens_seen": 19397040,
      "step": 914,
      "time_per_iteration": 2.5571024417877197
    },
    {
      "auxiliary_loss_clip": 0.01248567,
      "auxiliary_loss_mlp": 0.00764528,
      "balance_loss_clip": 1.07145095,
      "balance_loss_mlp": 1.00008321,
      "epoch": 0.11002224493476823,
      "flos": 17530117660800.0,
      "grad_norm": 1.9794344043463137,
      "language_loss": 0.87152588,
      "learning_rate": 3.9332890157555286e-06,
      "loss": 0.89165682,
      "num_input_tokens_seen": 19413975,
      "step": 915,
      "time_per_iteration": 2.4526405334472656
    },
    {
      "auxiliary_loss_clip": 0.01221672,
      "auxiliary_loss_mlp": 0.0104699,
      "balance_loss_clip": 1.0684762,
      "balance_loss_mlp": 1.03568292,
      "epoch": 0.11014248782540732,
      "flos": 12203093099520.0,
      "grad_norm": 2.289663525020898,
      "language_loss": 0.76395983,
      "learning_rate": 3.933089357472088e-06,
      "loss": 0.78664649,
      "num_input_tokens_seen": 19432005,
      "step": 916,
      "time_per_iteration": 2.4978785514831543
    },
    {
      "auxiliary_loss_clip": 0.01249824,
      "auxiliary_loss_mlp": 0.01044248,
      "balance_loss_clip": 1.0734427,
      "balance_loss_mlp": 1.03328633,
      "epoch": 0.11026273071604642,
      "flos": 22382977760640.0,
      "grad_norm": 1.7778952291835857,
      "language_loss": 0.85864955,
      "learning_rate": 3.932889405940529e-06,
      "loss": 0.88159031,
      "num_input_tokens_seen": 19450100,
      "step": 917,
      "time_per_iteration": 2.488654136657715
    },
    {
      "auxiliary_loss_clip": 0.01223596,
      "auxiliary_loss_mlp": 0.01042277,
      "balance_loss_clip": 1.07454467,
      "balance_loss_mlp": 1.03145242,
      "epoch": 0.1103829736066855,
      "flos": 19829046896640.0,
      "grad_norm": 2.164698477174344,
      "language_loss": 0.79746985,
      "learning_rate": 3.932689161191184e-06,
      "loss": 0.82012856,
      "num_input_tokens_seen": 19467805,
      "step": 918,
      "time_per_iteration": 2.527545928955078
    },
    {
      "auxiliary_loss_clip": 0.01230893,
      "auxiliary_loss_mlp": 0.01043044,
      "balance_loss_clip": 1.06721413,
      "balance_loss_mlp": 1.03174317,
      "epoch": 0.1105032164973246,
      "flos": 22669616292480.0,
      "grad_norm": 2.9836227759873966,
      "language_loss": 0.87975824,
      "learning_rate": 3.93248862325443e-06,
      "loss": 0.90249759,
      "num_input_tokens_seen": 19486710,
      "step": 919,
      "time_per_iteration": 2.527946949005127
    },
    {
      "auxiliary_loss_clip": 0.01130113,
      "auxiliary_loss_mlp": 0.01007582,
      "balance_loss_clip": 1.03725338,
      "balance_loss_mlp": 1.00422049,
      "epoch": 0.11062345938796368,
      "flos": 66483507876480.0,
      "grad_norm": 0.9480432606464496,
      "language_loss": 0.6446889,
      "learning_rate": 3.932287792160688e-06,
      "loss": 0.66606587,
      "num_input_tokens_seen": 19545170,
      "step": 920,
      "time_per_iteration": 3.0086216926574707
    },
    {
      "auxiliary_loss_clip": 0.01233536,
      "auxiliary_loss_mlp": 0.01057737,
      "balance_loss_clip": 1.0672487,
      "balance_loss_mlp": 1.04593492,
      "epoch": 0.11074370227860278,
      "flos": 21907771804800.0,
      "grad_norm": 2.3336433656492157,
      "language_loss": 0.80610073,
      "learning_rate": 3.932086667940424e-06,
      "loss": 0.82901341,
      "num_input_tokens_seen": 19561875,
      "step": 921,
      "time_per_iteration": 2.5176546573638916
    },
    {
      "auxiliary_loss_clip": 0.01232069,
      "auxiliary_loss_mlp": 0.00764056,
      "balance_loss_clip": 1.07112169,
      "balance_loss_mlp": 1.00010467,
      "epoch": 0.11086394516924186,
      "flos": 28658115763200.0,
      "grad_norm": 1.7945901445490888,
      "language_loss": 0.81595868,
      "learning_rate": 3.93188525062415e-06,
      "loss": 0.83591998,
      "num_input_tokens_seen": 19582340,
      "step": 922,
      "time_per_iteration": 2.5978198051452637
    },
    {
      "auxiliary_loss_clip": 0.0123443,
      "auxiliary_loss_mlp": 0.01047624,
      "balance_loss_clip": 1.06924939,
      "balance_loss_mlp": 1.03533983,
      "epoch": 0.11098418805988096,
      "flos": 24535247765760.0,
      "grad_norm": 2.0473126143712803,
      "language_loss": 0.85970914,
      "learning_rate": 3.931683540242418e-06,
      "loss": 0.88252968,
      "num_input_tokens_seen": 19603405,
      "step": 923,
      "time_per_iteration": 2.57163143157959
    },
    {
      "auxiliary_loss_clip": 0.01228415,
      "auxiliary_loss_mlp": 0.01044187,
      "balance_loss_clip": 1.06761456,
      "balance_loss_mlp": 1.03316021,
      "epoch": 0.11110443095052006,
      "flos": 22960384888320.0,
      "grad_norm": 2.5377431758959603,
      "language_loss": 0.9105643,
      "learning_rate": 3.9314815368258295e-06,
      "loss": 0.93329036,
      "num_input_tokens_seen": 19619885,
      "step": 924,
      "time_per_iteration": 2.5258777141571045
    },
    {
      "auxiliary_loss_clip": 0.01238387,
      "auxiliary_loss_mlp": 0.01038982,
      "balance_loss_clip": 1.07390344,
      "balance_loss_mlp": 1.02868247,
      "epoch": 0.11122467384115914,
      "flos": 18950025265920.0,
      "grad_norm": 1.7252943364941904,
      "language_loss": 0.78803492,
      "learning_rate": 3.9312792404050275e-06,
      "loss": 0.81080866,
      "num_input_tokens_seen": 19637940,
      "step": 925,
      "time_per_iteration": 2.5064358711242676
    },
    {
      "auxiliary_loss_clip": 0.01244987,
      "auxiliary_loss_mlp": 0.01047185,
      "balance_loss_clip": 1.07118607,
      "balance_loss_mlp": 1.03727221,
      "epoch": 0.11134491673179824,
      "flos": 25082957324160.0,
      "grad_norm": 5.115479374243775,
      "language_loss": 0.77323759,
      "learning_rate": 3.9310766510107e-06,
      "loss": 0.79615927,
      "num_input_tokens_seen": 19657115,
      "step": 926,
      "time_per_iteration": 2.513786554336548
    },
    {
      "auxiliary_loss_clip": 0.01204415,
      "auxiliary_loss_mlp": 0.01046335,
      "balance_loss_clip": 1.06521225,
      "balance_loss_mlp": 1.03349042,
      "epoch": 0.11146515962243732,
      "flos": 24499121662080.0,
      "grad_norm": 1.8579085080426099,
      "language_loss": 0.9196015,
      "learning_rate": 3.9308737686735806e-06,
      "loss": 0.94210899,
      "num_input_tokens_seen": 19677075,
      "step": 927,
      "time_per_iteration": 2.6081809997558594
    },
    {
      "auxiliary_loss_clip": 0.0124958,
      "auxiliary_loss_mlp": 0.01050704,
      "balance_loss_clip": 1.07328749,
      "balance_loss_mlp": 1.03933096,
      "epoch": 0.11158540251307641,
      "flos": 22343763087360.0,
      "grad_norm": 2.2441634463238933,
      "language_loss": 0.82748544,
      "learning_rate": 3.9306705934244455e-06,
      "loss": 0.85048831,
      "num_input_tokens_seen": 19697155,
      "step": 928,
      "time_per_iteration": 3.2197580337524414
    },
    {
      "auxiliary_loss_clip": 0.01204324,
      "auxiliary_loss_mlp": 0.01050717,
      "balance_loss_clip": 1.06646705,
      "balance_loss_mlp": 1.04034007,
      "epoch": 0.11170564540371551,
      "flos": 19902304684800.0,
      "grad_norm": 1.9821112777867815,
      "language_loss": 0.88482773,
      "learning_rate": 3.930467125294116e-06,
      "loss": 0.90737808,
      "num_input_tokens_seen": 19716705,
      "step": 929,
      "time_per_iteration": 2.556999683380127
    },
    {
      "auxiliary_loss_clip": 0.01069391,
      "auxiliary_loss_mlp": 0.01007343,
      "balance_loss_clip": 1.03120768,
      "balance_loss_mlp": 1.00386238,
      "epoch": 0.1118258882943546,
      "flos": 64586239499520.0,
      "grad_norm": 0.9265973861644086,
      "language_loss": 0.60494685,
      "learning_rate": 3.930263364313458e-06,
      "loss": 0.62571418,
      "num_input_tokens_seen": 19767275,
      "step": 930,
      "time_per_iteration": 3.0353214740753174
    },
    {
      "auxiliary_loss_clip": 0.01201746,
      "auxiliary_loss_mlp": 0.01042379,
      "balance_loss_clip": 1.06627882,
      "balance_loss_mlp": 1.03148973,
      "epoch": 0.11194613118499369,
      "flos": 17201965985280.0,
      "grad_norm": 1.886443152541927,
      "language_loss": 0.83034509,
      "learning_rate": 3.930059310513384e-06,
      "loss": 0.85278636,
      "num_input_tokens_seen": 19786315,
      "step": 931,
      "time_per_iteration": 3.982970714569092
    },
    {
      "auxiliary_loss_clip": 0.01188293,
      "auxiliary_loss_mlp": 0.00764389,
      "balance_loss_clip": 1.0641613,
      "balance_loss_mlp": 1.00009131,
      "epoch": 0.11206637407563277,
      "flos": 31863465728640.0,
      "grad_norm": 1.8127654205846255,
      "language_loss": 0.83984828,
      "learning_rate": 3.929854963924846e-06,
      "loss": 0.85937512,
      "num_input_tokens_seen": 19806580,
      "step": 932,
      "time_per_iteration": 2.6635072231292725
    },
    {
      "auxiliary_loss_clip": 0.01199909,
      "auxiliary_loss_mlp": 0.0104049,
      "balance_loss_clip": 1.06283259,
      "balance_loss_mlp": 1.03011894,
      "epoch": 0.11218661696627187,
      "flos": 21945621761280.0,
      "grad_norm": 1.9112604564073543,
      "language_loss": 0.77346182,
      "learning_rate": 3.929650324578845e-06,
      "loss": 0.79586577,
      "num_input_tokens_seen": 19826045,
      "step": 933,
      "time_per_iteration": 2.5690670013427734
    },
    {
      "auxiliary_loss_clip": 0.01217117,
      "auxiliary_loss_mlp": 0.01048947,
      "balance_loss_clip": 1.0674305,
      "balance_loss_mlp": 1.03661537,
      "epoch": 0.11230685985691095,
      "flos": 25878198481920.0,
      "grad_norm": 3.5620467060466203,
      "language_loss": 0.82116652,
      "learning_rate": 3.929445392506423e-06,
      "loss": 0.84382713,
      "num_input_tokens_seen": 19843985,
      "step": 934,
      "time_per_iteration": 3.3918347358703613
    },
    {
      "auxiliary_loss_clip": 0.01234024,
      "auxiliary_loss_mlp": 0.01045207,
      "balance_loss_clip": 1.07558608,
      "balance_loss_mlp": 1.03468692,
      "epoch": 0.11242710274755005,
      "flos": 22231506107520.0,
      "grad_norm": 1.9692801396890847,
      "language_loss": 0.76019627,
      "learning_rate": 3.92924016773867e-06,
      "loss": 0.78298855,
      "num_input_tokens_seen": 19860480,
      "step": 935,
      "time_per_iteration": 2.5194613933563232
    },
    {
      "auxiliary_loss_clip": 0.0121292,
      "auxiliary_loss_mlp": 0.00763087,
      "balance_loss_clip": 1.06406093,
      "balance_loss_mlp": 1.00007117,
      "epoch": 0.11254734563818915,
      "flos": 17712184723200.0,
      "grad_norm": 2.277876458523468,
      "language_loss": 0.7327826,
      "learning_rate": 3.9290346503067175e-06,
      "loss": 0.75254261,
      "num_input_tokens_seen": 19877145,
      "step": 936,
      "time_per_iteration": 2.5319368839263916
    },
    {
      "auxiliary_loss_clip": 0.01233592,
      "auxiliary_loss_mlp": 0.01048839,
      "balance_loss_clip": 1.06894588,
      "balance_loss_mlp": 1.03779435,
      "epoch": 0.11266758852882823,
      "flos": 54930397334400.0,
      "grad_norm": 1.737440878259158,
      "language_loss": 0.78951848,
      "learning_rate": 3.9288288402417415e-06,
      "loss": 0.81234282,
      "num_input_tokens_seen": 19903405,
      "step": 937,
      "time_per_iteration": 2.8238298892974854
    },
    {
      "auxiliary_loss_clip": 0.01236841,
      "auxiliary_loss_mlp": 0.01042997,
      "balance_loss_clip": 1.07390261,
      "balance_loss_mlp": 1.03097486,
      "epoch": 0.11278783141946733,
      "flos": 18878132194560.0,
      "grad_norm": 2.259250823817212,
      "language_loss": 0.70620245,
      "learning_rate": 3.928622737574964e-06,
      "loss": 0.72900081,
      "num_input_tokens_seen": 19918740,
      "step": 938,
      "time_per_iteration": 2.4704511165618896
    },
    {
      "auxiliary_loss_clip": 0.01216082,
      "auxiliary_loss_mlp": 0.01050647,
      "balance_loss_clip": 1.06690454,
      "balance_loss_mlp": 1.04009724,
      "epoch": 0.11290807431010641,
      "flos": 26469252777600.0,
      "grad_norm": 1.9025600073114202,
      "language_loss": 0.90966594,
      "learning_rate": 3.928416342337652e-06,
      "loss": 0.93233329,
      "num_input_tokens_seen": 19938475,
      "step": 939,
      "time_per_iteration": 2.5807857513427734
    },
    {
      "auxiliary_loss_clip": 0.01219274,
      "auxiliary_loss_mlp": 0.01047018,
      "balance_loss_clip": 1.06977141,
      "balance_loss_mlp": 1.03699839,
      "epoch": 0.1130283172007455,
      "flos": 22710590732160.0,
      "grad_norm": 1.8355422218335449,
      "language_loss": 0.82640743,
      "learning_rate": 3.928209654561113e-06,
      "loss": 0.84907043,
      "num_input_tokens_seen": 19959310,
      "step": 940,
      "time_per_iteration": 2.5475242137908936
    },
    {
      "auxiliary_loss_clip": 0.01210881,
      "auxiliary_loss_mlp": 0.01051042,
      "balance_loss_clip": 1.06972957,
      "balance_loss_mlp": 1.0413444,
      "epoch": 0.1131485600913846,
      "flos": 23219911630080.0,
      "grad_norm": 2.4165339146653837,
      "language_loss": 0.81280476,
      "learning_rate": 3.928002674276703e-06,
      "loss": 0.83542395,
      "num_input_tokens_seen": 19978700,
      "step": 941,
      "time_per_iteration": 2.561915636062622
    },
    {
      "auxiliary_loss_clip": 0.01160727,
      "auxiliary_loss_mlp": 0.0103903,
      "balance_loss_clip": 1.05465233,
      "balance_loss_mlp": 1.02787828,
      "epoch": 0.11326880298202369,
      "flos": 14064271286400.0,
      "grad_norm": 2.277653261927664,
      "language_loss": 0.75449318,
      "learning_rate": 3.92779540151582e-06,
      "loss": 0.77649075,
      "num_input_tokens_seen": 19995785,
      "step": 942,
      "time_per_iteration": 2.5538668632507324
    },
    {
      "auxiliary_loss_clip": 0.01214587,
      "auxiliary_loss_mlp": 0.01047391,
      "balance_loss_clip": 1.06747031,
      "balance_loss_mlp": 1.03654921,
      "epoch": 0.11338904587266278,
      "flos": 16325386479360.0,
      "grad_norm": 1.7691617912251245,
      "language_loss": 0.85655558,
      "learning_rate": 3.927587836309907e-06,
      "loss": 0.87917536,
      "num_input_tokens_seen": 20013615,
      "step": 943,
      "time_per_iteration": 2.5838091373443604
    },
    {
      "auxiliary_loss_clip": 0.01207,
      "auxiliary_loss_mlp": 0.01049268,
      "balance_loss_clip": 1.06298161,
      "balance_loss_mlp": 1.03833663,
      "epoch": 0.11350928876330187,
      "flos": 24426258923520.0,
      "grad_norm": 14.847520825832634,
      "language_loss": 0.78070915,
      "learning_rate": 3.927379978690452e-06,
      "loss": 0.80327189,
      "num_input_tokens_seen": 20032880,
      "step": 944,
      "time_per_iteration": 2.5879645347595215
    },
    {
      "auxiliary_loss_clip": 0.01187581,
      "auxiliary_loss_mlp": 0.01042993,
      "balance_loss_clip": 1.05752409,
      "balance_loss_mlp": 1.03296208,
      "epoch": 0.11362953165394096,
      "flos": 24497074586880.0,
      "grad_norm": 2.6487092013253197,
      "language_loss": 0.87392259,
      "learning_rate": 3.927171828688987e-06,
      "loss": 0.89622831,
      "num_input_tokens_seen": 20052405,
      "step": 945,
      "time_per_iteration": 2.6133930683135986
    },
    {
      "auxiliary_loss_clip": 0.01243782,
      "auxiliary_loss_mlp": 0.01036419,
      "balance_loss_clip": 1.07018054,
      "balance_loss_mlp": 1.02614963,
      "epoch": 0.11374977454458005,
      "flos": 24060831909120.0,
      "grad_norm": 2.312353961959972,
      "language_loss": 0.82289994,
      "learning_rate": 3.926963386337088e-06,
      "loss": 0.84570193,
      "num_input_tokens_seen": 20070635,
      "step": 946,
      "time_per_iteration": 2.535149097442627
    },
    {
      "auxiliary_loss_clip": 0.01251538,
      "auxiliary_loss_mlp": 0.01042055,
      "balance_loss_clip": 1.07483101,
      "balance_loss_mlp": 1.03039074,
      "epoch": 0.11387001743521914,
      "flos": 39457638967680.0,
      "grad_norm": 2.2371452498724365,
      "language_loss": 0.70287466,
      "learning_rate": 3.926754651666375e-06,
      "loss": 0.72581053,
      "num_input_tokens_seen": 20091195,
      "step": 947,
      "time_per_iteration": 2.6291024684906006
    },
    {
      "auxiliary_loss_clip": 0.01201283,
      "auxiliary_loss_mlp": 0.01045291,
      "balance_loss_clip": 1.06741762,
      "balance_loss_mlp": 1.03469276,
      "epoch": 0.11399026032585824,
      "flos": 25082454533760.0,
      "grad_norm": 2.4956148790224066,
      "language_loss": 0.78069115,
      "learning_rate": 3.926545624708513e-06,
      "loss": 0.80315685,
      "num_input_tokens_seen": 20110435,
      "step": 948,
      "time_per_iteration": 2.598841667175293
    },
    {
      "auxiliary_loss_clip": 0.01199208,
      "auxiliary_loss_mlp": 0.01041078,
      "balance_loss_clip": 1.06460309,
      "balance_loss_mlp": 1.03085542,
      "epoch": 0.11411050321649732,
      "flos": 17961835224960.0,
      "grad_norm": 2.0515818943790105,
      "language_loss": 0.85824728,
      "learning_rate": 3.926336305495213e-06,
      "loss": 0.88065004,
      "num_input_tokens_seen": 20128995,
      "step": 949,
      "time_per_iteration": 2.575958490371704
    },
    {
      "auxiliary_loss_clip": 0.01186199,
      "auxiliary_loss_mlp": 0.01043695,
      "balance_loss_clip": 1.06077933,
      "balance_loss_mlp": 1.03335392,
      "epoch": 0.11423074610713642,
      "flos": 22455409536000.0,
      "grad_norm": 3.102117660624367,
      "language_loss": 0.88800848,
      "learning_rate": 3.926126694058226e-06,
      "loss": 0.91030741,
      "num_input_tokens_seen": 20148145,
      "step": 950,
      "time_per_iteration": 2.5694007873535156
    },
    {
      "auxiliary_loss_clip": 0.01187571,
      "auxiliary_loss_mlp": 0.01051143,
      "balance_loss_clip": 1.06797028,
      "balance_loss_mlp": 1.04102802,
      "epoch": 0.1143509889977755,
      "flos": 19717687756800.0,
      "grad_norm": 1.432414656962617,
      "language_loss": 0.82136917,
      "learning_rate": 3.92591679042935e-06,
      "loss": 0.8437562,
      "num_input_tokens_seen": 20168035,
      "step": 951,
      "time_per_iteration": 2.606205940246582
    },
    {
      "auxiliary_loss_clip": 0.01230191,
      "auxiliary_loss_mlp": 0.01041722,
      "balance_loss_clip": 1.07095098,
      "balance_loss_mlp": 1.02980101,
      "epoch": 0.1144712318884146,
      "flos": 19822869757440.0,
      "grad_norm": 1.5893476569464955,
      "language_loss": 0.82427156,
      "learning_rate": 3.92570659464043e-06,
      "loss": 0.8469907,
      "num_input_tokens_seen": 20186095,
      "step": 952,
      "time_per_iteration": 2.5002236366271973
    },
    {
      "auxiliary_loss_clip": 0.01228701,
      "auxiliary_loss_mlp": 0.00763462,
      "balance_loss_clip": 1.07038617,
      "balance_loss_mlp": 1.00008583,
      "epoch": 0.1145914747790537,
      "flos": 14939198766720.0,
      "grad_norm": 2.090783299752373,
      "language_loss": 0.79642951,
      "learning_rate": 3.925496106723349e-06,
      "loss": 0.81635112,
      "num_input_tokens_seen": 20203535,
      "step": 953,
      "time_per_iteration": 2.4995129108428955
    },
    {
      "auxiliary_loss_clip": 0.01228695,
      "auxiliary_loss_mlp": 0.01048317,
      "balance_loss_clip": 1.06869578,
      "balance_loss_mlp": 1.03825557,
      "epoch": 0.11471171766969278,
      "flos": 19865029345920.0,
      "grad_norm": 2.6282307248863397,
      "language_loss": 0.83615541,
      "learning_rate": 3.9252853267100405e-06,
      "loss": 0.85892558,
      "num_input_tokens_seen": 20222780,
      "step": 954,
      "time_per_iteration": 3.296628713607788
    },
    {
      "auxiliary_loss_clip": 0.0119258,
      "auxiliary_loss_mlp": 0.01042857,
      "balance_loss_clip": 1.06709409,
      "balance_loss_mlp": 1.03197908,
      "epoch": 0.11483196056033187,
      "flos": 22526476594560.0,
      "grad_norm": 1.789309610338752,
      "language_loss": 0.8371259,
      "learning_rate": 3.9250742546324786e-06,
      "loss": 0.85948026,
      "num_input_tokens_seen": 20243015,
      "step": 955,
      "time_per_iteration": 2.604696035385132
    },
    {
      "auxiliary_loss_clip": 0.01212603,
      "auxiliary_loss_mlp": 0.0104542,
      "balance_loss_clip": 1.06500053,
      "balance_loss_mlp": 1.03526366,
      "epoch": 0.11495220345097096,
      "flos": 28220292887040.0,
      "grad_norm": 1.6901784695171744,
      "language_loss": 0.8700887,
      "learning_rate": 3.924862890522683e-06,
      "loss": 0.89266896,
      "num_input_tokens_seen": 20263025,
      "step": 956,
      "time_per_iteration": 2.6163079738616943
    },
    {
      "auxiliary_loss_clip": 0.01227848,
      "auxiliary_loss_mlp": 0.01057026,
      "balance_loss_clip": 1.06500506,
      "balance_loss_mlp": 1.04516435,
      "epoch": 0.11507244634161005,
      "flos": 17492267704320.0,
      "grad_norm": 2.1419889500632032,
      "language_loss": 0.85912532,
      "learning_rate": 3.9246512344127174e-06,
      "loss": 0.88197404,
      "num_input_tokens_seen": 20280685,
      "step": 957,
      "time_per_iteration": 4.208640098571777
    },
    {
      "auxiliary_loss_clip": 0.01148945,
      "auxiliary_loss_mlp": 0.01037617,
      "balance_loss_clip": 1.056409,
      "balance_loss_mlp": 1.02759731,
      "epoch": 0.11519268923224914,
      "flos": 22564937082240.0,
      "grad_norm": 2.104975868089068,
      "language_loss": 0.81794298,
      "learning_rate": 3.9244392863346895e-06,
      "loss": 0.83980858,
      "num_input_tokens_seen": 20300090,
      "step": 958,
      "time_per_iteration": 2.8621292114257812
    },
    {
      "auxiliary_loss_clip": 0.01216496,
      "auxiliary_loss_mlp": 0.01053741,
      "balance_loss_clip": 1.07067525,
      "balance_loss_mlp": 1.04220712,
      "epoch": 0.11531293212288823,
      "flos": 16982839065600.0,
      "grad_norm": 1.9126115079367951,
      "language_loss": 0.92378235,
      "learning_rate": 3.9242270463207524e-06,
      "loss": 0.94648468,
      "num_input_tokens_seen": 20318480,
      "step": 959,
      "time_per_iteration": 2.5266430377960205
    },
    {
      "auxiliary_loss_clip": 0.01169277,
      "auxiliary_loss_mlp": 0.01041489,
      "balance_loss_clip": 1.06025088,
      "balance_loss_mlp": 1.0302062,
      "epoch": 0.11543317501352733,
      "flos": 12422004537600.0,
      "grad_norm": 2.612402565174101,
      "language_loss": 0.85360587,
      "learning_rate": 3.924014514403102e-06,
      "loss": 0.87571359,
      "num_input_tokens_seen": 20334635,
      "step": 960,
      "time_per_iteration": 3.428123950958252
    },
    {
      "auxiliary_loss_clip": 0.01170613,
      "auxiliary_loss_mlp": 0.01047243,
      "balance_loss_clip": 1.05839956,
      "balance_loss_mlp": 1.03558445,
      "epoch": 0.11555341790416641,
      "flos": 19821648695040.0,
      "grad_norm": 1.9705249159668405,
      "language_loss": 0.9130134,
      "learning_rate": 3.92380169061398e-06,
      "loss": 0.93519199,
      "num_input_tokens_seen": 20352415,
      "step": 961,
      "time_per_iteration": 2.6064140796661377
    },
    {
      "auxiliary_loss_clip": 0.01188076,
      "auxiliary_loss_mlp": 0.00764472,
      "balance_loss_clip": 1.05784774,
      "balance_loss_mlp": 1.00009096,
      "epoch": 0.11567366079480551,
      "flos": 25738865625600.0,
      "grad_norm": 1.9084322123045003,
      "language_loss": 0.83613479,
      "learning_rate": 3.9235885749856705e-06,
      "loss": 0.85566032,
      "num_input_tokens_seen": 20371095,
      "step": 962,
      "time_per_iteration": 2.6071629524230957
    },
    {
      "auxiliary_loss_clip": 0.01214638,
      "auxiliary_loss_mlp": 0.01044543,
      "balance_loss_clip": 1.06970787,
      "balance_loss_mlp": 1.03370142,
      "epoch": 0.1157939036854446,
      "flos": 18223301301120.0,
      "grad_norm": 2.0163073596184913,
      "language_loss": 0.82715619,
      "learning_rate": 3.9233751675505035e-06,
      "loss": 0.84974802,
      "num_input_tokens_seen": 20389805,
      "step": 963,
      "time_per_iteration": 2.5311121940612793
    },
    {
      "auxiliary_loss_clip": 0.01211543,
      "auxiliary_loss_mlp": 0.01047984,
      "balance_loss_clip": 1.06771255,
      "balance_loss_mlp": 1.03649259,
      "epoch": 0.11591414657608369,
      "flos": 23073755189760.0,
      "grad_norm": 2.3879310253878088,
      "language_loss": 0.85124761,
      "learning_rate": 3.923161468340853e-06,
      "loss": 0.87384284,
      "num_input_tokens_seen": 20409640,
      "step": 964,
      "time_per_iteration": 2.548724412918091
    },
    {
      "auxiliary_loss_clip": 0.01164596,
      "auxiliary_loss_mlp": 0.01047491,
      "balance_loss_clip": 1.05488515,
      "balance_loss_mlp": 1.03645849,
      "epoch": 0.11603438946672277,
      "flos": 19461716461440.0,
      "grad_norm": 2.045262166896079,
      "language_loss": 0.81566608,
      "learning_rate": 3.9229474773891374e-06,
      "loss": 0.83778691,
      "num_input_tokens_seen": 20428180,
      "step": 965,
      "time_per_iteration": 2.6004765033721924
    },
    {
      "auxiliary_loss_clip": 0.01202082,
      "auxiliary_loss_mlp": 0.01041699,
      "balance_loss_clip": 1.05937195,
      "balance_loss_mlp": 1.02982545,
      "epoch": 0.11615463235736187,
      "flos": 26831986272000.0,
      "grad_norm": 2.0188323925548213,
      "language_loss": 0.83729315,
      "learning_rate": 3.922733194727818e-06,
      "loss": 0.85973096,
      "num_input_tokens_seen": 20447975,
      "step": 966,
      "time_per_iteration": 2.6260299682617188
    },
    {
      "auxiliary_loss_clip": 0.01233903,
      "auxiliary_loss_mlp": 0.01049225,
      "balance_loss_clip": 1.07061553,
      "balance_loss_mlp": 1.037745,
      "epoch": 0.11627487524800097,
      "flos": 18580324533120.0,
      "grad_norm": 1.9403762773585176,
      "language_loss": 0.87334752,
      "learning_rate": 3.922518620389402e-06,
      "loss": 0.89617884,
      "num_input_tokens_seen": 20464840,
      "step": 967,
      "time_per_iteration": 2.509111166000366
    },
    {
      "auxiliary_loss_clip": 0.01124698,
      "auxiliary_loss_mlp": 0.01043963,
      "balance_loss_clip": 1.05446291,
      "balance_loss_mlp": 1.03324056,
      "epoch": 0.11639511813864005,
      "flos": 18150474476160.0,
      "grad_norm": 1.7170641848430415,
      "language_loss": 0.89445019,
      "learning_rate": 3.922303754406439e-06,
      "loss": 0.91613686,
      "num_input_tokens_seen": 20482680,
      "step": 968,
      "time_per_iteration": 2.72664737701416
    },
    {
      "auxiliary_loss_clip": 0.01172959,
      "auxiliary_loss_mlp": 0.01043741,
      "balance_loss_clip": 1.05503273,
      "balance_loss_mlp": 1.03269625,
      "epoch": 0.11651536102927915,
      "flos": 20922023888640.0,
      "grad_norm": 2.0080600610882526,
      "language_loss": 0.78988111,
      "learning_rate": 3.922088596811526e-06,
      "loss": 0.81204808,
      "num_input_tokens_seen": 20501810,
      "step": 969,
      "time_per_iteration": 2.740123987197876
    },
    {
      "auxiliary_loss_clip": 0.01214807,
      "auxiliary_loss_mlp": 0.01049977,
      "balance_loss_clip": 1.06466746,
      "balance_loss_mlp": 1.03863406,
      "epoch": 0.11663560391991823,
      "flos": 16508602776960.0,
      "grad_norm": 2.35869317590607,
      "language_loss": 0.86652577,
      "learning_rate": 3.9218731476373e-06,
      "loss": 0.88917363,
      "num_input_tokens_seen": 20517995,
      "step": 970,
      "time_per_iteration": 2.4864609241485596
    },
    {
      "auxiliary_loss_clip": 0.01229663,
      "auxiliary_loss_mlp": 0.01037193,
      "balance_loss_clip": 1.06900525,
      "balance_loss_mlp": 1.02540922,
      "epoch": 0.11675584681055733,
      "flos": 19865029345920.0,
      "grad_norm": 2.180462380128972,
      "language_loss": 0.84922028,
      "learning_rate": 3.9216574069164455e-06,
      "loss": 0.87188888,
      "num_input_tokens_seen": 20536970,
      "step": 971,
      "time_per_iteration": 2.5146942138671875
    },
    {
      "auxiliary_loss_clip": 0.01239151,
      "auxiliary_loss_mlp": 0.01043031,
      "balance_loss_clip": 1.06715441,
      "balance_loss_mlp": 1.03293383,
      "epoch": 0.11687608970119642,
      "flos": 21944364785280.0,
      "grad_norm": 1.5684329703262403,
      "language_loss": 0.80336547,
      "learning_rate": 3.921441374681691e-06,
      "loss": 0.82618725,
      "num_input_tokens_seen": 20557030,
      "step": 972,
      "time_per_iteration": 2.496046304702759
    },
    {
      "auxiliary_loss_clip": 0.01209078,
      "auxiliary_loss_mlp": 0.01039586,
      "balance_loss_clip": 1.06617749,
      "balance_loss_mlp": 1.02924454,
      "epoch": 0.1169963325918355,
      "flos": 24061155131520.0,
      "grad_norm": 1.815288909968163,
      "language_loss": 0.64879394,
      "learning_rate": 3.921225050965808e-06,
      "loss": 0.6712805,
      "num_input_tokens_seen": 20576915,
      "step": 973,
      "time_per_iteration": 2.567025661468506
    },
    {
      "auxiliary_loss_clip": 0.011969,
      "auxiliary_loss_mlp": 0.01044693,
      "balance_loss_clip": 1.06524003,
      "balance_loss_mlp": 1.0336777,
      "epoch": 0.1171165754824746,
      "flos": 23368151059200.0,
      "grad_norm": 2.046361113843374,
      "language_loss": 0.75033188,
      "learning_rate": 3.921008435801612e-06,
      "loss": 0.77274776,
      "num_input_tokens_seen": 20596000,
      "step": 974,
      "time_per_iteration": 2.5719377994537354
    },
    {
      "auxiliary_loss_clip": 0.01213001,
      "auxiliary_loss_mlp": 0.01047457,
      "balance_loss_clip": 1.06314492,
      "balance_loss_mlp": 1.03598344,
      "epoch": 0.11723681837311369,
      "flos": 18552243075840.0,
      "grad_norm": 2.1182193404652154,
      "language_loss": 0.7572149,
      "learning_rate": 3.920791529221963e-06,
      "loss": 0.77981949,
      "num_input_tokens_seen": 20614675,
      "step": 975,
      "time_per_iteration": 2.5028796195983887
    },
    {
      "auxiliary_loss_clip": 0.01215737,
      "auxiliary_loss_mlp": 0.00764091,
      "balance_loss_clip": 1.06563377,
      "balance_loss_mlp": 1.00011647,
      "epoch": 0.11735706126375278,
      "flos": 23550541344000.0,
      "grad_norm": 1.9582907001725058,
      "language_loss": 0.7651614,
      "learning_rate": 3.920574331259768e-06,
      "loss": 0.78495967,
      "num_input_tokens_seen": 20635875,
      "step": 976,
      "time_per_iteration": 2.5761897563934326
    },
    {
      "auxiliary_loss_clip": 0.01202108,
      "auxiliary_loss_mlp": 0.01045077,
      "balance_loss_clip": 1.06470919,
      "balance_loss_mlp": 1.03384185,
      "epoch": 0.11747730415439187,
      "flos": 22381541216640.0,
      "grad_norm": 2.122575776477431,
      "language_loss": 0.79257643,
      "learning_rate": 3.9203568419479716e-06,
      "loss": 0.81504828,
      "num_input_tokens_seen": 20656430,
      "step": 977,
      "time_per_iteration": 2.534876823425293
    },
    {
      "auxiliary_loss_clip": 0.01209215,
      "auxiliary_loss_mlp": 0.01040546,
      "balance_loss_clip": 1.06407452,
      "balance_loss_mlp": 1.02985883,
      "epoch": 0.11759754704503096,
      "flos": 22200731130240.0,
      "grad_norm": 1.830706593722671,
      "language_loss": 0.74941134,
      "learning_rate": 3.92013906131957e-06,
      "loss": 0.77190888,
      "num_input_tokens_seen": 20675360,
      "step": 978,
      "time_per_iteration": 2.526524066925049
    },
    {
      "auxiliary_loss_clip": 0.01193855,
      "auxiliary_loss_mlp": 0.01049449,
      "balance_loss_clip": 1.06303024,
      "balance_loss_mlp": 1.03899395,
      "epoch": 0.11771778993567006,
      "flos": 22309755886080.0,
      "grad_norm": 1.615912471773369,
      "language_loss": 0.82558942,
      "learning_rate": 3.9199209894076e-06,
      "loss": 0.84802246,
      "num_input_tokens_seen": 20695675,
      "step": 979,
      "time_per_iteration": 2.5733649730682373
    },
    {
      "auxiliary_loss_clip": 0.0124346,
      "auxiliary_loss_mlp": 0.01050297,
      "balance_loss_clip": 1.06771302,
      "balance_loss_mlp": 1.0378691,
      "epoch": 0.11783803282630914,
      "flos": 21288169175040.0,
      "grad_norm": 1.8940638460998847,
      "language_loss": 0.90368426,
      "learning_rate": 3.919702626245142e-06,
      "loss": 0.92662185,
      "num_input_tokens_seen": 20715330,
      "step": 980,
      "time_per_iteration": 3.227565050125122
    },
    {
      "auxiliary_loss_clip": 0.01196298,
      "auxiliary_loss_mlp": 0.01041078,
      "balance_loss_clip": 1.06167197,
      "balance_loss_mlp": 1.03058147,
      "epoch": 0.11795827571694824,
      "flos": 25371535190400.0,
      "grad_norm": 2.0361380209453577,
      "language_loss": 0.66328055,
      "learning_rate": 3.919483971865322e-06,
      "loss": 0.68565434,
      "num_input_tokens_seen": 20735325,
      "step": 981,
      "time_per_iteration": 2.5614726543426514
    },
    {
      "auxiliary_loss_clip": 0.01210102,
      "auxiliary_loss_mlp": 0.01042892,
      "balance_loss_clip": 1.06693888,
      "balance_loss_mlp": 1.03231215,
      "epoch": 0.11807851860758732,
      "flos": 23622218933760.0,
      "grad_norm": 2.1589731525977682,
      "language_loss": 0.87997955,
      "learning_rate": 3.91926502630131e-06,
      "loss": 0.90250951,
      "num_input_tokens_seen": 20755940,
      "step": 982,
      "time_per_iteration": 3.320653200149536
    },
    {
      "auxiliary_loss_clip": 0.01231064,
      "auxiliary_loss_mlp": 0.01040098,
      "balance_loss_clip": 1.07202613,
      "balance_loss_mlp": 1.03068686,
      "epoch": 0.11819876149822642,
      "flos": 24972496024320.0,
      "grad_norm": 2.2612598626659897,
      "language_loss": 0.72188574,
      "learning_rate": 3.91904578958632e-06,
      "loss": 0.74459732,
      "num_input_tokens_seen": 20775355,
      "step": 983,
      "time_per_iteration": 3.2702271938323975
    },
    {
      "auxiliary_loss_clip": 0.01243829,
      "auxiliary_loss_mlp": 0.0104274,
      "balance_loss_clip": 1.06949043,
      "balance_loss_mlp": 1.0311408,
      "epoch": 0.11831900438886551,
      "flos": 23003226835200.0,
      "grad_norm": 2.1903680872947096,
      "language_loss": 0.84238833,
      "learning_rate": 3.918826261753608e-06,
      "loss": 0.86525404,
      "num_input_tokens_seen": 20794935,
      "step": 984,
      "time_per_iteration": 2.4896678924560547
    },
    {
      "auxiliary_loss_clip": 0.01210961,
      "auxiliary_loss_mlp": 0.01046308,
      "balance_loss_clip": 1.06473994,
      "balance_loss_mlp": 1.03587687,
      "epoch": 0.1184392472795046,
      "flos": 27965147604480.0,
      "grad_norm": 3.364835766975984,
      "language_loss": 0.71299851,
      "learning_rate": 3.918606442836478e-06,
      "loss": 0.73557115,
      "num_input_tokens_seen": 20817155,
      "step": 985,
      "time_per_iteration": 3.4297571182250977
    },
    {
      "auxiliary_loss_clip": 0.01227096,
      "auxiliary_loss_mlp": 0.01047707,
      "balance_loss_clip": 1.07068181,
      "balance_loss_mlp": 1.0372349,
      "epoch": 0.1185594901701437,
      "flos": 19898497843200.0,
      "grad_norm": 1.7814807041830716,
      "language_loss": 0.77578145,
      "learning_rate": 3.918386332868277e-06,
      "loss": 0.79852951,
      "num_input_tokens_seen": 20835125,
      "step": 986,
      "time_per_iteration": 2.496600866317749
    },
    {
      "auxiliary_loss_clip": 0.01212346,
      "auxiliary_loss_mlp": 0.01043894,
      "balance_loss_clip": 1.06438839,
      "balance_loss_mlp": 1.0332185,
      "epoch": 0.11867973306078278,
      "flos": 18912354877440.0,
      "grad_norm": 1.9041496624056555,
      "language_loss": 0.94322979,
      "learning_rate": 3.918165931882394e-06,
      "loss": 0.96579218,
      "num_input_tokens_seen": 20853525,
      "step": 987,
      "time_per_iteration": 2.4803590774536133
    },
    {
      "auxiliary_loss_clip": 0.01153236,
      "auxiliary_loss_mlp": 0.01046308,
      "balance_loss_clip": 1.05304921,
      "balance_loss_mlp": 1.03509068,
      "epoch": 0.11879997595142187,
      "flos": 16982803152000.0,
      "grad_norm": 17.62611678046567,
      "language_loss": 0.7547071,
      "learning_rate": 3.917945239912264e-06,
      "loss": 0.77670258,
      "num_input_tokens_seen": 20871000,
      "step": 988,
      "time_per_iteration": 2.614936113357544
    },
    {
      "auxiliary_loss_clip": 0.01178602,
      "auxiliary_loss_mlp": 0.01039397,
      "balance_loss_clip": 1.0602349,
      "balance_loss_mlp": 1.02894831,
      "epoch": 0.11892021884206096,
      "flos": 17530369056000.0,
      "grad_norm": 2.0324308378220324,
      "language_loss": 0.75453484,
      "learning_rate": 3.917724256991367e-06,
      "loss": 0.77671486,
      "num_input_tokens_seen": 20889745,
      "step": 989,
      "time_per_iteration": 2.578970193862915
    },
    {
      "auxiliary_loss_clip": 0.01196501,
      "auxiliary_loss_mlp": 0.01036995,
      "balance_loss_clip": 1.06219161,
      "balance_loss_mlp": 1.02698135,
      "epoch": 0.11904046173270005,
      "flos": 30955895763840.0,
      "grad_norm": 3.1331028476364158,
      "language_loss": 0.81216979,
      "learning_rate": 3.9175029831532245e-06,
      "loss": 0.83450472,
      "num_input_tokens_seen": 20909260,
      "step": 990,
      "time_per_iteration": 2.6072514057159424
    },
    {
      "auxiliary_loss_clip": 0.01197633,
      "auxiliary_loss_mlp": 0.01038387,
      "balance_loss_clip": 1.0647769,
      "balance_loss_mlp": 1.02765238,
      "epoch": 0.11916070462333915,
      "flos": 20157234485760.0,
      "grad_norm": 2.0083885017395473,
      "language_loss": 0.88945067,
      "learning_rate": 3.917281418431404e-06,
      "loss": 0.91181087,
      "num_input_tokens_seen": 20928305,
      "step": 991,
      "time_per_iteration": 2.577902317047119
    },
    {
      "auxiliary_loss_clip": 0.0120928,
      "auxiliary_loss_mlp": 0.01050016,
      "balance_loss_clip": 1.06682706,
      "balance_loss_mlp": 1.03953743,
      "epoch": 0.11928094751397823,
      "flos": 23551115961600.0,
      "grad_norm": 2.4028194144844246,
      "language_loss": 0.76611543,
      "learning_rate": 3.917059562859516e-06,
      "loss": 0.78870845,
      "num_input_tokens_seen": 20947630,
      "step": 992,
      "time_per_iteration": 2.555708169937134
    },
    {
      "auxiliary_loss_clip": 0.0120201,
      "auxiliary_loss_mlp": 0.01045265,
      "balance_loss_clip": 1.06535006,
      "balance_loss_mlp": 1.03535318,
      "epoch": 0.11940119040461733,
      "flos": 23908426502400.0,
      "grad_norm": 2.1386527664444506,
      "language_loss": 0.88685012,
      "learning_rate": 3.916837416471218e-06,
      "loss": 0.90932292,
      "num_input_tokens_seen": 20964250,
      "step": 993,
      "time_per_iteration": 2.573564291000366
    },
    {
      "auxiliary_loss_clip": 0.01218084,
      "auxiliary_loss_mlp": 0.01037131,
      "balance_loss_clip": 1.06228542,
      "balance_loss_mlp": 1.02732015,
      "epoch": 0.11952143329525641,
      "flos": 13844533835520.0,
      "grad_norm": 2.363845444000928,
      "language_loss": 0.72483838,
      "learning_rate": 3.916614979300207e-06,
      "loss": 0.74739045,
      "num_input_tokens_seen": 20979095,
      "step": 994,
      "time_per_iteration": 2.48179030418396
    },
    {
      "auxiliary_loss_clip": 0.01170348,
      "auxiliary_loss_mlp": 0.01047764,
      "balance_loss_clip": 1.0617075,
      "balance_loss_mlp": 1.03817368,
      "epoch": 0.11964167618589551,
      "flos": 27015525792000.0,
      "grad_norm": 2.350948582833249,
      "language_loss": 0.78669196,
      "learning_rate": 3.9163922513802274e-06,
      "loss": 0.80887306,
      "num_input_tokens_seen": 21001430,
      "step": 995,
      "time_per_iteration": 2.669786214828491
    },
    {
      "auxiliary_loss_clip": 0.01241704,
      "auxiliary_loss_mlp": 0.0104019,
      "balance_loss_clip": 1.0674206,
      "balance_loss_mlp": 1.02889538,
      "epoch": 0.1197619190765346,
      "flos": 12567622273920.0,
      "grad_norm": 5.665754895251741,
      "language_loss": 0.82662559,
      "learning_rate": 3.916169232745067e-06,
      "loss": 0.84944451,
      "num_input_tokens_seen": 21019105,
      "step": 996,
      "time_per_iteration": 2.449256658554077
    },
    {
      "auxiliary_loss_clip": 0.01193978,
      "auxiliary_loss_mlp": 0.01042703,
      "balance_loss_clip": 1.05947232,
      "balance_loss_mlp": 1.03119898,
      "epoch": 0.11988216196717369,
      "flos": 16909437623040.0,
      "grad_norm": 2.7984057589196656,
      "language_loss": 0.92141402,
      "learning_rate": 3.915945923428559e-06,
      "loss": 0.9437809,
      "num_input_tokens_seen": 21035630,
      "step": 997,
      "time_per_iteration": 2.51652455329895
    },
    {
      "auxiliary_loss_clip": 0.01219222,
      "auxiliary_loss_mlp": 0.01036305,
      "balance_loss_clip": 1.06218028,
      "balance_loss_mlp": 1.02565324,
      "epoch": 0.12000240485781279,
      "flos": 16216577205120.0,
      "grad_norm": 2.196823570379757,
      "language_loss": 0.83077949,
      "learning_rate": 3.915722323464577e-06,
      "loss": 0.85333478,
      "num_input_tokens_seen": 21054235,
      "step": 998,
      "time_per_iteration": 2.489825487136841
    },
    {
      "auxiliary_loss_clip": 0.01224061,
      "auxiliary_loss_mlp": 0.01039539,
      "balance_loss_clip": 1.06665778,
      "balance_loss_mlp": 1.02829814,
      "epoch": 0.12012264774845187,
      "flos": 49344887525760.0,
      "grad_norm": 2.287014030481074,
      "language_loss": 0.70666665,
      "learning_rate": 3.91549843288704e-06,
      "loss": 0.72930264,
      "num_input_tokens_seen": 21077915,
      "step": 999,
      "time_per_iteration": 2.747929573059082
    },
    {
      "auxiliary_loss_clip": 0.01194854,
      "auxiliary_loss_mlp": 0.00764529,
      "balance_loss_clip": 1.05817819,
      "balance_loss_mlp": 1.00008881,
      "epoch": 0.12024289063909097,
      "flos": 26979435601920.0,
      "grad_norm": 2.205937005047137,
      "language_loss": 0.78925496,
      "learning_rate": 3.915274251729916e-06,
      "loss": 0.8088488,
      "num_input_tokens_seen": 21099205,
      "step": 1000,
      "time_per_iteration": 2.6223549842834473
    },
    {
      "auxiliary_loss_clip": 0.0120193,
      "auxiliary_loss_mlp": 0.01039759,
      "balance_loss_clip": 1.06504798,
      "balance_loss_mlp": 1.02836823,
      "epoch": 0.12036313352973005,
      "flos": 19537308633600.0,
      "grad_norm": 1.8510918185789669,
      "language_loss": 0.90102053,
      "learning_rate": 3.91504978002721e-06,
      "loss": 0.92343736,
      "num_input_tokens_seen": 21118260,
      "step": 1001,
      "time_per_iteration": 2.5627899169921875
    },
    {
      "auxiliary_loss_clip": 0.0120985,
      "auxiliary_loss_mlp": 0.00764003,
      "balance_loss_clip": 1.06157613,
      "balance_loss_mlp": 1.00008309,
      "epoch": 0.12048337642036915,
      "flos": 17268256535040.0,
      "grad_norm": 1.8351333856395746,
      "language_loss": 0.76219308,
      "learning_rate": 3.914825017812974e-06,
      "loss": 0.78193158,
      "num_input_tokens_seen": 21134910,
      "step": 1002,
      "time_per_iteration": 2.506192445755005
    },
    {
      "auxiliary_loss_clip": 0.01210395,
      "auxiliary_loss_mlp": 0.01034836,
      "balance_loss_clip": 1.06747937,
      "balance_loss_mlp": 1.0246377,
      "epoch": 0.12060361931100824,
      "flos": 22856962654080.0,
      "grad_norm": 2.3618947865169138,
      "language_loss": 0.72538304,
      "learning_rate": 3.9145999651213065e-06,
      "loss": 0.7478354,
      "num_input_tokens_seen": 21154150,
      "step": 1003,
      "time_per_iteration": 2.5486679077148438
    },
    {
      "auxiliary_loss_clip": 0.01228804,
      "auxiliary_loss_mlp": 0.01045953,
      "balance_loss_clip": 1.07009506,
      "balance_loss_mlp": 1.03474689,
      "epoch": 0.12072386220164733,
      "flos": 16726795943040.0,
      "grad_norm": 2.941037161534408,
      "language_loss": 0.88428086,
      "learning_rate": 3.9143746219863465e-06,
      "loss": 0.90702844,
      "num_input_tokens_seen": 21171255,
      "step": 1004,
      "time_per_iteration": 2.478365421295166
    },
    {
      "auxiliary_loss_clip": 0.01127148,
      "auxiliary_loss_mlp": 0.01010518,
      "balance_loss_clip": 1.04374146,
      "balance_loss_mlp": 1.00746667,
      "epoch": 0.12084410509228642,
      "flos": 55144176105600.0,
      "grad_norm": 0.946783332347811,
      "language_loss": 0.64760125,
      "learning_rate": 3.914148988442278e-06,
      "loss": 0.66897792,
      "num_input_tokens_seen": 21227045,
      "step": 1005,
      "time_per_iteration": 3.1101934909820557
    },
    {
      "auxiliary_loss_clip": 0.01193567,
      "auxiliary_loss_mlp": 0.01043695,
      "balance_loss_clip": 1.06080675,
      "balance_loss_mlp": 1.03229272,
      "epoch": 0.1209643479829255,
      "flos": 26760236855040.0,
      "grad_norm": 2.557274589996241,
      "language_loss": 0.94781184,
      "learning_rate": 3.91392306452333e-06,
      "loss": 0.9701845,
      "num_input_tokens_seen": 21244120,
      "step": 1006,
      "time_per_iteration": 3.2965023517608643
    },
    {
      "auxiliary_loss_clip": 0.01241209,
      "auxiliary_loss_mlp": 0.01041381,
      "balance_loss_clip": 1.06694961,
      "balance_loss_mlp": 1.0306108,
      "epoch": 0.1210845908735646,
      "flos": 11035026725760.0,
      "grad_norm": 3.089521166468796,
      "language_loss": 0.66314483,
      "learning_rate": 3.913696850263774e-06,
      "loss": 0.68597078,
      "num_input_tokens_seen": 21258485,
      "step": 1007,
      "time_per_iteration": 2.4378108978271484
    },
    {
      "auxiliary_loss_clip": 0.01226125,
      "auxiliary_loss_mlp": 0.01034648,
      "balance_loss_clip": 1.06631136,
      "balance_loss_mlp": 1.02385378,
      "epoch": 0.1212048337642037,
      "flos": 20484631975680.0,
      "grad_norm": 1.948507140410761,
      "language_loss": 0.79312527,
      "learning_rate": 3.913470345697929e-06,
      "loss": 0.81573302,
      "num_input_tokens_seen": 21277115,
      "step": 1008,
      "time_per_iteration": 2.5120370388031006
    },
    {
      "auxiliary_loss_clip": 0.0117809,
      "auxiliary_loss_mlp": 0.01042639,
      "balance_loss_clip": 1.06074333,
      "balance_loss_mlp": 1.0321312,
      "epoch": 0.12132507665484278,
      "flos": 22346061557760.0,
      "grad_norm": 2.147806731130658,
      "language_loss": 0.85453367,
      "learning_rate": 3.913243550860153e-06,
      "loss": 0.87674093,
      "num_input_tokens_seen": 21294880,
      "step": 1009,
      "time_per_iteration": 4.116698980331421
    },
    {
      "auxiliary_loss_clip": 0.01228214,
      "auxiliary_loss_mlp": 0.01040269,
      "balance_loss_clip": 1.07086015,
      "balance_loss_mlp": 1.02935588,
      "epoch": 0.12144531954548188,
      "flos": 29314957818240.0,
      "grad_norm": 2.842664618129696,
      "language_loss": 0.75874376,
      "learning_rate": 3.913016465784852e-06,
      "loss": 0.78142858,
      "num_input_tokens_seen": 21315555,
      "step": 1010,
      "time_per_iteration": 2.5714364051818848
    },
    {
      "auxiliary_loss_clip": 0.01174774,
      "auxiliary_loss_mlp": 0.01042177,
      "balance_loss_clip": 1.05781555,
      "balance_loss_mlp": 1.03135252,
      "epoch": 0.12156556243612096,
      "flos": 20485242506880.0,
      "grad_norm": 2.91900781367874,
      "language_loss": 0.72013491,
      "learning_rate": 3.912789090506474e-06,
      "loss": 0.74230444,
      "num_input_tokens_seen": 21334815,
      "step": 1011,
      "time_per_iteration": 3.4006738662719727
    },
    {
      "auxiliary_loss_clip": 0.01200214,
      "auxiliary_loss_mlp": 0.01036179,
      "balance_loss_clip": 1.0606755,
      "balance_loss_mlp": 1.02527761,
      "epoch": 0.12168580532676006,
      "flos": 16472009796480.0,
      "grad_norm": 4.9035041887337485,
      "language_loss": 0.71690369,
      "learning_rate": 3.9125614250595114e-06,
      "loss": 0.73926765,
      "num_input_tokens_seen": 21351025,
      "step": 1012,
      "time_per_iteration": 2.533656358718872
    },
    {
      "auxiliary_loss_clip": 0.01227168,
      "auxiliary_loss_mlp": 0.01049969,
      "balance_loss_clip": 1.06686234,
      "balance_loss_mlp": 1.03878081,
      "epoch": 0.12180604821739914,
      "flos": 15341290588800.0,
      "grad_norm": 2.8939235399973793,
      "language_loss": 0.89178646,
      "learning_rate": 3.912333469478502e-06,
      "loss": 0.91455787,
      "num_input_tokens_seen": 21368990,
      "step": 1013,
      "time_per_iteration": 2.47973895072937
    },
    {
      "auxiliary_loss_clip": 0.01212338,
      "auxiliary_loss_mlp": 0.01042517,
      "balance_loss_clip": 1.0658381,
      "balance_loss_mlp": 1.03148365,
      "epoch": 0.12192629110803824,
      "flos": 19318038059520.0,
      "grad_norm": 3.3512641495084448,
      "language_loss": 0.78254145,
      "learning_rate": 3.912105223798025e-06,
      "loss": 0.80509001,
      "num_input_tokens_seen": 21388410,
      "step": 1014,
      "time_per_iteration": 2.5451502799987793
    },
    {
      "auxiliary_loss_clip": 0.0111427,
      "auxiliary_loss_mlp": 0.01007977,
      "balance_loss_clip": 1.04266477,
      "balance_loss_mlp": 1.00494945,
      "epoch": 0.12204653399867733,
      "flos": 47725354085760.0,
      "grad_norm": 0.999236496528554,
      "language_loss": 0.6771003,
      "learning_rate": 3.9118766880527065e-06,
      "loss": 0.69832277,
      "num_input_tokens_seen": 21442845,
      "step": 1015,
      "time_per_iteration": 3.0877015590667725
    },
    {
      "auxiliary_loss_clip": 0.01169396,
      "auxiliary_loss_mlp": 0.01040262,
      "balance_loss_clip": 1.05801344,
      "balance_loss_mlp": 1.03053463,
      "epoch": 0.12216677688931642,
      "flos": 18221936584320.0,
      "grad_norm": 1.8313106676261872,
      "language_loss": 0.73843783,
      "learning_rate": 3.9116478622772145e-06,
      "loss": 0.76053441,
      "num_input_tokens_seen": 21461420,
      "step": 1016,
      "time_per_iteration": 2.5957770347595215
    },
    {
      "auxiliary_loss_clip": 0.01221427,
      "auxiliary_loss_mlp": 0.01044564,
      "balance_loss_clip": 1.06845319,
      "balance_loss_mlp": 1.03365088,
      "epoch": 0.12228701977995551,
      "flos": 27525636789120.0,
      "grad_norm": 1.9327462647473441,
      "language_loss": 0.88105464,
      "learning_rate": 3.911418746506261e-06,
      "loss": 0.90371454,
      "num_input_tokens_seen": 21481550,
      "step": 1017,
      "time_per_iteration": 2.591132879257202
    },
    {
      "auxiliary_loss_clip": 0.01227371,
      "auxiliary_loss_mlp": 0.01043338,
      "balance_loss_clip": 1.07011354,
      "balance_loss_mlp": 1.03272271,
      "epoch": 0.1224072626705946,
      "flos": 21798136517760.0,
      "grad_norm": 3.5264767260250314,
      "language_loss": 0.78603655,
      "learning_rate": 3.911189340774604e-06,
      "loss": 0.8087436,
      "num_input_tokens_seen": 21501680,
      "step": 1018,
      "time_per_iteration": 2.509046792984009
    },
    {
      "auxiliary_loss_clip": 0.01214909,
      "auxiliary_loss_mlp": 0.01041383,
      "balance_loss_clip": 1.06339788,
      "balance_loss_mlp": 1.03122067,
      "epoch": 0.1225275055612337,
      "flos": 20703758895360.0,
      "grad_norm": 2.291850811442564,
      "language_loss": 0.79665935,
      "learning_rate": 3.910959645117043e-06,
      "loss": 0.81922233,
      "num_input_tokens_seen": 21521015,
      "step": 1019,
      "time_per_iteration": 2.5594072341918945
    },
    {
      "auxiliary_loss_clip": 0.01115963,
      "auxiliary_loss_mlp": 0.00754196,
      "balance_loss_clip": 1.03989565,
      "balance_loss_mlp": 0.99988312,
      "epoch": 0.12264774845187278,
      "flos": 57745294462080.0,
      "grad_norm": 0.8158658496248952,
      "language_loss": 0.56706023,
      "learning_rate": 3.910729659568423e-06,
      "loss": 0.58576179,
      "num_input_tokens_seen": 21578200,
      "step": 1020,
      "time_per_iteration": 3.125624179840088
    },
    {
      "auxiliary_loss_clip": 0.01209109,
      "auxiliary_loss_mlp": 0.01050486,
      "balance_loss_clip": 1.06729126,
      "balance_loss_mlp": 1.04064512,
      "epoch": 0.12276799134251187,
      "flos": 26396282298240.0,
      "grad_norm": 2.3552641417573708,
      "language_loss": 0.82110727,
      "learning_rate": 3.9104993841636344e-06,
      "loss": 0.84370321,
      "num_input_tokens_seen": 21598770,
      "step": 1021,
      "time_per_iteration": 2.6077752113342285
    },
    {
      "auxiliary_loss_clip": 0.01206334,
      "auxiliary_loss_mlp": 0.00763009,
      "balance_loss_clip": 1.07008576,
      "balance_loss_mlp": 1.00006676,
      "epoch": 0.12288823423315097,
      "flos": 21064193919360.0,
      "grad_norm": 1.7764777241659957,
      "language_loss": 0.81089461,
      "learning_rate": 3.910268818937608e-06,
      "loss": 0.8305881,
      "num_input_tokens_seen": 21616925,
      "step": 1022,
      "time_per_iteration": 2.589017629623413
    },
    {
      "auxiliary_loss_clip": 0.01182627,
      "auxiliary_loss_mlp": 0.0104498,
      "balance_loss_clip": 1.06673908,
      "balance_loss_mlp": 1.03384018,
      "epoch": 0.12300847712379005,
      "flos": 12312441077760.0,
      "grad_norm": 2.272365399970682,
      "language_loss": 0.87539548,
      "learning_rate": 3.9100379639253196e-06,
      "loss": 0.89767152,
      "num_input_tokens_seen": 21633645,
      "step": 1023,
      "time_per_iteration": 2.594097375869751
    },
    {
      "auxiliary_loss_clip": 0.01208385,
      "auxiliary_loss_mlp": 0.01040962,
      "balance_loss_clip": 1.0630374,
      "balance_loss_mlp": 1.03034019,
      "epoch": 0.12312872001442915,
      "flos": 16762239688320.0,
      "grad_norm": 2.63593625940969,
      "language_loss": 0.86253965,
      "learning_rate": 3.909806819161791e-06,
      "loss": 0.88503313,
      "num_input_tokens_seen": 21649120,
      "step": 1024,
      "time_per_iteration": 2.507352590560913
    },
    {
      "auxiliary_loss_clip": 0.01197829,
      "auxiliary_loss_mlp": 0.01039948,
      "balance_loss_clip": 1.06184983,
      "balance_loss_mlp": 1.02923715,
      "epoch": 0.12324896290506823,
      "flos": 18404937400320.0,
      "grad_norm": 2.0377459484418408,
      "language_loss": 0.85922503,
      "learning_rate": 3.909575384682086e-06,
      "loss": 0.88160288,
      "num_input_tokens_seen": 21668000,
      "step": 1025,
      "time_per_iteration": 2.5611326694488525
    },
    {
      "auxiliary_loss_clip": 0.01227751,
      "auxiliary_loss_mlp": 0.01032966,
      "balance_loss_clip": 1.06705713,
      "balance_loss_mlp": 1.02218926,
      "epoch": 0.12336920579570733,
      "flos": 18915407533440.0,
      "grad_norm": 1.8388627303701173,
      "language_loss": 0.69163442,
      "learning_rate": 3.9093436605213144e-06,
      "loss": 0.71424156,
      "num_input_tokens_seen": 21688500,
      "step": 1026,
      "time_per_iteration": 2.537940740585327
    },
    {
      "auxiliary_loss_clip": 0.01203698,
      "auxiliary_loss_mlp": 0.01041527,
      "balance_loss_clip": 1.06201303,
      "balance_loss_mlp": 1.03127527,
      "epoch": 0.12348944868634643,
      "flos": 23878369797120.0,
      "grad_norm": 2.0210308161222366,
      "language_loss": 0.7967819,
      "learning_rate": 3.909111646714627e-06,
      "loss": 0.81923413,
      "num_input_tokens_seen": 21709345,
      "step": 1027,
      "time_per_iteration": 2.579461097717285
    },
    {
      "auxiliary_loss_clip": 0.01238131,
      "auxiliary_loss_mlp": 0.01042365,
      "balance_loss_clip": 1.06863713,
      "balance_loss_mlp": 1.03266764,
      "epoch": 0.12360969157698551,
      "flos": 19026084314880.0,
      "grad_norm": 2.1369194900366577,
      "language_loss": 0.72323263,
      "learning_rate": 3.9088793432972206e-06,
      "loss": 0.74603754,
      "num_input_tokens_seen": 21728165,
      "step": 1028,
      "time_per_iteration": 2.4654159545898438
    },
    {
      "auxiliary_loss_clip": 0.01182774,
      "auxiliary_loss_mlp": 0.01046604,
      "balance_loss_clip": 1.06353712,
      "balance_loss_mlp": 1.03705001,
      "epoch": 0.1237299344676246,
      "flos": 13224607983360.0,
      "grad_norm": 2.131551928801967,
      "language_loss": 0.82159871,
      "learning_rate": 3.908646750304336e-06,
      "loss": 0.84389251,
      "num_input_tokens_seen": 21745850,
      "step": 1029,
      "time_per_iteration": 2.57853627204895
    },
    {
      "auxiliary_loss_clip": 0.01211947,
      "auxiliary_loss_mlp": 0.0103993,
      "balance_loss_clip": 1.06763446,
      "balance_loss_mlp": 1.02903426,
      "epoch": 0.12385017735826369,
      "flos": 20485673470080.0,
      "grad_norm": 1.5975521024949157,
      "language_loss": 0.87238926,
      "learning_rate": 3.908413867771257e-06,
      "loss": 0.89490801,
      "num_input_tokens_seen": 21764760,
      "step": 1030,
      "time_per_iteration": 2.540369987487793
    },
    {
      "auxiliary_loss_clip": 0.01226985,
      "auxiliary_loss_mlp": 0.01050138,
      "balance_loss_clip": 1.07096887,
      "balance_loss_mlp": 1.03969574,
      "epoch": 0.12397042024890279,
      "flos": 17347835116800.0,
      "grad_norm": 2.2459092524997883,
      "language_loss": 0.80491227,
      "learning_rate": 3.908180695733311e-06,
      "loss": 0.82768351,
      "num_input_tokens_seen": 21784250,
      "step": 1031,
      "time_per_iteration": 2.5008444786071777
    },
    {
      "auxiliary_loss_clip": 0.01151949,
      "auxiliary_loss_mlp": 0.01038405,
      "balance_loss_clip": 1.05252504,
      "balance_loss_mlp": 1.02744436,
      "epoch": 0.12409066313954187,
      "flos": 20412343854720.0,
      "grad_norm": 1.9526292427774656,
      "language_loss": 0.82673395,
      "learning_rate": 3.907947234225871e-06,
      "loss": 0.84863746,
      "num_input_tokens_seen": 21803260,
      "step": 1032,
      "time_per_iteration": 3.433749198913574
    },
    {
      "auxiliary_loss_clip": 0.01157618,
      "auxiliary_loss_mlp": 0.01037487,
      "balance_loss_clip": 1.05775177,
      "balance_loss_mlp": 1.02688885,
      "epoch": 0.12421090603018096,
      "flos": 20736688688640.0,
      "grad_norm": 1.7882049050897433,
      "language_loss": 0.87298417,
      "learning_rate": 3.907713483284352e-06,
      "loss": 0.89493525,
      "num_input_tokens_seen": 21822735,
      "step": 1033,
      "time_per_iteration": 2.632786512374878
    },
    {
      "auxiliary_loss_clip": 0.01131701,
      "auxiliary_loss_mlp": 0.01048709,
      "balance_loss_clip": 1.05019569,
      "balance_loss_mlp": 1.03846884,
      "epoch": 0.12433114892082006,
      "flos": 24498834353280.0,
      "grad_norm": 2.1218305339539834,
      "language_loss": 0.97514492,
      "learning_rate": 3.907479442944216e-06,
      "loss": 0.99694896,
      "num_input_tokens_seen": 21841140,
      "step": 1034,
      "time_per_iteration": 3.471487283706665
    },
    {
      "auxiliary_loss_clip": 0.01223738,
      "auxiliary_loss_mlp": 0.01045104,
      "balance_loss_clip": 1.0684098,
      "balance_loss_mlp": 1.03527486,
      "epoch": 0.12445139181145914,
      "flos": 19682315838720.0,
      "grad_norm": 2.3013008268563735,
      "language_loss": 0.92720854,
      "learning_rate": 3.907245113240963e-06,
      "loss": 0.94989699,
      "num_input_tokens_seen": 21859260,
      "step": 1035,
      "time_per_iteration": 3.2621138095855713
    },
    {
      "auxiliary_loss_clip": 0.01192164,
      "auxiliary_loss_mlp": 0.0104114,
      "balance_loss_clip": 1.06099451,
      "balance_loss_mlp": 1.03011334,
      "epoch": 0.12457163470209824,
      "flos": 46423087522560.0,
      "grad_norm": 1.7731773297659463,
      "language_loss": 0.7389465,
      "learning_rate": 3.907010494210144e-06,
      "loss": 0.76127958,
      "num_input_tokens_seen": 21881920,
      "step": 1036,
      "time_per_iteration": 2.8302087783813477
    },
    {
      "auxiliary_loss_clip": 0.01226856,
      "auxiliary_loss_mlp": 0.01040102,
      "balance_loss_clip": 1.06795359,
      "balance_loss_mlp": 1.02981997,
      "epoch": 0.12469187759273732,
      "flos": 20376289578240.0,
      "grad_norm": 2.3599893918052364,
      "language_loss": 0.92430198,
      "learning_rate": 3.9067755858873495e-06,
      "loss": 0.94697154,
      "num_input_tokens_seen": 21898720,
      "step": 1037,
      "time_per_iteration": 2.5211563110351562
    },
    {
      "auxiliary_loss_clip": 0.01097662,
      "auxiliary_loss_mlp": 0.01006655,
      "balance_loss_clip": 1.03346586,
      "balance_loss_mlp": 1.00376976,
      "epoch": 0.12481212048337642,
      "flos": 69224641447680.0,
      "grad_norm": 0.8616946533442321,
      "language_loss": 0.6283766,
      "learning_rate": 3.906540388308214e-06,
      "loss": 0.64941978,
      "num_input_tokens_seen": 21958305,
      "step": 1038,
      "time_per_iteration": 3.9670016765594482
    },
    {
      "auxiliary_loss_clip": 0.01162288,
      "auxiliary_loss_mlp": 0.01033035,
      "balance_loss_clip": 1.0609616,
      "balance_loss_mlp": 1.0230329,
      "epoch": 0.12493236337401552,
      "flos": 18223696350720.0,
      "grad_norm": 1.6842429156144205,
      "language_loss": 0.81365609,
      "learning_rate": 3.906304901508417e-06,
      "loss": 0.83560938,
      "num_input_tokens_seen": 21977205,
      "step": 1039,
      "time_per_iteration": 2.600846290588379
    },
    {
      "auxiliary_loss_clip": 0.01227438,
      "auxiliary_loss_mlp": 0.010384,
      "balance_loss_clip": 1.07246244,
      "balance_loss_mlp": 1.02747488,
      "epoch": 0.12505260626465461,
      "flos": 30044375303040.0,
      "grad_norm": 2.4189520001496287,
      "language_loss": 0.76166576,
      "learning_rate": 3.9060691255236835e-06,
      "loss": 0.78432417,
      "num_input_tokens_seen": 21997770,
      "step": 1040,
      "time_per_iteration": 2.5914833545684814
    },
    {
      "auxiliary_loss_clip": 0.0121671,
      "auxiliary_loss_mlp": 0.01061994,
      "balance_loss_clip": 1.06214261,
      "balance_loss_mlp": 1.05161726,
      "epoch": 0.1251728491552937,
      "flos": 24433980347520.0,
      "grad_norm": 1.7032147333213183,
      "language_loss": 0.80534184,
      "learning_rate": 3.905833060389778e-06,
      "loss": 0.82812893,
      "num_input_tokens_seen": 22021890,
      "step": 1041,
      "time_per_iteration": 2.648988723754883
    },
    {
      "auxiliary_loss_clip": 0.01238007,
      "auxiliary_loss_mlp": 0.00763197,
      "balance_loss_clip": 1.06839764,
      "balance_loss_mlp": 1.0000236,
      "epoch": 0.12529309204593278,
      "flos": 27119809952640.0,
      "grad_norm": 14.611580676521498,
      "language_loss": 0.78505158,
      "learning_rate": 3.905596706142513e-06,
      "loss": 0.80506361,
      "num_input_tokens_seen": 22043300,
      "step": 1042,
      "time_per_iteration": 2.55263614654541
    },
    {
      "auxiliary_loss_clip": 0.01187099,
      "auxiliary_loss_mlp": 0.01039232,
      "balance_loss_clip": 1.06089473,
      "balance_loss_mlp": 1.02934933,
      "epoch": 0.12541333493657186,
      "flos": 30774151923840.0,
      "grad_norm": 2.127013420047938,
      "language_loss": 0.86045521,
      "learning_rate": 3.9053600628177435e-06,
      "loss": 0.8827185,
      "num_input_tokens_seen": 22062910,
      "step": 1043,
      "time_per_iteration": 2.642418146133423
    },
    {
      "auxiliary_loss_clip": 0.01235644,
      "auxiliary_loss_mlp": 0.01035037,
      "balance_loss_clip": 1.06703401,
      "balance_loss_mlp": 1.02561402,
      "epoch": 0.12553357782721097,
      "flos": 23659566099840.0,
      "grad_norm": 2.2044162268617598,
      "language_loss": 0.84380829,
      "learning_rate": 3.905123130451367e-06,
      "loss": 0.86651516,
      "num_input_tokens_seen": 22084010,
      "step": 1044,
      "time_per_iteration": 2.5027005672454834
    },
    {
      "auxiliary_loss_clip": 0.01239681,
      "auxiliary_loss_mlp": 0.01040725,
      "balance_loss_clip": 1.07033145,
      "balance_loss_mlp": 1.03039598,
      "epoch": 0.12565382071785006,
      "flos": 24863758577280.0,
      "grad_norm": 1.8941177059674226,
      "language_loss": 0.79382074,
      "learning_rate": 3.904885909079326e-06,
      "loss": 0.81662488,
      "num_input_tokens_seen": 22102795,
      "step": 1045,
      "time_per_iteration": 2.5005042552948
    },
    {
      "auxiliary_loss_clip": 0.01223366,
      "auxiliary_loss_mlp": 0.01041051,
      "balance_loss_clip": 1.06666172,
      "balance_loss_mlp": 1.03070986,
      "epoch": 0.12577406360848914,
      "flos": 21360780518400.0,
      "grad_norm": 2.507624517364454,
      "language_loss": 0.78011411,
      "learning_rate": 3.904648398737607e-06,
      "loss": 0.80275834,
      "num_input_tokens_seen": 22121360,
      "step": 1046,
      "time_per_iteration": 2.5218708515167236
    },
    {
      "auxiliary_loss_clip": 0.01234951,
      "auxiliary_loss_mlp": 0.01034782,
      "balance_loss_clip": 1.06715405,
      "balance_loss_mlp": 1.02464294,
      "epoch": 0.12589430649912825,
      "flos": 36138056774400.0,
      "grad_norm": 1.9353736721021235,
      "language_loss": 0.7848376,
      "learning_rate": 3.9044105994622406e-06,
      "loss": 0.80753493,
      "num_input_tokens_seen": 22142505,
      "step": 1047,
      "time_per_iteration": 2.5868380069732666
    },
    {
      "auxiliary_loss_clip": 0.01209905,
      "auxiliary_loss_mlp": 0.00763811,
      "balance_loss_clip": 1.06370974,
      "balance_loss_mlp": 1.00001907,
      "epoch": 0.12601454938976733,
      "flos": 25337671643520.0,
      "grad_norm": 1.8846752401754707,
      "language_loss": 0.81593323,
      "learning_rate": 3.9041725112893005e-06,
      "loss": 0.83567035,
      "num_input_tokens_seen": 22163730,
      "step": 1048,
      "time_per_iteration": 2.592961311340332
    },
    {
      "auxiliary_loss_clip": 0.01184319,
      "auxiliary_loss_mlp": 0.01036099,
      "balance_loss_clip": 1.06369758,
      "balance_loss_mlp": 1.02505994,
      "epoch": 0.12613479228040642,
      "flos": 15560094286080.0,
      "grad_norm": 3.8366249066535074,
      "language_loss": 0.75072438,
      "learning_rate": 3.903934134254904e-06,
      "loss": 0.7729286,
      "num_input_tokens_seen": 22181520,
      "step": 1049,
      "time_per_iteration": 2.52116060256958
    },
    {
      "auxiliary_loss_clip": 0.01227174,
      "auxiliary_loss_mlp": 0.01036044,
      "balance_loss_clip": 1.06580555,
      "balance_loss_mlp": 1.02560163,
      "epoch": 0.1262550351710455,
      "flos": 21470595373440.0,
      "grad_norm": 2.1002996944746184,
      "language_loss": 0.84769142,
      "learning_rate": 3.903695468395213e-06,
      "loss": 0.87032366,
      "num_input_tokens_seen": 22199390,
      "step": 1050,
      "time_per_iteration": 2.5044331550598145
    },
    {
      "auxiliary_loss_clip": 0.01210677,
      "auxiliary_loss_mlp": 0.01039978,
      "balance_loss_clip": 1.06258106,
      "balance_loss_mlp": 1.0303582,
      "epoch": 0.1263752780616846,
      "flos": 31576719456000.0,
      "grad_norm": 2.113461247603634,
      "language_loss": 0.55644429,
      "learning_rate": 3.903456513746434e-06,
      "loss": 0.57895082,
      "num_input_tokens_seen": 22220365,
      "step": 1051,
      "time_per_iteration": 2.604372024536133
    },
    {
      "auxiliary_loss_clip": 0.01234611,
      "auxiliary_loss_mlp": 0.01038971,
      "balance_loss_clip": 1.06648862,
      "balance_loss_mlp": 1.02904713,
      "epoch": 0.1264955209523237,
      "flos": 28768217927040.0,
      "grad_norm": 1.8333412222344823,
      "language_loss": 0.87386739,
      "learning_rate": 3.903217270344815e-06,
      "loss": 0.89660329,
      "num_input_tokens_seen": 22240615,
      "step": 1052,
      "time_per_iteration": 2.533623218536377
    },
    {
      "auxiliary_loss_clip": 0.01188388,
      "auxiliary_loss_mlp": 0.0103917,
      "balance_loss_clip": 1.06217551,
      "balance_loss_mlp": 1.02996135,
      "epoch": 0.12661576384296278,
      "flos": 29241125412480.0,
      "grad_norm": 1.7217997555068911,
      "language_loss": 0.82403624,
      "learning_rate": 3.902977738226648e-06,
      "loss": 0.84631181,
      "num_input_tokens_seen": 22261350,
      "step": 1053,
      "time_per_iteration": 2.62927508354187
    },
    {
      "auxiliary_loss_clip": 0.01226531,
      "auxiliary_loss_mlp": 0.01037201,
      "balance_loss_clip": 1.06914687,
      "balance_loss_mlp": 1.02600074,
      "epoch": 0.12673600673360189,
      "flos": 20850346298880.0,
      "grad_norm": 1.8901247559226004,
      "language_loss": 0.90926504,
      "learning_rate": 3.902737917428273e-06,
      "loss": 0.93190235,
      "num_input_tokens_seen": 22279515,
      "step": 1054,
      "time_per_iteration": 2.5070865154266357
    },
    {
      "auxiliary_loss_clip": 0.01235273,
      "auxiliary_loss_mlp": 0.01040541,
      "balance_loss_clip": 1.06722105,
      "balance_loss_mlp": 1.03018808,
      "epoch": 0.12685624962424097,
      "flos": 25263695583360.0,
      "grad_norm": 1.6935110264801863,
      "language_loss": 0.83799517,
      "learning_rate": 3.902497807986068e-06,
      "loss": 0.86075336,
      "num_input_tokens_seen": 22299535,
      "step": 1055,
      "time_per_iteration": 2.501563549041748
    },
    {
      "auxiliary_loss_clip": 0.01194053,
      "auxiliary_loss_mlp": 0.01039177,
      "balance_loss_clip": 1.05978036,
      "balance_loss_mlp": 1.02860951,
      "epoch": 0.12697649251488005,
      "flos": 27527109246720.0,
      "grad_norm": 1.741663241716953,
      "language_loss": 0.8377679,
      "learning_rate": 3.902257409936458e-06,
      "loss": 0.86010021,
      "num_input_tokens_seen": 22320300,
      "step": 1056,
      "time_per_iteration": 2.649744987487793
    },
    {
      "auxiliary_loss_clip": 0.01210589,
      "auxiliary_loss_mlp": 0.01042695,
      "balance_loss_clip": 1.06760371,
      "balance_loss_mlp": 1.03259778,
      "epoch": 0.12709673540551916,
      "flos": 21251863503360.0,
      "grad_norm": 1.9954526018250913,
      "language_loss": 0.83935523,
      "learning_rate": 3.902016723315912e-06,
      "loss": 0.86188811,
      "num_input_tokens_seen": 22338240,
      "step": 1057,
      "time_per_iteration": 2.5422284603118896
    },
    {
      "auxiliary_loss_clip": 0.01217533,
      "auxiliary_loss_mlp": 0.01046021,
      "balance_loss_clip": 1.06248784,
      "balance_loss_mlp": 1.03604937,
      "epoch": 0.12721697829615825,
      "flos": 25337707557120.0,
      "grad_norm": 4.1453064297143944,
      "language_loss": 0.69615984,
      "learning_rate": 3.901775748160941e-06,
      "loss": 0.7187953,
      "num_input_tokens_seen": 22357420,
      "step": 1058,
      "time_per_iteration": 3.2716434001922607
    },
    {
      "auxiliary_loss_clip": 0.01111976,
      "auxiliary_loss_mlp": 0.01009213,
      "balance_loss_clip": 1.04346955,
      "balance_loss_mlp": 1.00648272,
      "epoch": 0.12733722118679733,
      "flos": 61943287754880.0,
      "grad_norm": 0.8154381214633049,
      "language_loss": 0.60871136,
      "learning_rate": 3.901534484508101e-06,
      "loss": 0.62992334,
      "num_input_tokens_seen": 22420095,
      "step": 1059,
      "time_per_iteration": 3.1701419353485107
    },
    {
      "auxiliary_loss_clip": 0.01196456,
      "auxiliary_loss_mlp": 0.01042208,
      "balance_loss_clip": 1.0625875,
      "balance_loss_mlp": 1.03169942,
      "epoch": 0.1274574640774364,
      "flos": 26976742081920.0,
      "grad_norm": 1.9840385864703185,
      "language_loss": 0.74875391,
      "learning_rate": 3.901292932393991e-06,
      "loss": 0.77114058,
      "num_input_tokens_seen": 22438975,
      "step": 1060,
      "time_per_iteration": 3.372079372406006
    },
    {
      "auxiliary_loss_clip": 0.01242457,
      "auxiliary_loss_mlp": 0.01041294,
      "balance_loss_clip": 1.0721159,
      "balance_loss_mlp": 1.03060734,
      "epoch": 0.12757770696807552,
      "flos": 22236318529920.0,
      "grad_norm": 3.0989016028262593,
      "language_loss": 0.85295761,
      "learning_rate": 3.9010510918552555e-06,
      "loss": 0.87579513,
      "num_input_tokens_seen": 22458050,
      "step": 1061,
      "time_per_iteration": 3.159975290298462
    },
    {
      "auxiliary_loss_clip": 0.01205721,
      "auxiliary_loss_mlp": 0.01055676,
      "balance_loss_clip": 1.06359768,
      "balance_loss_mlp": 1.04441035,
      "epoch": 0.1276979498587146,
      "flos": 28547905858560.0,
      "grad_norm": 2.251265036280461,
      "language_loss": 0.74592358,
      "learning_rate": 3.900808962928581e-06,
      "loss": 0.76853752,
      "num_input_tokens_seen": 22475665,
      "step": 1062,
      "time_per_iteration": 2.6017189025878906
    },
    {
      "auxiliary_loss_clip": 0.01239666,
      "auxiliary_loss_mlp": 0.01040674,
      "balance_loss_clip": 1.07162142,
      "balance_loss_mlp": 1.03125095,
      "epoch": 0.1278181927493537,
      "flos": 17420338719360.0,
      "grad_norm": 2.4556067199211773,
      "language_loss": 0.89300942,
      "learning_rate": 3.900566545650698e-06,
      "loss": 0.91581285,
      "num_input_tokens_seen": 22493335,
      "step": 1063,
      "time_per_iteration": 2.456106185913086
    },
    {
      "auxiliary_loss_clip": 0.0122523,
      "auxiliary_loss_mlp": 0.01046825,
      "balance_loss_clip": 1.06893086,
      "balance_loss_mlp": 1.0362215,
      "epoch": 0.1279384356399928,
      "flos": 21138636856320.0,
      "grad_norm": 2.2882648592835912,
      "language_loss": 0.8186565,
      "learning_rate": 3.900323840058381e-06,
      "loss": 0.84137702,
      "num_input_tokens_seen": 22511045,
      "step": 1064,
      "time_per_iteration": 3.282580852508545
    },
    {
      "auxiliary_loss_clip": 0.01220223,
      "auxiliary_loss_mlp": 0.01034484,
      "balance_loss_clip": 1.06271076,
      "balance_loss_mlp": 1.02486968,
      "epoch": 0.12805867853063188,
      "flos": 26576733248640.0,
      "grad_norm": 1.8128715621103602,
      "language_loss": 0.81777501,
      "learning_rate": 3.900080846188449e-06,
      "loss": 0.84032208,
      "num_input_tokens_seen": 22529635,
      "step": 1065,
      "time_per_iteration": 2.559678554534912
    },
    {
      "auxiliary_loss_clip": 0.01239158,
      "auxiliary_loss_mlp": 0.01041953,
      "balance_loss_clip": 1.06922078,
      "balance_loss_mlp": 1.03142071,
      "epoch": 0.12817892142127096,
      "flos": 16436206915200.0,
      "grad_norm": 1.8277312969692305,
      "language_loss": 0.8130101,
      "learning_rate": 3.8998375640777625e-06,
      "loss": 0.83582121,
      "num_input_tokens_seen": 22547505,
      "step": 1066,
      "time_per_iteration": 2.471086025238037
    },
    {
      "auxiliary_loss_clip": 0.01100713,
      "auxiliary_loss_mlp": 0.01003675,
      "balance_loss_clip": 1.03002143,
      "balance_loss_mlp": 1.00096846,
      "epoch": 0.12829916431191005,
      "flos": 60757049099520.0,
      "grad_norm": 0.7055097184507033,
      "language_loss": 0.52666527,
      "learning_rate": 3.899593993763229e-06,
      "loss": 0.54770911,
      "num_input_tokens_seen": 22608465,
      "step": 1067,
      "time_per_iteration": 3.065671443939209
    },
    {
      "auxiliary_loss_clip": 0.01185337,
      "auxiliary_loss_mlp": 0.01043484,
      "balance_loss_clip": 1.0615418,
      "balance_loss_mlp": 1.03236735,
      "epoch": 0.12841940720254916,
      "flos": 29786895636480.0,
      "grad_norm": 2.905354981767814,
      "language_loss": 0.81025183,
      "learning_rate": 3.899350135281796e-06,
      "loss": 0.83254004,
      "num_input_tokens_seen": 22629465,
      "step": 1068,
      "time_per_iteration": 2.6390085220336914
    },
    {
      "auxiliary_loss_clip": 0.01198111,
      "auxiliary_loss_mlp": 0.010461,
      "balance_loss_clip": 1.06490064,
      "balance_loss_mlp": 1.03622901,
      "epoch": 0.12853965009318824,
      "flos": 25951851319680.0,
      "grad_norm": 2.030057409363176,
      "language_loss": 0.79829299,
      "learning_rate": 3.8991059886704585e-06,
      "loss": 0.8207351,
      "num_input_tokens_seen": 22648970,
      "step": 1069,
      "time_per_iteration": 2.5977070331573486
    },
    {
      "auxiliary_loss_clip": 0.01187355,
      "auxiliary_loss_mlp": 0.01042341,
      "balance_loss_clip": 1.06336772,
      "balance_loss_mlp": 1.03122497,
      "epoch": 0.12865989298382732,
      "flos": 30846871008000.0,
      "grad_norm": 1.9402292555019003,
      "language_loss": 0.83154994,
      "learning_rate": 3.898861553966252e-06,
      "loss": 0.85384691,
      "num_input_tokens_seen": 22668620,
      "step": 1070,
      "time_per_iteration": 2.651322603225708
    },
    {
      "auxiliary_loss_clip": 0.01145557,
      "auxiliary_loss_mlp": 0.01048347,
      "balance_loss_clip": 1.05353045,
      "balance_loss_mlp": 1.03677213,
      "epoch": 0.12878013587446643,
      "flos": 25885776251520.0,
      "grad_norm": 1.8593134196282446,
      "language_loss": 0.88117647,
      "learning_rate": 3.898616831206257e-06,
      "loss": 0.90311563,
      "num_input_tokens_seen": 22689045,
      "step": 1071,
      "time_per_iteration": 2.7072017192840576
    },
    {
      "auxiliary_loss_clip": 0.01187902,
      "auxiliary_loss_mlp": 0.0104513,
      "balance_loss_clip": 1.05970454,
      "balance_loss_mlp": 1.03456807,
      "epoch": 0.12890037876510552,
      "flos": 23333138277120.0,
      "grad_norm": 2.5660865153847987,
      "language_loss": 0.76886845,
      "learning_rate": 3.8983718204276e-06,
      "loss": 0.79119885,
      "num_input_tokens_seen": 22711265,
      "step": 1072,
      "time_per_iteration": 2.6430046558380127
    },
    {
      "auxiliary_loss_clip": 0.01207625,
      "auxiliary_loss_mlp": 0.01049004,
      "balance_loss_clip": 1.06569839,
      "balance_loss_mlp": 1.03885889,
      "epoch": 0.1290206216557446,
      "flos": 23587242065280.0,
      "grad_norm": 1.7357677556958448,
      "language_loss": 0.82732749,
      "learning_rate": 3.898126521667446e-06,
      "loss": 0.84989381,
      "num_input_tokens_seen": 22731420,
      "step": 1073,
      "time_per_iteration": 2.5688769817352295
    },
    {
      "auxiliary_loss_clip": 0.01225172,
      "auxiliary_loss_mlp": 0.01042169,
      "balance_loss_clip": 1.06762969,
      "balance_loss_mlp": 1.03099346,
      "epoch": 0.12914086454638368,
      "flos": 24170610850560.0,
      "grad_norm": 1.6040887869025755,
      "language_loss": 0.8330344,
      "learning_rate": 3.897880934963007e-06,
      "loss": 0.85570788,
      "num_input_tokens_seen": 22750970,
      "step": 1074,
      "time_per_iteration": 2.5432510375976562
    },
    {
      "auxiliary_loss_clip": 0.01203278,
      "auxiliary_loss_mlp": 0.01040597,
      "balance_loss_clip": 1.06255579,
      "balance_loss_mlp": 1.03034496,
      "epoch": 0.1292611074370228,
      "flos": 20267157081600.0,
      "grad_norm": 2.223900936652385,
      "language_loss": 0.7885235,
      "learning_rate": 3.89763506035154e-06,
      "loss": 0.8109622,
      "num_input_tokens_seen": 22768820,
      "step": 1075,
      "time_per_iteration": 2.55460524559021
    },
    {
      "auxiliary_loss_clip": 0.01208227,
      "auxiliary_loss_mlp": 0.01042921,
      "balance_loss_clip": 1.06471598,
      "balance_loss_mlp": 1.03299665,
      "epoch": 0.12938135032766188,
      "flos": 27377684668800.0,
      "grad_norm": 1.9846482448184568,
      "language_loss": 0.81389189,
      "learning_rate": 3.897388897870343e-06,
      "loss": 0.83640337,
      "num_input_tokens_seen": 22789460,
      "step": 1076,
      "time_per_iteration": 2.5815446376800537
    },
    {
      "auxiliary_loss_clip": 0.01218741,
      "auxiliary_loss_mlp": 0.01038614,
      "balance_loss_clip": 1.06675339,
      "balance_loss_mlp": 1.02762282,
      "epoch": 0.12950159321830096,
      "flos": 29277107861760.0,
      "grad_norm": 2.010272583759834,
      "language_loss": 0.74907184,
      "learning_rate": 3.89714244755676e-06,
      "loss": 0.77164537,
      "num_input_tokens_seen": 22810820,
      "step": 1077,
      "time_per_iteration": 2.6025137901306152
    },
    {
      "auxiliary_loss_clip": 0.01158042,
      "auxiliary_loss_mlp": 0.01046506,
      "balance_loss_clip": 1.0539825,
      "balance_loss_mlp": 1.03581285,
      "epoch": 0.12962183610894007,
      "flos": 24534888629760.0,
      "grad_norm": 2.788753932859496,
      "language_loss": 0.86120832,
      "learning_rate": 3.896895709448175e-06,
      "loss": 0.88325381,
      "num_input_tokens_seen": 22830570,
      "step": 1078,
      "time_per_iteration": 2.6355793476104736
    },
    {
      "auxiliary_loss_clip": 0.01150097,
      "auxiliary_loss_mlp": 0.01039747,
      "balance_loss_clip": 1.05335402,
      "balance_loss_mlp": 1.0289706,
      "epoch": 0.12974207899957915,
      "flos": 11215944552960.0,
      "grad_norm": 2.506821613467787,
      "language_loss": 0.7715832,
      "learning_rate": 3.896648683582019e-06,
      "loss": 0.79348165,
      "num_input_tokens_seen": 22845905,
      "step": 1079,
      "time_per_iteration": 2.6121151447296143
    },
    {
      "auxiliary_loss_clip": 0.01175785,
      "auxiliary_loss_mlp": 0.01037341,
      "balance_loss_clip": 1.06349158,
      "balance_loss_mlp": 1.0271368,
      "epoch": 0.12986232189021824,
      "flos": 24717889445760.0,
      "grad_norm": 2.077727509405393,
      "language_loss": 0.81146133,
      "learning_rate": 3.896401369995766e-06,
      "loss": 0.83359259,
      "num_input_tokens_seen": 22865710,
      "step": 1080,
      "time_per_iteration": 2.652203321456909
    },
    {
      "auxiliary_loss_clip": 0.01235741,
      "auxiliary_loss_mlp": 0.01048084,
      "balance_loss_clip": 1.0676527,
      "balance_loss_mlp": 1.03718805,
      "epoch": 0.12998256478085732,
      "flos": 23915357827200.0,
      "grad_norm": 1.8472068778746615,
      "language_loss": 0.79423982,
      "learning_rate": 3.896153768726932e-06,
      "loss": 0.81707811,
      "num_input_tokens_seen": 22886020,
      "step": 1081,
      "time_per_iteration": 2.5074172019958496
    },
    {
      "auxiliary_loss_clip": 0.01219518,
      "auxiliary_loss_mlp": 0.01041225,
      "balance_loss_clip": 1.06879735,
      "balance_loss_mlp": 1.03098512,
      "epoch": 0.13010280767149643,
      "flos": 18624207974400.0,
      "grad_norm": 3.457752169268043,
      "language_loss": 0.87808055,
      "learning_rate": 3.8959058798130806e-06,
      "loss": 0.90068793,
      "num_input_tokens_seen": 22903995,
      "step": 1082,
      "time_per_iteration": 2.500200033187866
    },
    {
      "auxiliary_loss_clip": 0.01204605,
      "auxiliary_loss_mlp": 0.00764392,
      "balance_loss_clip": 1.06292176,
      "balance_loss_mlp": 1.00007427,
      "epoch": 0.1302230505621355,
      "flos": 22783992174720.0,
      "grad_norm": 1.671046759421625,
      "language_loss": 0.7482087,
      "learning_rate": 3.895657703291814e-06,
      "loss": 0.76789868,
      "num_input_tokens_seen": 22924100,
      "step": 1083,
      "time_per_iteration": 2.5692949295043945
    },
    {
      "auxiliary_loss_clip": 0.01213587,
      "auxiliary_loss_mlp": 0.0104095,
      "balance_loss_clip": 1.06223917,
      "balance_loss_mlp": 1.02898169,
      "epoch": 0.1303432934527746,
      "flos": 21323612920320.0,
      "grad_norm": 3.09108124320713,
      "language_loss": 0.78934336,
      "learning_rate": 3.895409239200781e-06,
      "loss": 0.81188869,
      "num_input_tokens_seen": 22939985,
      "step": 1084,
      "time_per_iteration": 3.2918388843536377
    },
    {
      "auxiliary_loss_clip": 0.01214201,
      "auxiliary_loss_mlp": 0.01037566,
      "balance_loss_clip": 1.06372011,
      "balance_loss_mlp": 1.02671838,
      "epoch": 0.1304635363434137,
      "flos": 20922490765440.0,
      "grad_norm": 2.6420160555123697,
      "language_loss": 0.91216254,
      "learning_rate": 3.895160487577673e-06,
      "loss": 0.93468016,
      "num_input_tokens_seen": 22957555,
      "step": 1085,
      "time_per_iteration": 2.5015358924865723
    },
    {
      "auxiliary_loss_clip": 0.01117955,
      "auxiliary_loss_mlp": 0.01003347,
      "balance_loss_clip": 1.0354588,
      "balance_loss_mlp": 1.0004977,
      "epoch": 0.1305837792340528,
      "flos": 63245659080960.0,
      "grad_norm": 0.7826108519944255,
      "language_loss": 0.60902047,
      "learning_rate": 3.894911448460226e-06,
      "loss": 0.63023353,
      "num_input_tokens_seen": 23016870,
      "step": 1086,
      "time_per_iteration": 3.7135608196258545
    },
    {
      "auxiliary_loss_clip": 0.01126396,
      "auxiliary_loss_mlp": 0.01039791,
      "balance_loss_clip": 1.05444038,
      "balance_loss_mlp": 1.02948511,
      "epoch": 0.13070402212469187,
      "flos": 26428852955520.0,
      "grad_norm": 3.093885844408734,
      "language_loss": 0.72794288,
      "learning_rate": 3.8946621218862195e-06,
      "loss": 0.7496047,
      "num_input_tokens_seen": 23037870,
      "step": 1087,
      "time_per_iteration": 3.542111396789551
    },
    {
      "auxiliary_loss_clip": 0.01187018,
      "auxiliary_loss_mlp": 0.01045685,
      "balance_loss_clip": 1.06092954,
      "balance_loss_mlp": 1.0347594,
      "epoch": 0.13082426501533098,
      "flos": 27673409341440.0,
      "grad_norm": 2.0505530708368536,
      "language_loss": 0.89069784,
      "learning_rate": 3.894412507893475e-06,
      "loss": 0.91302478,
      "num_input_tokens_seen": 23058150,
      "step": 1088,
      "time_per_iteration": 2.645631790161133
    },
    {
      "auxiliary_loss_clip": 0.01181856,
      "auxiliary_loss_mlp": 0.01041655,
      "balance_loss_clip": 1.06087172,
      "balance_loss_mlp": 1.03062272,
      "epoch": 0.13094450790597006,
      "flos": 24826770547200.0,
      "grad_norm": 1.997153205952805,
      "language_loss": 0.72069687,
      "learning_rate": 3.894162606519859e-06,
      "loss": 0.74293196,
      "num_input_tokens_seen": 23077100,
      "step": 1089,
      "time_per_iteration": 2.639298439025879
    },
    {
      "auxiliary_loss_clip": 0.01170312,
      "auxiliary_loss_mlp": 0.01033489,
      "balance_loss_clip": 1.05938792,
      "balance_loss_mlp": 1.02311826,
      "epoch": 0.13106475079660915,
      "flos": 19062605468160.0,
      "grad_norm": 2.0633290717525665,
      "language_loss": 0.77350956,
      "learning_rate": 3.893912417803282e-06,
      "loss": 0.7955476,
      "num_input_tokens_seen": 23096815,
      "step": 1090,
      "time_per_iteration": 3.3392324447631836
    },
    {
      "auxiliary_loss_clip": 0.01169929,
      "auxiliary_loss_mlp": 0.01039163,
      "balance_loss_clip": 1.05380213,
      "balance_loss_mlp": 1.02805877,
      "epoch": 0.13118499368724823,
      "flos": 28913189218560.0,
      "grad_norm": 1.8559563495343867,
      "language_loss": 0.76453829,
      "learning_rate": 3.8936619417816975e-06,
      "loss": 0.7866292,
      "num_input_tokens_seen": 23117145,
      "step": 1091,
      "time_per_iteration": 2.6592886447906494
    },
    {
      "auxiliary_loss_clip": 0.01191227,
      "auxiliary_loss_mlp": 0.01049684,
      "balance_loss_clip": 1.06508636,
      "balance_loss_mlp": 1.03860319,
      "epoch": 0.13130523657788734,
      "flos": 14283398206080.0,
      "grad_norm": 1.8131272923169133,
      "language_loss": 0.71543336,
      "learning_rate": 3.8934111784931015e-06,
      "loss": 0.73784244,
      "num_input_tokens_seen": 23134595,
      "step": 1092,
      "time_per_iteration": 2.555454730987549
    },
    {
      "auxiliary_loss_clip": 0.01104841,
      "auxiliary_loss_mlp": 0.0100439,
      "balance_loss_clip": 1.03043973,
      "balance_loss_mlp": 1.0016129,
      "epoch": 0.13142547946852642,
      "flos": 70174155519360.0,
      "grad_norm": 0.990191517391737,
      "language_loss": 0.59061873,
      "learning_rate": 3.893160127975535e-06,
      "loss": 0.61171103,
      "num_input_tokens_seen": 23195285,
      "step": 1093,
      "time_per_iteration": 3.2444469928741455
    },
    {
      "auxiliary_loss_clip": 0.01180352,
      "auxiliary_loss_mlp": 0.01038853,
      "balance_loss_clip": 1.05835319,
      "balance_loss_mlp": 1.02877378,
      "epoch": 0.1315457223591655,
      "flos": 45805998844800.0,
      "grad_norm": 2.2641655821412545,
      "language_loss": 0.81359255,
      "learning_rate": 3.8929087902670826e-06,
      "loss": 0.83578455,
      "num_input_tokens_seen": 23216915,
      "step": 1094,
      "time_per_iteration": 2.8174192905426025
    },
    {
      "auxiliary_loss_clip": 0.01115483,
      "auxiliary_loss_mlp": 0.01003374,
      "balance_loss_clip": 1.0285337,
      "balance_loss_mlp": 1.0005368,
      "epoch": 0.13166596524980462,
      "flos": 62881165820160.0,
      "grad_norm": 0.9259716268126945,
      "language_loss": 0.60692918,
      "learning_rate": 3.8926571654058715e-06,
      "loss": 0.6281178,
      "num_input_tokens_seen": 23273560,
      "step": 1095,
      "time_per_iteration": 3.031236171722412
    },
    {
      "auxiliary_loss_clip": 0.01188913,
      "auxiliary_loss_mlp": 0.01041946,
      "balance_loss_clip": 1.06352246,
      "balance_loss_mlp": 1.03181326,
      "epoch": 0.1317862081404437,
      "flos": 23586523793280.0,
      "grad_norm": 3.8247342624576754,
      "language_loss": 0.77083004,
      "learning_rate": 3.892405253430074e-06,
      "loss": 0.79313862,
      "num_input_tokens_seen": 23291080,
      "step": 1096,
      "time_per_iteration": 2.707878351211548
    },
    {
      "auxiliary_loss_clip": 0.0121278,
      "auxiliary_loss_mlp": 0.00764253,
      "balance_loss_clip": 1.06629777,
      "balance_loss_mlp": 1.00004077,
      "epoch": 0.13190645103108278,
      "flos": 20260764460800.0,
      "grad_norm": 2.432152014739372,
      "language_loss": 0.82273233,
      "learning_rate": 3.892153054377904e-06,
      "loss": 0.84250265,
      "num_input_tokens_seen": 23308485,
      "step": 1097,
      "time_per_iteration": 2.5972392559051514
    },
    {
      "auxiliary_loss_clip": 0.01054078,
      "auxiliary_loss_mlp": 0.0100616,
      "balance_loss_clip": 1.02388835,
      "balance_loss_mlp": 1.00325131,
      "epoch": 0.13202669392172187,
      "flos": 53455440136320.0,
      "grad_norm": 0.9991126831087038,
      "language_loss": 0.59444469,
      "learning_rate": 3.891900568287619e-06,
      "loss": 0.61504704,
      "num_input_tokens_seen": 23360870,
      "step": 1098,
      "time_per_iteration": 3.1282007694244385
    },
    {
      "auxiliary_loss_clip": 0.01192207,
      "auxiliary_loss_mlp": 0.01040911,
      "balance_loss_clip": 1.05981302,
      "balance_loss_mlp": 1.0293839,
      "epoch": 0.13214693681236098,
      "flos": 15851293845120.0,
      "grad_norm": 2.707852829563897,
      "language_loss": 0.72586048,
      "learning_rate": 3.891647795197523e-06,
      "loss": 0.7481916,
      "num_input_tokens_seen": 23376910,
      "step": 1099,
      "time_per_iteration": 2.841768741607666
    },
    {
      "auxiliary_loss_clip": 0.01194541,
      "auxiliary_loss_mlp": 0.01042848,
      "balance_loss_clip": 1.05951071,
      "balance_loss_mlp": 1.03201747,
      "epoch": 0.13226717970300006,
      "flos": 19353840940800.0,
      "grad_norm": 2.1859448089054485,
      "language_loss": 0.68508744,
      "learning_rate": 3.8913947351459605e-06,
      "loss": 0.70746136,
      "num_input_tokens_seen": 23394450,
      "step": 1100,
      "time_per_iteration": 2.5741093158721924
    },
    {
      "auxiliary_loss_clip": 0.01236321,
      "auxiliary_loss_mlp": 0.01038801,
      "balance_loss_clip": 1.06783998,
      "balance_loss_mlp": 1.02913857,
      "epoch": 0.13238742259363914,
      "flos": 20698084546560.0,
      "grad_norm": 1.8962882495268178,
      "language_loss": 0.67583054,
      "learning_rate": 3.89114138817132e-06,
      "loss": 0.6985817,
      "num_input_tokens_seen": 23411115,
      "step": 1101,
      "time_per_iteration": 2.493234634399414
    },
    {
      "auxiliary_loss_clip": 0.01219882,
      "auxiliary_loss_mlp": 0.01041359,
      "balance_loss_clip": 1.06832194,
      "balance_loss_mlp": 1.03082693,
      "epoch": 0.13250766548427825,
      "flos": 21032449274880.0,
      "grad_norm": 1.9517497232671868,
      "language_loss": 0.84359211,
      "learning_rate": 3.890887754312035e-06,
      "loss": 0.8662045,
      "num_input_tokens_seen": 23429360,
      "step": 1102,
      "time_per_iteration": 2.4972715377807617
    },
    {
      "auxiliary_loss_clip": 0.01199824,
      "auxiliary_loss_mlp": 0.01038425,
      "balance_loss_clip": 1.06074739,
      "balance_loss_mlp": 1.02841735,
      "epoch": 0.13262790837491734,
      "flos": 22637871648000.0,
      "grad_norm": 1.7829331486085258,
      "language_loss": 0.87633824,
      "learning_rate": 3.890633833606581e-06,
      "loss": 0.89872074,
      "num_input_tokens_seen": 23449050,
      "step": 1103,
      "time_per_iteration": 2.5372328758239746
    },
    {
      "auxiliary_loss_clip": 0.01222158,
      "auxiliary_loss_mlp": 0.01041444,
      "balance_loss_clip": 1.07158792,
      "balance_loss_mlp": 1.03104854,
      "epoch": 0.13274815126555642,
      "flos": 19683141851520.0,
      "grad_norm": 1.7276098298992784,
      "language_loss": 0.69412512,
      "learning_rate": 3.890379626093477e-06,
      "loss": 0.71676111,
      "num_input_tokens_seen": 23468800,
      "step": 1104,
      "time_per_iteration": 2.506258487701416
    },
    {
      "auxiliary_loss_clip": 0.0116264,
      "auxiliary_loss_mlp": 0.01032905,
      "balance_loss_clip": 1.05792308,
      "balance_loss_mlp": 1.02291548,
      "epoch": 0.1328683941561955,
      "flos": 21317687176320.0,
      "grad_norm": 2.8742001804637454,
      "language_loss": 0.92538899,
      "learning_rate": 3.890125131811287e-06,
      "loss": 0.94734442,
      "num_input_tokens_seen": 23486850,
      "step": 1105,
      "time_per_iteration": 2.6051981449127197
    },
    {
      "auxiliary_loss_clip": 0.01189605,
      "auxiliary_loss_mlp": 0.01037476,
      "balance_loss_clip": 1.05710018,
      "balance_loss_mlp": 1.02773702,
      "epoch": 0.1329886370468346,
      "flos": 13699131580800.0,
      "grad_norm": 2.2387493368120492,
      "language_loss": 0.75461066,
      "learning_rate": 3.889870350798618e-06,
      "loss": 0.77688146,
      "num_input_tokens_seen": 23504195,
      "step": 1106,
      "time_per_iteration": 2.520110607147217
    },
    {
      "auxiliary_loss_clip": 0.01234204,
      "auxiliary_loss_mlp": 0.01039318,
      "balance_loss_clip": 1.06679511,
      "balance_loss_mlp": 1.02877378,
      "epoch": 0.1331088799374737,
      "flos": 21032413361280.0,
      "grad_norm": 1.5903168467201718,
      "language_loss": 0.78506261,
      "learning_rate": 3.889615283094119e-06,
      "loss": 0.80779779,
      "num_input_tokens_seen": 23523385,
      "step": 1107,
      "time_per_iteration": 2.4824838638305664
    },
    {
      "auxiliary_loss_clip": 0.01239381,
      "auxiliary_loss_mlp": 0.01042967,
      "balance_loss_clip": 1.06683755,
      "balance_loss_mlp": 1.03141594,
      "epoch": 0.13322912282811278,
      "flos": 18260432985600.0,
      "grad_norm": 1.991906230940626,
      "language_loss": 0.84380919,
      "learning_rate": 3.889359928736485e-06,
      "loss": 0.8666327,
      "num_input_tokens_seen": 23541330,
      "step": 1108,
      "time_per_iteration": 2.462399959564209
    },
    {
      "auxiliary_loss_clip": 0.01198381,
      "auxiliary_loss_mlp": 0.00764162,
      "balance_loss_clip": 1.06358278,
      "balance_loss_mlp": 1.00002074,
      "epoch": 0.1333493657187519,
      "flos": 24460876656000.0,
      "grad_norm": 2.165911575523885,
      "language_loss": 0.90957469,
      "learning_rate": 3.889104287764451e-06,
      "loss": 0.92920011,
      "num_input_tokens_seen": 23561705,
      "step": 1109,
      "time_per_iteration": 2.594787836074829
    },
    {
      "auxiliary_loss_clip": 0.0120574,
      "auxiliary_loss_mlp": 0.01041373,
      "balance_loss_clip": 1.06926918,
      "balance_loss_mlp": 1.03079331,
      "epoch": 0.13346960860939097,
      "flos": 22158930677760.0,
      "grad_norm": 1.9917991605200263,
      "language_loss": 0.90840089,
      "learning_rate": 3.888848360216798e-06,
      "loss": 0.93087208,
      "num_input_tokens_seen": 23579350,
      "step": 1110,
      "time_per_iteration": 3.3396666049957275
    },
    {
      "auxiliary_loss_clip": 0.01100919,
      "auxiliary_loss_mlp": 0.01006612,
      "balance_loss_clip": 1.02441216,
      "balance_loss_mlp": 1.00371504,
      "epoch": 0.13358985150003005,
      "flos": 67931212608000.0,
      "grad_norm": 0.7895816040523768,
      "language_loss": 0.56576914,
      "learning_rate": 3.888592146132351e-06,
      "loss": 0.5868445,
      "num_input_tokens_seen": 23640620,
      "step": 1111,
      "time_per_iteration": 3.2206993103027344
    },
    {
      "auxiliary_loss_clip": 0.01221945,
      "auxiliary_loss_mlp": 0.01036491,
      "balance_loss_clip": 1.0686655,
      "balance_loss_mlp": 1.02615571,
      "epoch": 0.13371009439066917,
      "flos": 26834284742400.0,
      "grad_norm": 2.083562897806881,
      "language_loss": 0.78610349,
      "learning_rate": 3.888335645549978e-06,
      "loss": 0.80868787,
      "num_input_tokens_seen": 23661040,
      "step": 1112,
      "time_per_iteration": 3.315871238708496
    },
    {
      "auxiliary_loss_clip": 0.01238626,
      "auxiliary_loss_mlp": 0.01042496,
      "balance_loss_clip": 1.07068491,
      "balance_loss_mlp": 1.03269696,
      "epoch": 0.13383033728130825,
      "flos": 26322844942080.0,
      "grad_norm": 2.3037942782484135,
      "language_loss": 0.81557512,
      "learning_rate": 3.888078858508588e-06,
      "loss": 0.83838642,
      "num_input_tokens_seen": 23680900,
      "step": 1113,
      "time_per_iteration": 3.2923686504364014
    },
    {
      "auxiliary_loss_clip": 0.01205822,
      "auxiliary_loss_mlp": 0.01041946,
      "balance_loss_clip": 1.0692637,
      "balance_loss_mlp": 1.03245056,
      "epoch": 0.13395058017194733,
      "flos": 22563931501440.0,
      "grad_norm": 1.6757867007411082,
      "language_loss": 0.84435284,
      "learning_rate": 3.8878217850471365e-06,
      "loss": 0.86683047,
      "num_input_tokens_seen": 23700815,
      "step": 1114,
      "time_per_iteration": 2.63095760345459
    },
    {
      "auxiliary_loss_clip": 0.01243009,
      "auxiliary_loss_mlp": 0.01048852,
      "balance_loss_clip": 1.07256651,
      "balance_loss_mlp": 1.03854585,
      "epoch": 0.13407082306258641,
      "flos": 25810938264960.0,
      "grad_norm": 1.8682284066975048,
      "language_loss": 0.73856801,
      "learning_rate": 3.887564425204621e-06,
      "loss": 0.76148665,
      "num_input_tokens_seen": 23722500,
      "step": 1115,
      "time_per_iteration": 2.524057626724243
    },
    {
      "auxiliary_loss_clip": 0.01082079,
      "auxiliary_loss_mlp": 0.01002566,
      "balance_loss_clip": 1.02439404,
      "balance_loss_mlp": 0.9995262,
      "epoch": 0.13419106595322552,
      "flos": 68338365269760.0,
      "grad_norm": 0.8576483780228971,
      "language_loss": 0.54734468,
      "learning_rate": 3.887306779020083e-06,
      "loss": 0.56819117,
      "num_input_tokens_seen": 23777155,
      "step": 1116,
      "time_per_iteration": 3.838433265686035
    },
    {
      "auxiliary_loss_clip": 0.01225783,
      "auxiliary_loss_mlp": 0.01040903,
      "balance_loss_clip": 1.06887877,
      "balance_loss_mlp": 1.0300138,
      "epoch": 0.1343113088438646,
      "flos": 20449080489600.0,
      "grad_norm": 2.3780901386180773,
      "language_loss": 0.70458007,
      "learning_rate": 3.887048846532608e-06,
      "loss": 0.727247,
      "num_input_tokens_seen": 23794130,
      "step": 1117,
      "time_per_iteration": 2.5200228691101074
    },
    {
      "auxiliary_loss_clip": 0.01082167,
      "auxiliary_loss_mlp": 0.01003955,
      "balance_loss_clip": 1.020033,
      "balance_loss_mlp": 1.00120151,
      "epoch": 0.1344315517345037,
      "flos": 67389784951680.0,
      "grad_norm": 0.7615194366654615,
      "language_loss": 0.5816226,
      "learning_rate": 3.8867906277813224e-06,
      "loss": 0.60248387,
      "num_input_tokens_seen": 23852285,
      "step": 1118,
      "time_per_iteration": 3.0330288410186768
    },
    {
      "auxiliary_loss_clip": 0.01228704,
      "auxiliary_loss_mlp": 0.00763465,
      "balance_loss_clip": 1.07133436,
      "balance_loss_mlp": 1.00006819,
      "epoch": 0.1345517946251428,
      "flos": 40734442788480.0,
      "grad_norm": 2.2327492687396324,
      "language_loss": 0.73994875,
      "learning_rate": 3.886532122805399e-06,
      "loss": 0.75987041,
      "num_input_tokens_seen": 23874765,
      "step": 1119,
      "time_per_iteration": 2.670726776123047
    },
    {
      "auxiliary_loss_clip": 0.01143759,
      "auxiliary_loss_mlp": 0.01040175,
      "balance_loss_clip": 1.05410326,
      "balance_loss_mlp": 1.02834988,
      "epoch": 0.13467203751578188,
      "flos": 22816850140800.0,
      "grad_norm": 2.035678919470623,
      "language_loss": 0.896725,
      "learning_rate": 3.886273331644053e-06,
      "loss": 0.91856432,
      "num_input_tokens_seen": 23893635,
      "step": 1120,
      "time_per_iteration": 2.6622493267059326
    },
    {
      "auxiliary_loss_clip": 0.01176451,
      "auxiliary_loss_mlp": 0.01038946,
      "balance_loss_clip": 1.06515288,
      "balance_loss_mlp": 1.02870011,
      "epoch": 0.13479228040642097,
      "flos": 17091576512640.0,
      "grad_norm": 2.230809638794972,
      "language_loss": 0.82186097,
      "learning_rate": 3.886014254336542e-06,
      "loss": 0.84401494,
      "num_input_tokens_seen": 23910110,
      "step": 1121,
      "time_per_iteration": 2.580785036087036
    },
    {
      "auxiliary_loss_clip": 0.01217711,
      "auxiliary_loss_mlp": 0.0103449,
      "balance_loss_clip": 1.06384087,
      "balance_loss_mlp": 1.02427387,
      "epoch": 0.13491252329706005,
      "flos": 23730525417600.0,
      "grad_norm": 1.7950515985489024,
      "language_loss": 0.9277209,
      "learning_rate": 3.885754890922168e-06,
      "loss": 0.95024294,
      "num_input_tokens_seen": 23930440,
      "step": 1122,
      "time_per_iteration": 2.546475410461426
    },
    {
      "auxiliary_loss_clip": 0.01134924,
      "auxiliary_loss_mlp": 0.01034934,
      "balance_loss_clip": 1.05812025,
      "balance_loss_mlp": 1.02390158,
      "epoch": 0.13503276618769916,
      "flos": 34127058960000.0,
      "grad_norm": 1.7578511347297217,
      "language_loss": 0.78192395,
      "learning_rate": 3.885495241440277e-06,
      "loss": 0.80362248,
      "num_input_tokens_seen": 23954535,
      "step": 1123,
      "time_per_iteration": 2.9330544471740723
    },
    {
      "auxiliary_loss_clip": 0.01238627,
      "auxiliary_loss_mlp": 0.01042118,
      "balance_loss_clip": 1.06967354,
      "balance_loss_mlp": 1.03066242,
      "epoch": 0.13515300907833824,
      "flos": 17712328377600.0,
      "grad_norm": 1.7088415671969657,
      "language_loss": 0.74298728,
      "learning_rate": 3.885235305930257e-06,
      "loss": 0.76579475,
      "num_input_tokens_seen": 23972735,
      "step": 1124,
      "time_per_iteration": 2.775200366973877
    },
    {
      "auxiliary_loss_clip": 0.01190843,
      "auxiliary_loss_mlp": 0.01048259,
      "balance_loss_clip": 1.06791425,
      "balance_loss_mlp": 1.03707731,
      "epoch": 0.13527325196897733,
      "flos": 20260872201600.0,
      "grad_norm": 1.9254887649025159,
      "language_loss": 0.85758221,
      "learning_rate": 3.884975084431539e-06,
      "loss": 0.87997329,
      "num_input_tokens_seen": 23987685,
      "step": 1125,
      "time_per_iteration": 2.5701098442077637
    },
    {
      "auxiliary_loss_clip": 0.01214913,
      "auxiliary_loss_mlp": 0.0076362,
      "balance_loss_clip": 1.06769311,
      "balance_loss_mlp": 1.00002646,
      "epoch": 0.13539349485961644,
      "flos": 18186492839040.0,
      "grad_norm": 2.7822546174349254,
      "language_loss": 0.9173485,
      "learning_rate": 3.8847145769836e-06,
      "loss": 0.93713379,
      "num_input_tokens_seen": 24004105,
      "step": 1126,
      "time_per_iteration": 2.4926645755767822
    },
    {
      "auxiliary_loss_clip": 0.01243262,
      "auxiliary_loss_mlp": 0.01044506,
      "balance_loss_clip": 1.07253468,
      "balance_loss_mlp": 1.03313386,
      "epoch": 0.13551373775025552,
      "flos": 19317463441920.0,
      "grad_norm": 3.0543462769511254,
      "language_loss": 0.66461194,
      "learning_rate": 3.884453783625959e-06,
      "loss": 0.68748963,
      "num_input_tokens_seen": 24021715,
      "step": 1127,
      "time_per_iteration": 2.468153715133667
    },
    {
      "auxiliary_loss_clip": 0.01202113,
      "auxiliary_loss_mlp": 0.01032469,
      "balance_loss_clip": 1.06510186,
      "balance_loss_mlp": 1.02213311,
      "epoch": 0.1356339806408946,
      "flos": 20850813175680.0,
      "grad_norm": 2.851723123638947,
      "language_loss": 0.84840971,
      "learning_rate": 3.884192704398176e-06,
      "loss": 0.87075555,
      "num_input_tokens_seen": 24038915,
      "step": 1128,
      "time_per_iteration": 2.5372021198272705
    },
    {
      "auxiliary_loss_clip": 0.01226204,
      "auxiliary_loss_mlp": 0.01043395,
      "balance_loss_clip": 1.06909537,
      "balance_loss_mlp": 1.03320897,
      "epoch": 0.13575422353153369,
      "flos": 50476037696640.0,
      "grad_norm": 1.6816622764400155,
      "language_loss": 0.74245942,
      "learning_rate": 3.883931339339858e-06,
      "loss": 0.76515543,
      "num_input_tokens_seen": 24063300,
      "step": 1129,
      "time_per_iteration": 2.783559799194336
    },
    {
      "auxiliary_loss_clip": 0.01230181,
      "auxiliary_loss_mlp": 0.01036587,
      "balance_loss_clip": 1.06877148,
      "balance_loss_mlp": 1.02492785,
      "epoch": 0.1358744664221728,
      "flos": 18150797698560.0,
      "grad_norm": 2.87518929652613,
      "language_loss": 0.78960055,
      "learning_rate": 3.883669688490654e-06,
      "loss": 0.81226814,
      "num_input_tokens_seen": 24081070,
      "step": 1130,
      "time_per_iteration": 2.5040786266326904
    },
    {
      "auxiliary_loss_clip": 0.01196868,
      "auxiliary_loss_mlp": 0.00763939,
      "balance_loss_clip": 1.06340885,
      "balance_loss_mlp": 1.00002766,
      "epoch": 0.13599470931281188,
      "flos": 18442966924800.0,
      "grad_norm": 1.760592807308845,
      "language_loss": 0.85254437,
      "learning_rate": 3.883407751890256e-06,
      "loss": 0.87215245,
      "num_input_tokens_seen": 24099675,
      "step": 1131,
      "time_per_iteration": 2.540595531463623
    },
    {
      "auxiliary_loss_clip": 0.01189569,
      "auxiliary_loss_mlp": 0.01053139,
      "balance_loss_clip": 1.0616672,
      "balance_loss_mlp": 1.04134369,
      "epoch": 0.13611495220345096,
      "flos": 26680766014080.0,
      "grad_norm": 1.7244084684688303,
      "language_loss": 0.85677087,
      "learning_rate": 3.8831455295783994e-06,
      "loss": 0.87919796,
      "num_input_tokens_seen": 24118925,
      "step": 1132,
      "time_per_iteration": 2.668065071105957
    },
    {
      "auxiliary_loss_clip": 0.01199704,
      "auxiliary_loss_mlp": 0.01041149,
      "balance_loss_clip": 1.06322074,
      "balance_loss_mlp": 1.03071237,
      "epoch": 0.13623519509409007,
      "flos": 21686238673920.0,
      "grad_norm": 1.6932830496390316,
      "language_loss": 0.74116409,
      "learning_rate": 3.882883021594864e-06,
      "loss": 0.76357269,
      "num_input_tokens_seen": 24137065,
      "step": 1133,
      "time_per_iteration": 2.5827267169952393
    },
    {
      "auxiliary_loss_clip": 0.01186156,
      "auxiliary_loss_mlp": 0.01039199,
      "balance_loss_clip": 1.06685448,
      "balance_loss_mlp": 1.0286665,
      "epoch": 0.13635543798472916,
      "flos": 14830389492480.0,
      "grad_norm": 2.133726437370247,
      "language_loss": 0.86785662,
      "learning_rate": 3.8826202279794705e-06,
      "loss": 0.89011019,
      "num_input_tokens_seen": 24154125,
      "step": 1134,
      "time_per_iteration": 2.5417089462280273
    },
    {
      "auxiliary_loss_clip": 0.01244427,
      "auxiliary_loss_mlp": 0.01038021,
      "balance_loss_clip": 1.07556391,
      "balance_loss_mlp": 1.02693439,
      "epoch": 0.13647568087536824,
      "flos": 22890323410560.0,
      "grad_norm": 1.9784777363543373,
      "language_loss": 0.69955075,
      "learning_rate": 3.882357148772085e-06,
      "loss": 0.72237521,
      "num_input_tokens_seen": 24171550,
      "step": 1135,
      "time_per_iteration": 2.4768245220184326
    },
    {
      "auxiliary_loss_clip": 0.01173109,
      "auxiliary_loss_mlp": 0.01036191,
      "balance_loss_clip": 1.06081605,
      "balance_loss_mlp": 1.02534878,
      "epoch": 0.13659592376600732,
      "flos": 19937927998080.0,
      "grad_norm": 2.1685532595731543,
      "language_loss": 0.84322059,
      "learning_rate": 3.882093784012617e-06,
      "loss": 0.86531359,
      "num_input_tokens_seen": 24190190,
      "step": 1136,
      "time_per_iteration": 3.279172420501709
    },
    {
      "auxiliary_loss_clip": 0.01205418,
      "auxiliary_loss_mlp": 0.01042745,
      "balance_loss_clip": 1.06602895,
      "balance_loss_mlp": 1.03200424,
      "epoch": 0.13671616665664643,
      "flos": 21428579439360.0,
      "grad_norm": 3.016829545914213,
      "language_loss": 0.84346777,
      "learning_rate": 3.881830133741019e-06,
      "loss": 0.86594939,
      "num_input_tokens_seen": 24209055,
      "step": 1137,
      "time_per_iteration": 2.554168224334717
    },
    {
      "auxiliary_loss_clip": 0.01190138,
      "auxiliary_loss_mlp": 0.01041915,
      "balance_loss_clip": 1.06542563,
      "balance_loss_mlp": 1.03112113,
      "epoch": 0.13683640954728551,
      "flos": 22778138257920.0,
      "grad_norm": 2.25583779416728,
      "language_loss": 0.76518047,
      "learning_rate": 3.881566197997285e-06,
      "loss": 0.78750098,
      "num_input_tokens_seen": 24225490,
      "step": 1138,
      "time_per_iteration": 3.2935361862182617
    },
    {
      "auxiliary_loss_clip": 0.0120532,
      "auxiliary_loss_mlp": 0.01034923,
      "balance_loss_clip": 1.07150459,
      "balance_loss_mlp": 1.02599454,
      "epoch": 0.1369566524379246,
      "flos": 21725884310400.0,
      "grad_norm": 1.5380594193272303,
      "language_loss": 0.74724793,
      "learning_rate": 3.881301976821456e-06,
      "loss": 0.76965034,
      "num_input_tokens_seen": 24245520,
      "step": 1139,
      "time_per_iteration": 2.569831132888794
    },
    {
      "auxiliary_loss_clip": 0.01220437,
      "auxiliary_loss_mlp": 0.01055003,
      "balance_loss_clip": 1.07040215,
      "balance_loss_mlp": 1.04451299,
      "epoch": 0.1370768953285637,
      "flos": 18624459369600.0,
      "grad_norm": 1.974003263037538,
      "language_loss": 0.90546453,
      "learning_rate": 3.881037470253612e-06,
      "loss": 0.9282189,
      "num_input_tokens_seen": 24265035,
      "step": 1140,
      "time_per_iteration": 3.2378902435302734
    },
    {
      "auxiliary_loss_clip": 0.01173965,
      "auxiliary_loss_mlp": 0.01041635,
      "balance_loss_clip": 1.0633359,
      "balance_loss_mlp": 1.03121591,
      "epoch": 0.1371971382192028,
      "flos": 14939521989120.0,
      "grad_norm": 2.551260192676724,
      "language_loss": 0.79472005,
      "learning_rate": 3.88077267833388e-06,
      "loss": 0.81687605,
      "num_input_tokens_seen": 24281550,
      "step": 1141,
      "time_per_iteration": 2.607044219970703
    },
    {
      "auxiliary_loss_clip": 0.01168256,
      "auxiliary_loss_mlp": 0.01045528,
      "balance_loss_clip": 1.06044364,
      "balance_loss_mlp": 1.03499007,
      "epoch": 0.13731738110984187,
      "flos": 19023785844480.0,
      "grad_norm": 2.4540920970050055,
      "language_loss": 0.83946609,
      "learning_rate": 3.880507601102427e-06,
      "loss": 0.86160398,
      "num_input_tokens_seen": 24299485,
      "step": 1142,
      "time_per_iteration": 3.3360414505004883
    },
    {
      "auxiliary_loss_clip": 0.01239483,
      "auxiliary_loss_mlp": 0.0104365,
      "balance_loss_clip": 1.07405138,
      "balance_loss_mlp": 1.0329864,
      "epoch": 0.13743762400048098,
      "flos": 18187462506240.0,
      "grad_norm": 1.9891838609193278,
      "language_loss": 0.82009637,
      "learning_rate": 3.880242238599467e-06,
      "loss": 0.84292769,
      "num_input_tokens_seen": 24316010,
      "step": 1143,
      "time_per_iteration": 2.464167594909668
    },
    {
      "auxiliary_loss_clip": 0.01237853,
      "auxiliary_loss_mlp": 0.01044802,
      "balance_loss_clip": 1.07210159,
      "balance_loss_mlp": 1.03452611,
      "epoch": 0.13755786689112007,
      "flos": 21031982398080.0,
      "grad_norm": 1.810489523737688,
      "language_loss": 0.83302999,
      "learning_rate": 3.879976590865254e-06,
      "loss": 0.85585654,
      "num_input_tokens_seen": 24335465,
      "step": 1144,
      "time_per_iteration": 2.5151267051696777
    },
    {
      "auxiliary_loss_clip": 0.01206537,
      "auxiliary_loss_mlp": 0.01043413,
      "balance_loss_clip": 1.06874466,
      "balance_loss_mlp": 1.03367376,
      "epoch": 0.13767810978175915,
      "flos": 21360636864000.0,
      "grad_norm": 2.4492752068655173,
      "language_loss": 0.87256116,
      "learning_rate": 3.879710657940087e-06,
      "loss": 0.89506072,
      "num_input_tokens_seen": 24354415,
      "step": 1145,
      "time_per_iteration": 2.5684032440185547
    },
    {
      "auxiliary_loss_clip": 0.01226121,
      "auxiliary_loss_mlp": 0.01045909,
      "balance_loss_clip": 1.07200193,
      "balance_loss_mlp": 1.03479266,
      "epoch": 0.13779835267239823,
      "flos": 30592084861440.0,
      "grad_norm": 1.8142344330606872,
      "language_loss": 0.70044303,
      "learning_rate": 3.879444439864308e-06,
      "loss": 0.72316331,
      "num_input_tokens_seen": 24373990,
      "step": 1146,
      "time_per_iteration": 2.6162970066070557
    },
    {
      "auxiliary_loss_clip": 0.01223864,
      "auxiliary_loss_mlp": 0.00764626,
      "balance_loss_clip": 1.06775951,
      "balance_loss_mlp": 1.00007868,
      "epoch": 0.13791859556303734,
      "flos": 22669867687680.0,
      "grad_norm": 1.8247518730885273,
      "language_loss": 0.85947877,
      "learning_rate": 3.879177936678301e-06,
      "loss": 0.87936372,
      "num_input_tokens_seen": 24392995,
      "step": 1147,
      "time_per_iteration": 2.558215379714966
    },
    {
      "auxiliary_loss_clip": 0.01211205,
      "auxiliary_loss_mlp": 0.01051362,
      "balance_loss_clip": 1.06756783,
      "balance_loss_mlp": 1.04016256,
      "epoch": 0.13803883845367643,
      "flos": 35224166016000.0,
      "grad_norm": 2.016657929155821,
      "language_loss": 0.7714982,
      "learning_rate": 3.878911148422496e-06,
      "loss": 0.79412389,
      "num_input_tokens_seen": 24414470,
      "step": 1148,
      "time_per_iteration": 2.679152727127075
    },
    {
      "auxiliary_loss_clip": 0.0122311,
      "auxiliary_loss_mlp": 0.01039296,
      "balance_loss_clip": 1.06740451,
      "balance_loss_mlp": 1.02801871,
      "epoch": 0.1381590813443155,
      "flos": 32014542332160.0,
      "grad_norm": 2.0965841062838457,
      "language_loss": 0.70386118,
      "learning_rate": 3.878644075137364e-06,
      "loss": 0.72648525,
      "num_input_tokens_seen": 24435120,
      "step": 1149,
      "time_per_iteration": 2.603072166442871
    },
    {
      "auxiliary_loss_clip": 0.01168865,
      "auxiliary_loss_mlp": 0.01037359,
      "balance_loss_clip": 1.05771565,
      "balance_loss_mlp": 1.02698135,
      "epoch": 0.13827932423495462,
      "flos": 17821855923840.0,
      "grad_norm": 1.9303235662594027,
      "language_loss": 0.79020268,
      "learning_rate": 3.878376716863418e-06,
      "loss": 0.81226492,
      "num_input_tokens_seen": 24451420,
      "step": 1150,
      "time_per_iteration": 2.568640947341919
    },
    {
      "auxiliary_loss_clip": 0.01208529,
      "auxiliary_loss_mlp": 0.01043499,
      "balance_loss_clip": 1.0686388,
      "balance_loss_mlp": 1.03267467,
      "epoch": 0.1383995671255937,
      "flos": 19427098728960.0,
      "grad_norm": 2.1386331431536547,
      "language_loss": 0.71652067,
      "learning_rate": 3.878109073641219e-06,
      "loss": 0.73904097,
      "num_input_tokens_seen": 24470450,
      "step": 1151,
      "time_per_iteration": 2.5820083618164062
    },
    {
      "auxiliary_loss_clip": 0.01171966,
      "auxiliary_loss_mlp": 0.01039776,
      "balance_loss_clip": 1.06081545,
      "balance_loss_mlp": 1.02931511,
      "epoch": 0.13851981001623279,
      "flos": 28296603331200.0,
      "grad_norm": 1.5526199993874867,
      "language_loss": 0.81295621,
      "learning_rate": 3.877841145511366e-06,
      "loss": 0.83507365,
      "num_input_tokens_seen": 24493190,
      "step": 1152,
      "time_per_iteration": 2.6953811645507812
    },
    {
      "auxiliary_loss_clip": 0.0122924,
      "auxiliary_loss_mlp": 0.01046464,
      "balance_loss_clip": 1.07070696,
      "balance_loss_mlp": 1.03554416,
      "epoch": 0.13864005290687187,
      "flos": 21213079793280.0,
      "grad_norm": 1.7498995785154863,
      "language_loss": 0.82878846,
      "learning_rate": 3.8775729325145035e-06,
      "loss": 0.85154557,
      "num_input_tokens_seen": 24512425,
      "step": 1153,
      "time_per_iteration": 2.505779981613159
    },
    {
      "auxiliary_loss_clip": 0.01074198,
      "auxiliary_loss_mlp": 0.0100266,
      "balance_loss_clip": 1.02419162,
      "balance_loss_mlp": 0.99961996,
      "epoch": 0.13876029579751098,
      "flos": 71653389413760.0,
      "grad_norm": 0.7906288474545474,
      "language_loss": 0.64768398,
      "learning_rate": 3.877304434691321e-06,
      "loss": 0.6684525,
      "num_input_tokens_seen": 24579275,
      "step": 1154,
      "time_per_iteration": 3.3098843097686768
    },
    {
      "auxiliary_loss_clip": 0.01190417,
      "auxiliary_loss_mlp": 0.01044166,
      "balance_loss_clip": 1.06470239,
      "balance_loss_mlp": 1.03418803,
      "epoch": 0.13888053868815006,
      "flos": 21941348042880.0,
      "grad_norm": 1.728192063580859,
      "language_loss": 0.79743266,
      "learning_rate": 3.877035652082548e-06,
      "loss": 0.81977856,
      "num_input_tokens_seen": 24598720,
      "step": 1155,
      "time_per_iteration": 2.687314510345459
    },
    {
      "auxiliary_loss_clip": 0.01196053,
      "auxiliary_loss_mlp": 0.01045991,
      "balance_loss_clip": 1.06727493,
      "balance_loss_mlp": 1.0335989,
      "epoch": 0.13900078157878915,
      "flos": 19608627087360.0,
      "grad_norm": 1.8732829073156738,
      "language_loss": 0.85319138,
      "learning_rate": 3.87676658472896e-06,
      "loss": 0.87561178,
      "num_input_tokens_seen": 24617530,
      "step": 1156,
      "time_per_iteration": 2.5447165966033936
    },
    {
      "auxiliary_loss_clip": 0.01219975,
      "auxiliary_loss_mlp": 0.01036379,
      "balance_loss_clip": 1.06328905,
      "balance_loss_mlp": 1.02615714,
      "epoch": 0.13912102446942826,
      "flos": 22638051216000.0,
      "grad_norm": 1.922781142927629,
      "language_loss": 0.85300469,
      "learning_rate": 3.876497232671372e-06,
      "loss": 0.87556827,
      "num_input_tokens_seen": 24637485,
      "step": 1157,
      "time_per_iteration": 2.5204851627349854
    },
    {
      "auxiliary_loss_clip": 0.01177775,
      "auxiliary_loss_mlp": 0.01036177,
      "balance_loss_clip": 1.05973625,
      "balance_loss_mlp": 1.02524543,
      "epoch": 0.13924126736006734,
      "flos": 29643324975360.0,
      "grad_norm": 2.1414461794580153,
      "language_loss": 0.83581114,
      "learning_rate": 3.876227595950647e-06,
      "loss": 0.85795063,
      "num_input_tokens_seen": 24656915,
      "step": 1158,
      "time_per_iteration": 2.6877052783966064
    },
    {
      "auxiliary_loss_clip": 0.01239745,
      "auxiliary_loss_mlp": 0.01036426,
      "balance_loss_clip": 1.07319832,
      "balance_loss_mlp": 1.0257926,
      "epoch": 0.13936151025070642,
      "flos": 27417653527680.0,
      "grad_norm": 1.7322000058818168,
      "language_loss": 0.78887689,
      "learning_rate": 3.875957674607686e-06,
      "loss": 0.81163859,
      "num_input_tokens_seen": 24679190,
      "step": 1159,
      "time_per_iteration": 2.5359959602355957
    },
    {
      "auxiliary_loss_clip": 0.01211797,
      "auxiliary_loss_mlp": 0.00764896,
      "balance_loss_clip": 1.06291389,
      "balance_loss_mlp": 1.00008035,
      "epoch": 0.1394817531413455,
      "flos": 16399326625920.0,
      "grad_norm": 1.9577591347752015,
      "language_loss": 0.87827551,
      "learning_rate": 3.8756874686834386e-06,
      "loss": 0.89804244,
      "num_input_tokens_seen": 24697405,
      "step": 1160,
      "time_per_iteration": 2.481790542602539
    },
    {
      "auxiliary_loss_clip": 0.01228954,
      "auxiliary_loss_mlp": 0.00764404,
      "balance_loss_clip": 1.06827223,
      "balance_loss_mlp": 1.00008917,
      "epoch": 0.13960199603198462,
      "flos": 30922319525760.0,
      "grad_norm": 1.7368458239657774,
      "language_loss": 0.80511677,
      "learning_rate": 3.875416978218893e-06,
      "loss": 0.82505035,
      "num_input_tokens_seen": 24720600,
      "step": 1161,
      "time_per_iteration": 2.620042562484741
    },
    {
      "auxiliary_loss_clip": 0.0119759,
      "auxiliary_loss_mlp": 0.01044747,
      "balance_loss_clip": 1.05987275,
      "balance_loss_mlp": 1.03351712,
      "epoch": 0.1397222389226237,
      "flos": 18113773754880.0,
      "grad_norm": 2.4549737725805496,
      "language_loss": 0.83208954,
      "learning_rate": 3.8751462032550835e-06,
      "loss": 0.85451293,
      "num_input_tokens_seen": 24737605,
      "step": 1162,
      "time_per_iteration": 3.3946242332458496
    },
    {
      "auxiliary_loss_clip": 0.01206955,
      "auxiliary_loss_mlp": 0.01042705,
      "balance_loss_clip": 1.07017696,
      "balance_loss_mlp": 1.0326314,
      "epoch": 0.13984248181326278,
      "flos": 16872772815360.0,
      "grad_norm": 3.4371805308696306,
      "language_loss": 0.83049047,
      "learning_rate": 3.874875143833085e-06,
      "loss": 0.85298711,
      "num_input_tokens_seen": 24755845,
      "step": 1163,
      "time_per_iteration": 2.5443012714385986
    },
    {
      "auxiliary_loss_clip": 0.01226824,
      "auxiliary_loss_mlp": 0.01041089,
      "balance_loss_clip": 1.07003617,
      "balance_loss_mlp": 1.02994871,
      "epoch": 0.1399627247039019,
      "flos": 54121401267840.0,
      "grad_norm": 1.7728770708954875,
      "language_loss": 0.68531573,
      "learning_rate": 3.874603799994019e-06,
      "loss": 0.70799482,
      "num_input_tokens_seen": 24779380,
      "step": 1164,
      "time_per_iteration": 2.8329520225524902
    },
    {
      "auxiliary_loss_clip": 0.01184171,
      "auxiliary_loss_mlp": 0.01040275,
      "balance_loss_clip": 1.06203735,
      "balance_loss_mlp": 1.03094649,
      "epoch": 0.14008296759454097,
      "flos": 11765521618560.0,
      "grad_norm": 2.163132056055306,
      "language_loss": 0.8711552,
      "learning_rate": 3.874332171779046e-06,
      "loss": 0.89339966,
      "num_input_tokens_seen": 24794260,
      "step": 1165,
      "time_per_iteration": 3.364712953567505
    },
    {
      "auxiliary_loss_clip": 0.01186474,
      "auxiliary_loss_mlp": 0.01043632,
      "balance_loss_clip": 1.06126451,
      "balance_loss_mlp": 1.03299832,
      "epoch": 0.14020321048518006,
      "flos": 22017514832640.0,
      "grad_norm": 1.8784991732215723,
      "language_loss": 0.75701189,
      "learning_rate": 3.874060259229373e-06,
      "loss": 0.77931297,
      "num_input_tokens_seen": 24815835,
      "step": 1166,
      "time_per_iteration": 3.3699495792388916
    },
    {
      "auxiliary_loss_clip": 0.01227501,
      "auxiliary_loss_mlp": 0.01046504,
      "balance_loss_clip": 1.07141829,
      "balance_loss_mlp": 1.03630543,
      "epoch": 0.14032345337581917,
      "flos": 23404313076480.0,
      "grad_norm": 2.122148924545455,
      "language_loss": 0.93754208,
      "learning_rate": 3.873788062386249e-06,
      "loss": 0.96028209,
      "num_input_tokens_seen": 24834095,
      "step": 1167,
      "time_per_iteration": 3.337472677230835
    },
    {
      "auxiliary_loss_clip": 0.0120268,
      "auxiliary_loss_mlp": 0.01044965,
      "balance_loss_clip": 1.07090044,
      "balance_loss_mlp": 1.03410494,
      "epoch": 0.14044369626645825,
      "flos": 29645767100160.0,
      "grad_norm": 3.0740757612724146,
      "language_loss": 0.81968713,
      "learning_rate": 3.873515581290965e-06,
      "loss": 0.84216356,
      "num_input_tokens_seen": 24858900,
      "step": 1168,
      "time_per_iteration": 2.6590356826782227
    },
    {
      "auxiliary_loss_clip": 0.0119537,
      "auxiliary_loss_mlp": 0.01041121,
      "balance_loss_clip": 1.07136595,
      "balance_loss_mlp": 1.03074372,
      "epoch": 0.14056393915709733,
      "flos": 18332972501760.0,
      "grad_norm": 2.223487051503957,
      "language_loss": 0.75664914,
      "learning_rate": 3.8732428159848575e-06,
      "loss": 0.77901399,
      "num_input_tokens_seen": 24877875,
      "step": 1169,
      "time_per_iteration": 2.583380699157715
    },
    {
      "auxiliary_loss_clip": 0.01223989,
      "auxiliary_loss_mlp": 0.01040391,
      "balance_loss_clip": 1.07267499,
      "balance_loss_mlp": 1.02951908,
      "epoch": 0.14068418204773642,
      "flos": 26687517770880.0,
      "grad_norm": 1.9388476739443976,
      "language_loss": 0.78290719,
      "learning_rate": 3.872969766509304e-06,
      "loss": 0.80555099,
      "num_input_tokens_seen": 24898430,
      "step": 1170,
      "time_per_iteration": 2.564643621444702
    },
    {
      "auxiliary_loss_clip": 0.01077592,
      "auxiliary_loss_mlp": 0.01010244,
      "balance_loss_clip": 1.02460599,
      "balance_loss_mlp": 1.00743091,
      "epoch": 0.14080442493837553,
      "flos": 65259314501760.0,
      "grad_norm": 0.7649416942154666,
      "language_loss": 0.55689424,
      "learning_rate": 3.872696432905726e-06,
      "loss": 0.57777262,
      "num_input_tokens_seen": 24959250,
      "step": 1171,
      "time_per_iteration": 3.155531167984009
    },
    {
      "auxiliary_loss_clip": 0.01228327,
      "auxiliary_loss_mlp": 0.01043899,
      "balance_loss_clip": 1.06894374,
      "balance_loss_mlp": 1.03276491,
      "epoch": 0.1409246678290146,
      "flos": 25776715582080.0,
      "grad_norm": 2.296491865698173,
      "language_loss": 0.7160061,
      "learning_rate": 3.872422815215589e-06,
      "loss": 0.73872828,
      "num_input_tokens_seen": 24978330,
      "step": 1172,
      "time_per_iteration": 2.5400092601776123
    },
    {
      "auxiliary_loss_clip": 0.01215221,
      "auxiliary_loss_mlp": 0.01044628,
      "balance_loss_clip": 1.06396461,
      "balance_loss_mlp": 1.03275454,
      "epoch": 0.1410449107196537,
      "flos": 21868521217920.0,
      "grad_norm": 1.99115911783194,
      "language_loss": 0.74266034,
      "learning_rate": 3.8721489134803994e-06,
      "loss": 0.76525885,
      "num_input_tokens_seen": 24997120,
      "step": 1173,
      "time_per_iteration": 2.524367332458496
    },
    {
      "auxiliary_loss_clip": 0.01218942,
      "auxiliary_loss_mlp": 0.01048966,
      "balance_loss_clip": 1.06684709,
      "balance_loss_mlp": 1.03840971,
      "epoch": 0.1411651536102928,
      "flos": 16684133564160.0,
      "grad_norm": 2.790371512309975,
      "language_loss": 0.7261548,
      "learning_rate": 3.871874727741707e-06,
      "loss": 0.74883389,
      "num_input_tokens_seen": 25014350,
      "step": 1174,
      "time_per_iteration": 2.490488290786743
    },
    {
      "auxiliary_loss_clip": 0.0122222,
      "auxiliary_loss_mlp": 0.01044657,
      "balance_loss_clip": 1.0707432,
      "balance_loss_mlp": 1.03511441,
      "epoch": 0.1412853965009319,
      "flos": 20992264934400.0,
      "grad_norm": 1.8734726218812983,
      "language_loss": 0.96573216,
      "learning_rate": 3.871600258041108e-06,
      "loss": 0.98840094,
      "num_input_tokens_seen": 25033875,
      "step": 1175,
      "time_per_iteration": 2.518188714981079
    },
    {
      "auxiliary_loss_clip": 0.01204978,
      "auxiliary_loss_mlp": 0.0103693,
      "balance_loss_clip": 1.06504822,
      "balance_loss_mlp": 1.026088,
      "epoch": 0.14140563939157097,
      "flos": 20335279224960.0,
      "grad_norm": 10.392965848892324,
      "language_loss": 0.85461283,
      "learning_rate": 3.871325504420238e-06,
      "loss": 0.87703192,
      "num_input_tokens_seen": 25052865,
      "step": 1176,
      "time_per_iteration": 2.5256526470184326
    },
    {
      "auxiliary_loss_clip": 0.01239771,
      "auxiliary_loss_mlp": 0.01038329,
      "balance_loss_clip": 1.07256269,
      "balance_loss_mlp": 1.02729678,
      "epoch": 0.14152588228221005,
      "flos": 21068826773760.0,
      "grad_norm": 2.3339663899710685,
      "language_loss": 0.8182168,
      "learning_rate": 3.871050466920776e-06,
      "loss": 0.84099782,
      "num_input_tokens_seen": 25072770,
      "step": 1177,
      "time_per_iteration": 2.5006301403045654
    },
    {
      "auxiliary_loss_clip": 0.01181739,
      "auxiliary_loss_mlp": 0.01043194,
      "balance_loss_clip": 1.06085324,
      "balance_loss_mlp": 1.03341854,
      "epoch": 0.14164612517284916,
      "flos": 18223157646720.0,
      "grad_norm": 2.1126768482357976,
      "language_loss": 0.80001688,
      "learning_rate": 3.870775145584447e-06,
      "loss": 0.82226622,
      "num_input_tokens_seen": 25090550,
      "step": 1178,
      "time_per_iteration": 2.5550267696380615
    },
    {
      "auxiliary_loss_clip": 0.01216423,
      "auxiliary_loss_mlp": 0.01051215,
      "balance_loss_clip": 1.06946087,
      "balance_loss_mlp": 1.038221,
      "epoch": 0.14176636806348825,
      "flos": 22744454279040.0,
      "grad_norm": 3.800001600348268,
      "language_loss": 0.65203154,
      "learning_rate": 3.8704995404530145e-06,
      "loss": 0.67470789,
      "num_input_tokens_seen": 25106175,
      "step": 1179,
      "time_per_iteration": 2.550929546356201
    },
    {
      "auxiliary_loss_clip": 0.01237087,
      "auxiliary_loss_mlp": 0.01049896,
      "balance_loss_clip": 1.07175148,
      "balance_loss_mlp": 1.04045486,
      "epoch": 0.14188661095412733,
      "flos": 22091095843200.0,
      "grad_norm": 1.9055131790722095,
      "language_loss": 0.85245156,
      "learning_rate": 3.87022365156829e-06,
      "loss": 0.87532133,
      "num_input_tokens_seen": 25126890,
      "step": 1180,
      "time_per_iteration": 2.509553909301758
    },
    {
      "auxiliary_loss_clip": 0.01141299,
      "auxiliary_loss_mlp": 0.01046825,
      "balance_loss_clip": 1.055902,
      "balance_loss_mlp": 1.03624487,
      "epoch": 0.14200685384476644,
      "flos": 24352390604160.0,
      "grad_norm": 1.9999384973730896,
      "language_loss": 0.81007594,
      "learning_rate": 3.869947478972123e-06,
      "loss": 0.83195716,
      "num_input_tokens_seen": 25147915,
      "step": 1181,
      "time_per_iteration": 2.699185609817505
    },
    {
      "auxiliary_loss_clip": 0.01212855,
      "auxiliary_loss_mlp": 0.01050564,
      "balance_loss_clip": 1.06427288,
      "balance_loss_mlp": 1.04030013,
      "epoch": 0.14212709673540552,
      "flos": 24022048199040.0,
      "grad_norm": 1.894264614422643,
      "language_loss": 0.82204187,
      "learning_rate": 3.869671022706412e-06,
      "loss": 0.84467614,
      "num_input_tokens_seen": 25166645,
      "step": 1182,
      "time_per_iteration": 2.5384719371795654
    },
    {
      "auxiliary_loss_clip": 0.01158059,
      "auxiliary_loss_mlp": 0.01040314,
      "balance_loss_clip": 1.05718136,
      "balance_loss_mlp": 1.03021061,
      "epoch": 0.1422473396260446,
      "flos": 26431797870720.0,
      "grad_norm": 1.8391047519957893,
      "language_loss": 0.64955759,
      "learning_rate": 3.869394282813092e-06,
      "loss": 0.67154127,
      "num_input_tokens_seen": 25185845,
      "step": 1183,
      "time_per_iteration": 2.763465404510498
    },
    {
      "auxiliary_loss_clip": 0.01197477,
      "auxiliary_loss_mlp": 0.01040825,
      "balance_loss_clip": 1.06077528,
      "balance_loss_mlp": 1.02858281,
      "epoch": 0.1423675825166837,
      "flos": 17055306754560.0,
      "grad_norm": 3.0974082004399066,
      "language_loss": 0.88988018,
      "learning_rate": 3.869117259334147e-06,
      "loss": 0.91226315,
      "num_input_tokens_seen": 25203770,
      "step": 1184,
      "time_per_iteration": 2.560513734817505
    },
    {
      "auxiliary_loss_clip": 0.01217706,
      "auxiliary_loss_mlp": 0.01048486,
      "balance_loss_clip": 1.06657434,
      "balance_loss_mlp": 1.03709614,
      "epoch": 0.1424878254073228,
      "flos": 17929480049280.0,
      "grad_norm": 1.9723263223924634,
      "language_loss": 0.82074761,
      "learning_rate": 3.868839952311599e-06,
      "loss": 0.84340954,
      "num_input_tokens_seen": 25221725,
      "step": 1185,
      "time_per_iteration": 2.5030486583709717
    },
    {
      "auxiliary_loss_clip": 0.011988,
      "auxiliary_loss_mlp": 0.01048922,
      "balance_loss_clip": 1.06278062,
      "balance_loss_mlp": 1.03792453,
      "epoch": 0.14260806829796188,
      "flos": 20303606407680.0,
      "grad_norm": 2.322310254770216,
      "language_loss": 0.80318534,
      "learning_rate": 3.868562361787516e-06,
      "loss": 0.82566249,
      "num_input_tokens_seen": 25240855,
      "step": 1186,
      "time_per_iteration": 2.5204532146453857
    },
    {
      "auxiliary_loss_clip": 0.0113869,
      "auxiliary_loss_mlp": 0.01038683,
      "balance_loss_clip": 1.0556227,
      "balance_loss_mlp": 1.02836561,
      "epoch": 0.14272831118860096,
      "flos": 23185724860800.0,
      "grad_norm": 1.8978966597112648,
      "language_loss": 0.68900943,
      "learning_rate": 3.868284487804009e-06,
      "loss": 0.71078324,
      "num_input_tokens_seen": 25260085,
      "step": 1187,
      "time_per_iteration": 3.555598258972168
    },
    {
      "auxiliary_loss_clip": 0.01210845,
      "auxiliary_loss_mlp": 0.01046049,
      "balance_loss_clip": 1.06583333,
      "balance_loss_mlp": 1.03534389,
      "epoch": 0.14284855407924008,
      "flos": 27232210586880.0,
      "grad_norm": 1.5425162109390098,
      "language_loss": 0.78048992,
      "learning_rate": 3.86800633040323e-06,
      "loss": 0.80305886,
      "num_input_tokens_seen": 25280675,
      "step": 1188,
      "time_per_iteration": 2.708998203277588
    },
    {
      "auxiliary_loss_clip": 0.01206242,
      "auxiliary_loss_mlp": 0.00763831,
      "balance_loss_clip": 1.06910992,
      "balance_loss_mlp": 1.00011086,
      "epoch": 0.14296879696987916,
      "flos": 28184202696960.0,
      "grad_norm": 2.072506443447748,
      "language_loss": 0.78509831,
      "learning_rate": 3.867727889627376e-06,
      "loss": 0.80479908,
      "num_input_tokens_seen": 25300290,
      "step": 1189,
      "time_per_iteration": 2.614508628845215
    },
    {
      "auxiliary_loss_clip": 0.01182641,
      "auxiliary_loss_mlp": 0.0104288,
      "balance_loss_clip": 1.06495619,
      "balance_loss_mlp": 1.03233027,
      "epoch": 0.14308903986051824,
      "flos": 19390290266880.0,
      "grad_norm": 2.6694291891790622,
      "language_loss": 0.78084159,
      "learning_rate": 3.867449165518687e-06,
      "loss": 0.80309689,
      "num_input_tokens_seen": 25316760,
      "step": 1190,
      "time_per_iteration": 3.3341403007507324
    },
    {
      "auxiliary_loss_clip": 0.01237701,
      "auxiliary_loss_mlp": 0.00764448,
      "balance_loss_clip": 1.0679363,
      "balance_loss_mlp": 1.00010705,
      "epoch": 0.14320928275115732,
      "flos": 17457506317440.0,
      "grad_norm": 1.8571396496057127,
      "language_loss": 0.71168935,
      "learning_rate": 3.867170158119444e-06,
      "loss": 0.73171085,
      "num_input_tokens_seen": 25335760,
      "step": 1191,
      "time_per_iteration": 3.2022206783294678
    },
    {
      "auxiliary_loss_clip": 0.01241313,
      "auxiliary_loss_mlp": 0.01045407,
      "balance_loss_clip": 1.07175863,
      "balance_loss_mlp": 1.03440404,
      "epoch": 0.14332952564179643,
      "flos": 21466070259840.0,
      "grad_norm": 1.8351944652002223,
      "language_loss": 0.75704867,
      "learning_rate": 3.866890867471972e-06,
      "loss": 0.77991587,
      "num_input_tokens_seen": 25354230,
      "step": 1192,
      "time_per_iteration": 2.501220703125
    },
    {
      "auxiliary_loss_clip": 0.01200005,
      "auxiliary_loss_mlp": 0.01055107,
      "balance_loss_clip": 1.05999422,
      "balance_loss_mlp": 1.04385984,
      "epoch": 0.14344976853243552,
      "flos": 16396992241920.0,
      "grad_norm": 2.85677477603376,
      "language_loss": 0.89934504,
      "learning_rate": 3.86661129361864e-06,
      "loss": 0.92189622,
      "num_input_tokens_seen": 25368720,
      "step": 1193,
      "time_per_iteration": 3.2081103324890137
    },
    {
      "auxiliary_loss_clip": 0.01201997,
      "auxiliary_loss_mlp": 0.01045438,
      "balance_loss_clip": 1.06405044,
      "balance_loss_mlp": 1.0346849,
      "epoch": 0.1435700114230746,
      "flos": 18916736336640.0,
      "grad_norm": 2.302127986164106,
      "language_loss": 0.85733926,
      "learning_rate": 3.866331436601859e-06,
      "loss": 0.87981355,
      "num_input_tokens_seen": 25386715,
      "step": 1194,
      "time_per_iteration": 2.53012752532959
    },
    {
      "auxiliary_loss_clip": 0.01239177,
      "auxiliary_loss_mlp": 0.01049949,
      "balance_loss_clip": 1.07220721,
      "balance_loss_mlp": 1.03891611,
      "epoch": 0.1436902543137137,
      "flos": 19755394058880.0,
      "grad_norm": 2.0616442689449,
      "language_loss": 0.74070966,
      "learning_rate": 3.866051296464083e-06,
      "loss": 0.76360089,
      "num_input_tokens_seen": 25405550,
      "step": 1195,
      "time_per_iteration": 2.470418930053711
    },
    {
      "auxiliary_loss_clip": 0.01236414,
      "auxiliary_loss_mlp": 0.00764295,
      "balance_loss_clip": 1.06728888,
      "balance_loss_mlp": 1.00009096,
      "epoch": 0.1438104972043528,
      "flos": 14684807669760.0,
      "grad_norm": 7.806885811676236,
      "language_loss": 0.85434055,
      "learning_rate": 3.86577087324781e-06,
      "loss": 0.87434769,
      "num_input_tokens_seen": 25422040,
      "step": 1196,
      "time_per_iteration": 2.462236166000366
    },
    {
      "auxiliary_loss_clip": 0.01220617,
      "auxiliary_loss_mlp": 0.01042788,
      "balance_loss_clip": 1.06998634,
      "balance_loss_mlp": 1.03215468,
      "epoch": 0.14393074009499188,
      "flos": 17092330698240.0,
      "grad_norm": 1.9418607803400916,
      "language_loss": 0.780321,
      "learning_rate": 3.865490166995578e-06,
      "loss": 0.80295503,
      "num_input_tokens_seen": 25440270,
      "step": 1197,
      "time_per_iteration": 2.4981062412261963
    },
    {
      "auxiliary_loss_clip": 0.0122065,
      "auxiliary_loss_mlp": 0.01036547,
      "balance_loss_clip": 1.06787896,
      "balance_loss_mlp": 1.02559149,
      "epoch": 0.144050982985631,
      "flos": 30476200608000.0,
      "grad_norm": 2.1678999740330958,
      "language_loss": 0.83994919,
      "learning_rate": 3.86520917774997e-06,
      "loss": 0.86252117,
      "num_input_tokens_seen": 25459705,
      "step": 1198,
      "time_per_iteration": 2.5767323970794678
    },
    {
      "auxiliary_loss_clip": 0.01218813,
      "auxiliary_loss_mlp": 0.01036536,
      "balance_loss_clip": 1.0678308,
      "balance_loss_mlp": 1.02724934,
      "epoch": 0.14417122587627007,
      "flos": 17858484817920.0,
      "grad_norm": 7.437738764998897,
      "language_loss": 0.74914801,
      "learning_rate": 3.864927905553614e-06,
      "loss": 0.77170151,
      "num_input_tokens_seen": 25477615,
      "step": 1199,
      "time_per_iteration": 2.4848790168762207
    },
    {
      "auxiliary_loss_clip": 0.01188189,
      "auxiliary_loss_mlp": 0.01040452,
      "balance_loss_clip": 1.06330466,
      "balance_loss_mlp": 1.03073072,
      "epoch": 0.14429146876690915,
      "flos": 21613914639360.0,
      "grad_norm": 1.750155872338273,
      "language_loss": 0.89036977,
      "learning_rate": 3.8646463504491765e-06,
      "loss": 0.91265619,
      "num_input_tokens_seen": 25497750,
      "step": 1200,
      "time_per_iteration": 2.583354949951172
    },
    {
      "auxiliary_loss_clip": 0.01225046,
      "auxiliary_loss_mlp": 0.01042827,
      "balance_loss_clip": 1.07102084,
      "balance_loss_mlp": 1.03187132,
      "epoch": 0.14441171165754824,
      "flos": 23258120722560.0,
      "grad_norm": 1.8849858337742023,
      "language_loss": 0.83540046,
      "learning_rate": 3.8643645124793705e-06,
      "loss": 0.8580792,
      "num_input_tokens_seen": 25516650,
      "step": 1201,
      "time_per_iteration": 2.5331785678863525
    },
    {
      "auxiliary_loss_clip": 0.01218955,
      "auxiliary_loss_mlp": 0.01039641,
      "balance_loss_clip": 1.06797111,
      "balance_loss_mlp": 1.02948475,
      "epoch": 0.14453195454818735,
      "flos": 42854213963520.0,
      "grad_norm": 1.7525258514862025,
      "language_loss": 0.74943173,
      "learning_rate": 3.8640823916869515e-06,
      "loss": 0.77201772,
      "num_input_tokens_seen": 25540960,
      "step": 1202,
      "time_per_iteration": 2.73274827003479
    },
    {
      "auxiliary_loss_clip": 0.01238516,
      "auxiliary_loss_mlp": 0.01040989,
      "balance_loss_clip": 1.07061863,
      "balance_loss_mlp": 1.0304153,
      "epoch": 0.14465219743882643,
      "flos": 27235873774080.0,
      "grad_norm": 1.5101298519358748,
      "language_loss": 0.78354555,
      "learning_rate": 3.863799988114714e-06,
      "loss": 0.80634058,
      "num_input_tokens_seen": 25562990,
      "step": 1203,
      "time_per_iteration": 2.532618522644043
    },
    {
      "auxiliary_loss_clip": 0.01240154,
      "auxiliary_loss_mlp": 0.01049258,
      "balance_loss_clip": 1.07104063,
      "balance_loss_mlp": 1.03869033,
      "epoch": 0.1447724403294655,
      "flos": 16690705752960.0,
      "grad_norm": 2.358913495222585,
      "language_loss": 0.70430768,
      "learning_rate": 3.863517301805502e-06,
      "loss": 0.72720182,
      "num_input_tokens_seen": 25581380,
      "step": 1204,
      "time_per_iteration": 2.4636733531951904
    },
    {
      "auxiliary_loss_clip": 0.01189813,
      "auxiliary_loss_mlp": 0.01052736,
      "balance_loss_clip": 1.06480682,
      "balance_loss_mlp": 1.04174495,
      "epoch": 0.14489268322010462,
      "flos": 20073741321600.0,
      "grad_norm": 2.4236540586033355,
      "language_loss": 0.97206497,
      "learning_rate": 3.863234332802196e-06,
      "loss": 0.9944905,
      "num_input_tokens_seen": 25593585,
      "step": 1205,
      "time_per_iteration": 2.5376169681549072
    },
    {
      "auxiliary_loss_clip": 0.01205685,
      "auxiliary_loss_mlp": 0.01053954,
      "balance_loss_clip": 1.06634295,
      "balance_loss_mlp": 1.04424465,
      "epoch": 0.1450129261107437,
      "flos": 27125627955840.0,
      "grad_norm": 2.15950022428928,
      "language_loss": 0.73593235,
      "learning_rate": 3.862951081147723e-06,
      "loss": 0.75852877,
      "num_input_tokens_seen": 25613750,
      "step": 1206,
      "time_per_iteration": 2.6040773391723633
    },
    {
      "auxiliary_loss_clip": 0.01225383,
      "auxiliary_loss_mlp": 0.01029555,
      "balance_loss_clip": 1.07073784,
      "balance_loss_mlp": 1.02035797,
      "epoch": 0.1451331690013828,
      "flos": 25702344472320.0,
      "grad_norm": 2.21975543337457,
      "language_loss": 0.78156471,
      "learning_rate": 3.862667546885053e-06,
      "loss": 0.8041141,
      "num_input_tokens_seen": 25632300,
      "step": 1207,
      "time_per_iteration": 2.5649683475494385
    },
    {
      "auxiliary_loss_clip": 0.01208664,
      "auxiliary_loss_mlp": 0.01040224,
      "balance_loss_clip": 1.06342316,
      "balance_loss_mlp": 1.03009117,
      "epoch": 0.14525341189202187,
      "flos": 25737393168000.0,
      "grad_norm": 1.8263123138488657,
      "language_loss": 0.73672479,
      "learning_rate": 3.8623837300571965e-06,
      "loss": 0.75921375,
      "num_input_tokens_seen": 25651285,
      "step": 1208,
      "time_per_iteration": 2.573349714279175
    },
    {
      "auxiliary_loss_clip": 0.01236526,
      "auxiliary_loss_mlp": 0.01046693,
      "balance_loss_clip": 1.06831574,
      "balance_loss_mlp": 1.03651273,
      "epoch": 0.14537365478266098,
      "flos": 23073898844160.0,
      "grad_norm": 2.5398537289672287,
      "language_loss": 0.8397541,
      "learning_rate": 3.8620996307072085e-06,
      "loss": 0.86258626,
      "num_input_tokens_seen": 25671990,
      "step": 1209,
      "time_per_iteration": 2.5037333965301514
    },
    {
      "auxiliary_loss_clip": 0.01192445,
      "auxiliary_loss_mlp": 0.01038187,
      "balance_loss_clip": 1.05924892,
      "balance_loss_mlp": 1.0280242,
      "epoch": 0.14549389767330007,
      "flos": 20595021448320.0,
      "grad_norm": 1.8130490946862647,
      "language_loss": 0.64464813,
      "learning_rate": 3.861815248878188e-06,
      "loss": 0.6669544,
      "num_input_tokens_seen": 25689475,
      "step": 1210,
      "time_per_iteration": 2.573620319366455
    },
    {
      "auxiliary_loss_clip": 0.01198499,
      "auxiliary_loss_mlp": 0.01031449,
      "balance_loss_clip": 1.06632876,
      "balance_loss_mlp": 1.02128613,
      "epoch": 0.14561414056393915,
      "flos": 15121804533120.0,
      "grad_norm": 2.287243218526405,
      "language_loss": 0.79312414,
      "learning_rate": 3.861530584613274e-06,
      "loss": 0.81542361,
      "num_input_tokens_seen": 25707475,
      "step": 1211,
      "time_per_iteration": 2.524085760116577
    },
    {
      "auxiliary_loss_clip": 0.0121931,
      "auxiliary_loss_mlp": 0.0076427,
      "balance_loss_clip": 1.06680238,
      "balance_loss_mlp": 1.00008512,
      "epoch": 0.14573438345457826,
      "flos": 19427493778560.0,
      "grad_norm": 2.13414999522986,
      "language_loss": 0.82104164,
      "learning_rate": 3.86124563795565e-06,
      "loss": 0.84087741,
      "num_input_tokens_seen": 25726290,
      "step": 1212,
      "time_per_iteration": 2.518179178237915
    },
    {
      "auxiliary_loss_clip": 0.01235229,
      "auxiliary_loss_mlp": 0.01045457,
      "balance_loss_clip": 1.07143354,
      "balance_loss_mlp": 1.03541398,
      "epoch": 0.14585462634521734,
      "flos": 24828422572800.0,
      "grad_norm": 1.780015111174624,
      "language_loss": 0.70183754,
      "learning_rate": 3.860960408948543e-06,
      "loss": 0.72464436,
      "num_input_tokens_seen": 25748040,
      "step": 1213,
      "time_per_iteration": 3.2525079250335693
    },
    {
      "auxiliary_loss_clip": 0.01209535,
      "auxiliary_loss_mlp": 0.01040597,
      "balance_loss_clip": 1.06439734,
      "balance_loss_mlp": 1.03086925,
      "epoch": 0.14597486923585642,
      "flos": 15448627405440.0,
      "grad_norm": 2.505900079757993,
      "language_loss": 0.8986693,
      "learning_rate": 3.860674897635222e-06,
      "loss": 0.92117065,
      "num_input_tokens_seen": 25764525,
      "step": 1214,
      "time_per_iteration": 2.4681742191314697
    },
    {
      "auxiliary_loss_clip": 0.01218366,
      "auxiliary_loss_mlp": 0.01048691,
      "balance_loss_clip": 1.0658741,
      "balance_loss_mlp": 1.03795648,
      "epoch": 0.1460951121264955,
      "flos": 16655154266880.0,
      "grad_norm": 2.491793929092135,
      "language_loss": 0.8331666,
      "learning_rate": 3.860389104058998e-06,
      "loss": 0.85583723,
      "num_input_tokens_seen": 25782755,
      "step": 1215,
      "time_per_iteration": 2.4893887042999268
    },
    {
      "auxiliary_loss_clip": 0.01203695,
      "auxiliary_loss_mlp": 0.01041893,
      "balance_loss_clip": 1.06551313,
      "balance_loss_mlp": 1.03127742,
      "epoch": 0.14621535501713462,
      "flos": 24863291700480.0,
      "grad_norm": 2.4635722457433875,
      "language_loss": 0.72621191,
      "learning_rate": 3.860103028263227e-06,
      "loss": 0.74866772,
      "num_input_tokens_seen": 25805860,
      "step": 1216,
      "time_per_iteration": 2.5885047912597656
    },
    {
      "auxiliary_loss_clip": 0.01165983,
      "auxiliary_loss_mlp": 0.01042436,
      "balance_loss_clip": 1.05447531,
      "balance_loss_mlp": 1.03198743,
      "epoch": 0.1463355979077737,
      "flos": 25228000442880.0,
      "grad_norm": 2.0035692223552153,
      "language_loss": 0.70179909,
      "learning_rate": 3.859816670291304e-06,
      "loss": 0.72388327,
      "num_input_tokens_seen": 25824955,
      "step": 1217,
      "time_per_iteration": 3.4199001789093018
    },
    {
      "auxiliary_loss_clip": 0.01151335,
      "auxiliary_loss_mlp": 0.01045223,
      "balance_loss_clip": 1.05895269,
      "balance_loss_mlp": 1.03507257,
      "epoch": 0.14645584079841278,
      "flos": 22054143726720.0,
      "grad_norm": 2.4363594744974093,
      "language_loss": 0.9030624,
      "learning_rate": 3.859530030186672e-06,
      "loss": 0.92502797,
      "num_input_tokens_seen": 25841965,
      "step": 1218,
      "time_per_iteration": 3.3213143348693848
    },
    {
      "auxiliary_loss_clip": 0.0120925,
      "auxiliary_loss_mlp": 0.01039827,
      "balance_loss_clip": 1.06653953,
      "balance_loss_mlp": 1.02921176,
      "epoch": 0.1465760836890519,
      "flos": 23623870959360.0,
      "grad_norm": 126.12435513719466,
      "language_loss": 0.82851446,
      "learning_rate": 3.859243107992813e-06,
      "loss": 0.8510052,
      "num_input_tokens_seen": 25860770,
      "step": 1219,
      "time_per_iteration": 3.375338554382324
    },
    {
      "auxiliary_loss_clip": 0.01192572,
      "auxiliary_loss_mlp": 0.01040005,
      "balance_loss_clip": 1.0584774,
      "balance_loss_mlp": 1.02946711,
      "epoch": 0.14669632657969098,
      "flos": 37407893356800.0,
      "grad_norm": 2.3801636510324506,
      "language_loss": 0.7819373,
      "learning_rate": 3.858955903753252e-06,
      "loss": 0.80426306,
      "num_input_tokens_seen": 25879410,
      "step": 1220,
      "time_per_iteration": 2.7418296337127686
    },
    {
      "auxiliary_loss_clip": 0.01219585,
      "auxiliary_loss_mlp": 0.01037564,
      "balance_loss_clip": 1.06580448,
      "balance_loss_mlp": 1.02757978,
      "epoch": 0.14681656947033006,
      "flos": 28365910623360.0,
      "grad_norm": 1.6368415040399995,
      "language_loss": 0.8353225,
      "learning_rate": 3.858668417511559e-06,
      "loss": 0.857894,
      "num_input_tokens_seen": 25902160,
      "step": 1221,
      "time_per_iteration": 2.600917100906372
    },
    {
      "auxiliary_loss_clip": 0.01207971,
      "auxiliary_loss_mlp": 0.01037379,
      "balance_loss_clip": 1.06781805,
      "balance_loss_mlp": 1.02713943,
      "epoch": 0.14693681236096917,
      "flos": 18479488078080.0,
      "grad_norm": 2.008748524464274,
      "language_loss": 0.76309729,
      "learning_rate": 3.8583806493113445e-06,
      "loss": 0.78555083,
      "num_input_tokens_seen": 25920505,
      "step": 1222,
      "time_per_iteration": 2.532684087753296
    },
    {
      "auxiliary_loss_clip": 0.01217096,
      "auxiliary_loss_mlp": 0.01045283,
      "balance_loss_clip": 1.06718576,
      "balance_loss_mlp": 1.03492975,
      "epoch": 0.14705705525160825,
      "flos": 20777806782720.0,
      "grad_norm": 21.65130867512965,
      "language_loss": 0.82175773,
      "learning_rate": 3.858092599196263e-06,
      "loss": 0.84438151,
      "num_input_tokens_seen": 25938460,
      "step": 1223,
      "time_per_iteration": 2.520313024520874
    },
    {
      "auxiliary_loss_clip": 0.01214872,
      "auxiliary_loss_mlp": 0.01040273,
      "balance_loss_clip": 1.06513786,
      "balance_loss_mlp": 1.0298655,
      "epoch": 0.14717729814224734,
      "flos": 29932944336000.0,
      "grad_norm": 2.9650409227733205,
      "language_loss": 0.82650965,
      "learning_rate": 3.857804267210012e-06,
      "loss": 0.84906107,
      "num_input_tokens_seen": 25957760,
      "step": 1224,
      "time_per_iteration": 2.573158025741577
    },
    {
      "auxiliary_loss_clip": 0.01171219,
      "auxiliary_loss_mlp": 0.01037582,
      "balance_loss_clip": 1.05604887,
      "balance_loss_mlp": 1.02771163,
      "epoch": 0.14729754103288642,
      "flos": 20047491457920.0,
      "grad_norm": 1.9573344880507828,
      "language_loss": 0.88097692,
      "learning_rate": 3.857515653396331e-06,
      "loss": 0.90306491,
      "num_input_tokens_seen": 25974970,
      "step": 1225,
      "time_per_iteration": 2.564187526702881
    },
    {
      "auxiliary_loss_clip": 0.01174453,
      "auxiliary_loss_mlp": 0.01041378,
      "balance_loss_clip": 1.06007719,
      "balance_loss_mlp": 1.03162026,
      "epoch": 0.14741778392352553,
      "flos": 19281516906240.0,
      "grad_norm": 2.2131698639411095,
      "language_loss": 0.8671701,
      "learning_rate": 3.857226757799002e-06,
      "loss": 0.88932836,
      "num_input_tokens_seen": 25992525,
      "step": 1226,
      "time_per_iteration": 2.6035447120666504
    },
    {
      "auxiliary_loss_clip": 0.01202291,
      "auxiliary_loss_mlp": 0.01038263,
      "balance_loss_clip": 1.06191885,
      "balance_loss_mlp": 1.02787364,
      "epoch": 0.1475380268141646,
      "flos": 25411108999680.0,
      "grad_norm": 2.4199833198526,
      "language_loss": 0.74384409,
      "learning_rate": 3.85693758046185e-06,
      "loss": 0.76624966,
      "num_input_tokens_seen": 26010815,
      "step": 1227,
      "time_per_iteration": 2.591486692428589
    },
    {
      "auxiliary_loss_clip": 0.01235524,
      "auxiliary_loss_mlp": 0.0103081,
      "balance_loss_clip": 1.07165742,
      "balance_loss_mlp": 1.02157092,
      "epoch": 0.1476582697048037,
      "flos": 20847652778880.0,
      "grad_norm": 1.794290337170015,
      "language_loss": 0.8312875,
      "learning_rate": 3.8566481214287435e-06,
      "loss": 0.85395086,
      "num_input_tokens_seen": 26028935,
      "step": 1228,
      "time_per_iteration": 2.4958767890930176
    },
    {
      "auxiliary_loss_clip": 0.01177755,
      "auxiliary_loss_mlp": 0.01048504,
      "balance_loss_clip": 1.05671024,
      "balance_loss_mlp": 1.03927124,
      "epoch": 0.1477785125954428,
      "flos": 14028109269120.0,
      "grad_norm": 2.2192077802258328,
      "language_loss": 0.9022603,
      "learning_rate": 3.8563583807435935e-06,
      "loss": 0.92452288,
      "num_input_tokens_seen": 26045080,
      "step": 1229,
      "time_per_iteration": 2.5569844245910645
    },
    {
      "auxiliary_loss_clip": 0.01218102,
      "auxiliary_loss_mlp": 0.00763934,
      "balance_loss_clip": 1.065799,
      "balance_loss_mlp": 1.00011396,
      "epoch": 0.1478987554860819,
      "flos": 20516699842560.0,
      "grad_norm": 1.920996783308175,
      "language_loss": 0.77890736,
      "learning_rate": 3.856068358450353e-06,
      "loss": 0.79872775,
      "num_input_tokens_seen": 26065030,
      "step": 1230,
      "time_per_iteration": 2.5401666164398193
    },
    {
      "auxiliary_loss_clip": 0.01197051,
      "auxiliary_loss_mlp": 0.01046809,
      "balance_loss_clip": 1.06661689,
      "balance_loss_mlp": 1.03741515,
      "epoch": 0.14801899837672097,
      "flos": 17857012360320.0,
      "grad_norm": 1.8319963498551617,
      "language_loss": 0.85799837,
      "learning_rate": 3.8557780545930186e-06,
      "loss": 0.88043696,
      "num_input_tokens_seen": 26083445,
      "step": 1231,
      "time_per_iteration": 2.5340802669525146
    },
    {
      "auxiliary_loss_clip": 0.01201185,
      "auxiliary_loss_mlp": 0.01038438,
      "balance_loss_clip": 1.06662345,
      "balance_loss_mlp": 1.02823997,
      "epoch": 0.14813924126736006,
      "flos": 20881408584960.0,
      "grad_norm": 4.536479468467541,
      "language_loss": 0.79221082,
      "learning_rate": 3.855487469215628e-06,
      "loss": 0.81460696,
      "num_input_tokens_seen": 26102375,
      "step": 1232,
      "time_per_iteration": 2.5651161670684814
    },
    {
      "auxiliary_loss_clip": 0.01184046,
      "auxiliary_loss_mlp": 0.0103755,
      "balance_loss_clip": 1.06203723,
      "balance_loss_mlp": 1.02804315,
      "epoch": 0.14825948415799917,
      "flos": 37414070496000.0,
      "grad_norm": 2.3141789531318753,
      "language_loss": 0.72074735,
      "learning_rate": 3.855196602362264e-06,
      "loss": 0.74296331,
      "num_input_tokens_seen": 26125295,
      "step": 1233,
      "time_per_iteration": 2.7047815322875977
    },
    {
      "auxiliary_loss_clip": 0.0121672,
      "auxiliary_loss_mlp": 0.01034249,
      "balance_loss_clip": 1.06348538,
      "balance_loss_mlp": 1.02492046,
      "epoch": 0.14837972704863825,
      "flos": 22014641744640.0,
      "grad_norm": 2.759836005956032,
      "language_loss": 0.94438124,
      "learning_rate": 3.854905454077051e-06,
      "loss": 0.96689093,
      "num_input_tokens_seen": 26142905,
      "step": 1234,
      "time_per_iteration": 2.512117624282837
    },
    {
      "auxiliary_loss_clip": 0.01134758,
      "auxiliary_loss_mlp": 0.0103358,
      "balance_loss_clip": 1.05206752,
      "balance_loss_mlp": 1.02440667,
      "epoch": 0.14849996993927733,
      "flos": 20996323171200.0,
      "grad_norm": 1.799342135685345,
      "language_loss": 0.88395083,
      "learning_rate": 3.854614024404155e-06,
      "loss": 0.90563422,
      "num_input_tokens_seen": 26161215,
      "step": 1235,
      "time_per_iteration": 2.6659913063049316
    },
    {
      "auxiliary_loss_clip": 0.0118801,
      "auxiliary_loss_mlp": 0.01045039,
      "balance_loss_clip": 1.05992126,
      "balance_loss_mlp": 1.03572261,
      "epoch": 0.14862021282991644,
      "flos": 20047994248320.0,
      "grad_norm": 2.2246791207531884,
      "language_loss": 0.8939684,
      "learning_rate": 3.8543223133877865e-06,
      "loss": 0.91629887,
      "num_input_tokens_seen": 26179810,
      "step": 1236,
      "time_per_iteration": 2.5244293212890625
    },
    {
      "auxiliary_loss_clip": 0.01179603,
      "auxiliary_loss_mlp": 0.01046779,
      "balance_loss_clip": 1.0569551,
      "balance_loss_mlp": 1.03568697,
      "epoch": 0.14874045572055553,
      "flos": 22712027276160.0,
      "grad_norm": 1.9338246477101735,
      "language_loss": 0.88281071,
      "learning_rate": 3.854030321072198e-06,
      "loss": 0.90507454,
      "num_input_tokens_seen": 26199715,
      "step": 1237,
      "time_per_iteration": 2.5670957565307617
    },
    {
      "auxiliary_loss_clip": 0.01191532,
      "auxiliary_loss_mlp": 0.01044278,
      "balance_loss_clip": 1.06097507,
      "balance_loss_mlp": 1.03455615,
      "epoch": 0.1488606986111946,
      "flos": 25411288567680.0,
      "grad_norm": 1.9485664210385103,
      "language_loss": 0.73542875,
      "learning_rate": 3.853738047501682e-06,
      "loss": 0.75778687,
      "num_input_tokens_seen": 26220275,
      "step": 1238,
      "time_per_iteration": 2.622931480407715
    },
    {
      "auxiliary_loss_clip": 0.01220864,
      "auxiliary_loss_mlp": 0.0104049,
      "balance_loss_clip": 1.06830692,
      "balance_loss_mlp": 1.03018999,
      "epoch": 0.1489809415018337,
      "flos": 17018749687680.0,
      "grad_norm": 2.311163616947138,
      "language_loss": 0.77790856,
      "learning_rate": 3.85344549272058e-06,
      "loss": 0.80052209,
      "num_input_tokens_seen": 26238255,
      "step": 1239,
      "time_per_iteration": 3.325458288192749
    },
    {
      "auxiliary_loss_clip": 0.01213437,
      "auxiliary_loss_mlp": 0.01037293,
      "balance_loss_clip": 1.06554866,
      "balance_loss_mlp": 1.0273931,
      "epoch": 0.1491011843924728,
      "flos": 33659394860160.0,
      "grad_norm": 1.8156580564623166,
      "language_loss": 0.82844424,
      "learning_rate": 3.853152656773269e-06,
      "loss": 0.85095155,
      "num_input_tokens_seen": 26259690,
      "step": 1240,
      "time_per_iteration": 2.6338186264038086
    },
    {
      "auxiliary_loss_clip": 0.01199765,
      "auxiliary_loss_mlp": 0.01040952,
      "balance_loss_clip": 1.06367874,
      "balance_loss_mlp": 1.03128982,
      "epoch": 0.14922142728311188,
      "flos": 21179000764800.0,
      "grad_norm": 1.6569908625758358,
      "language_loss": 0.84640503,
      "learning_rate": 3.852859539704174e-06,
      "loss": 0.8688122,
      "num_input_tokens_seen": 26278990,
      "step": 1241,
      "time_per_iteration": 2.5596587657928467
    },
    {
      "auxiliary_loss_clip": 0.01168839,
      "auxiliary_loss_mlp": 0.01038059,
      "balance_loss_clip": 1.05789781,
      "balance_loss_mlp": 1.02797389,
      "epoch": 0.14934167017375097,
      "flos": 29860548474240.0,
      "grad_norm": 1.835476112114779,
      "language_loss": 0.7623288,
      "learning_rate": 3.85256614155776e-06,
      "loss": 0.78439778,
      "num_input_tokens_seen": 26299120,
      "step": 1242,
      "time_per_iteration": 3.413752317428589
    },
    {
      "auxiliary_loss_clip": 0.01213212,
      "auxiliary_loss_mlp": 0.01047164,
      "balance_loss_clip": 1.06097543,
      "balance_loss_mlp": 1.03741884,
      "epoch": 0.14946191306439008,
      "flos": 17019216564480.0,
      "grad_norm": 2.0492215053719023,
      "language_loss": 0.74554384,
      "learning_rate": 3.852272462378535e-06,
      "loss": 0.76814759,
      "num_input_tokens_seen": 26316995,
      "step": 1243,
      "time_per_iteration": 2.5034518241882324
    },
    {
      "auxiliary_loss_clip": 0.01199331,
      "auxiliary_loss_mlp": 0.01039098,
      "balance_loss_clip": 1.06175315,
      "balance_loss_mlp": 1.02981186,
      "epoch": 0.14958215595502916,
      "flos": 15669047214720.0,
      "grad_norm": 2.507181371958501,
      "language_loss": 0.77797937,
      "learning_rate": 3.85197850221105e-06,
      "loss": 0.80036366,
      "num_input_tokens_seen": 26333295,
      "step": 1244,
      "time_per_iteration": 3.9502391815185547
    },
    {
      "auxiliary_loss_clip": 0.01212757,
      "auxiliary_loss_mlp": 0.01035283,
      "balance_loss_clip": 1.06706762,
      "balance_loss_mlp": 1.02581203,
      "epoch": 0.14970239884566824,
      "flos": 33108560818560.0,
      "grad_norm": 2.3726623667044358,
      "language_loss": 0.75557154,
      "learning_rate": 3.851684261099899e-06,
      "loss": 0.77805191,
      "num_input_tokens_seen": 26355035,
      "step": 1245,
      "time_per_iteration": 2.6286158561706543
    },
    {
      "auxiliary_loss_clip": 0.01196867,
      "auxiliary_loss_mlp": 0.01034535,
      "balance_loss_clip": 1.06202912,
      "balance_loss_mlp": 1.02520132,
      "epoch": 0.14982264173630733,
      "flos": 17821245392640.0,
      "grad_norm": 2.06953269660528,
      "language_loss": 0.86732119,
      "learning_rate": 3.851389739089718e-06,
      "loss": 0.88963521,
      "num_input_tokens_seen": 26371655,
      "step": 1246,
      "time_per_iteration": 2.5257251262664795
    },
    {
      "auxiliary_loss_clip": 0.01218965,
      "auxiliary_loss_mlp": 0.01038042,
      "balance_loss_clip": 1.06917202,
      "balance_loss_mlp": 1.02779019,
      "epoch": 0.14994288462694644,
      "flos": 32409559175040.0,
      "grad_norm": 3.2010424294520945,
      "language_loss": 0.80093694,
      "learning_rate": 3.851094936225186e-06,
      "loss": 0.82350701,
      "num_input_tokens_seen": 26392540,
      "step": 1247,
      "time_per_iteration": 2.6059978008270264
    },
    {
      "auxiliary_loss_clip": 0.01197522,
      "auxiliary_loss_mlp": 0.01034969,
      "balance_loss_clip": 1.06480241,
      "balance_loss_mlp": 1.02514052,
      "epoch": 0.15006312751758552,
      "flos": 31794661226880.0,
      "grad_norm": 1.5221054215568754,
      "language_loss": 0.76697505,
      "learning_rate": 3.850799852551024e-06,
      "loss": 0.78929996,
      "num_input_tokens_seen": 26414960,
      "step": 1248,
      "time_per_iteration": 2.627321720123291
    },
    {
      "auxiliary_loss_clip": 0.01206191,
      "auxiliary_loss_mlp": 0.0104399,
      "balance_loss_clip": 1.06086278,
      "balance_loss_mlp": 1.03441155,
      "epoch": 0.1501833704082246,
      "flos": 16618022582400.0,
      "grad_norm": 2.2571696673926906,
      "language_loss": 0.85963368,
      "learning_rate": 3.850504488111995e-06,
      "loss": 0.88213551,
      "num_input_tokens_seen": 26431635,
      "step": 1249,
      "time_per_iteration": 2.5312087535858154
    },
    {
      "auxiliary_loss_clip": 0.01191656,
      "auxiliary_loss_mlp": 0.0103711,
      "balance_loss_clip": 1.05951965,
      "balance_loss_mlp": 1.02784169,
      "epoch": 0.15030361329886371,
      "flos": 23471178243840.0,
      "grad_norm": 1.8725446992473576,
      "language_loss": 0.82770312,
      "learning_rate": 3.850208842952907e-06,
      "loss": 0.84999079,
      "num_input_tokens_seen": 26450440,
      "step": 1250,
      "time_per_iteration": 2.6059553623199463
    },
    {
      "auxiliary_loss_clip": 0.0117614,
      "auxiliary_loss_mlp": 0.01034838,
      "balance_loss_clip": 1.05648148,
      "balance_loss_mlp": 1.02540874,
      "epoch": 0.1504238561895028,
      "flos": 25629409906560.0,
      "grad_norm": 1.8848417262908645,
      "language_loss": 0.79202056,
      "learning_rate": 3.849912917118608e-06,
      "loss": 0.81413037,
      "num_input_tokens_seen": 26471480,
      "step": 1251,
      "time_per_iteration": 2.6452410221099854
    },
    {
      "auxiliary_loss_clip": 0.01126418,
      "auxiliary_loss_mlp": 0.01037186,
      "balance_loss_clip": 1.03905678,
      "balance_loss_mlp": 1.03421807,
      "epoch": 0.15054409908014188,
      "flos": 52095146129280.0,
      "grad_norm": 0.8898160478092594,
      "language_loss": 0.59299958,
      "learning_rate": 3.849616710653992e-06,
      "loss": 0.61463559,
      "num_input_tokens_seen": 26532950,
      "step": 1252,
      "time_per_iteration": 3.123248815536499
    },
    {
      "auxiliary_loss_clip": 0.01215588,
      "auxiliary_loss_mlp": 0.01037339,
      "balance_loss_clip": 1.0654192,
      "balance_loss_mlp": 1.02740872,
      "epoch": 0.150664341970781,
      "flos": 18880251096960.0,
      "grad_norm": 1.6391089731606816,
      "language_loss": 0.74911976,
      "learning_rate": 3.84932022360399e-06,
      "loss": 0.771649,
      "num_input_tokens_seen": 26551615,
      "step": 1253,
      "time_per_iteration": 2.525238275527954
    },
    {
      "auxiliary_loss_clip": 0.0119814,
      "auxiliary_loss_mlp": 0.01043897,
      "balance_loss_clip": 1.06833661,
      "balance_loss_mlp": 1.0336628,
      "epoch": 0.15078458486142007,
      "flos": 22163240309760.0,
      "grad_norm": 2.8070992715039202,
      "language_loss": 0.84425253,
      "learning_rate": 3.849023456013581e-06,
      "loss": 0.86667287,
      "num_input_tokens_seen": 26569175,
      "step": 1254,
      "time_per_iteration": 2.5650267601013184
    },
    {
      "auxiliary_loss_clip": 0.01219972,
      "auxiliary_loss_mlp": 0.01046886,
      "balance_loss_clip": 1.06662714,
      "balance_loss_mlp": 1.03693163,
      "epoch": 0.15090482775205916,
      "flos": 26651894457600.0,
      "grad_norm": 2.977613070945915,
      "language_loss": 0.62039173,
      "learning_rate": 3.848726407927784e-06,
      "loss": 0.64306027,
      "num_input_tokens_seen": 26589560,
      "step": 1255,
      "time_per_iteration": 2.590996742248535
    },
    {
      "auxiliary_loss_clip": 0.01204443,
      "auxiliary_loss_mlp": 0.01041312,
      "balance_loss_clip": 1.06819761,
      "balance_loss_mlp": 1.03060734,
      "epoch": 0.15102507064269824,
      "flos": 21798998444160.0,
      "grad_norm": 2.324249326525796,
      "language_loss": 0.86564147,
      "learning_rate": 3.84842907939166e-06,
      "loss": 0.88809907,
      "num_input_tokens_seen": 26608785,
      "step": 1256,
      "time_per_iteration": 2.5579047203063965
    },
    {
      "auxiliary_loss_clip": 0.01179238,
      "auxiliary_loss_mlp": 0.01043049,
      "balance_loss_clip": 1.06031156,
      "balance_loss_mlp": 1.03298795,
      "epoch": 0.15114531353333735,
      "flos": 22820908377600.0,
      "grad_norm": 2.6541669305571123,
      "language_loss": 0.71345329,
      "learning_rate": 3.8481314704503146e-06,
      "loss": 0.73567617,
      "num_input_tokens_seen": 26628615,
      "step": 1257,
      "time_per_iteration": 2.6265079975128174
    },
    {
      "auxiliary_loss_clip": 0.0121475,
      "auxiliary_loss_mlp": 0.01054902,
      "balance_loss_clip": 1.06889272,
      "balance_loss_mlp": 1.04505491,
      "epoch": 0.15126555642397643,
      "flos": 19682674974720.0,
      "grad_norm": 2.2504887181899695,
      "language_loss": 0.8792153,
      "learning_rate": 3.847833581148895e-06,
      "loss": 0.90191185,
      "num_input_tokens_seen": 26647525,
      "step": 1258,
      "time_per_iteration": 2.530487060546875
    },
    {
      "auxiliary_loss_clip": 0.01231308,
      "auxiliary_loss_mlp": 0.01033265,
      "balance_loss_clip": 1.066185,
      "balance_loss_mlp": 1.02330554,
      "epoch": 0.15138579931461552,
      "flos": 28726022424960.0,
      "grad_norm": 2.207361799962119,
      "language_loss": 0.80591643,
      "learning_rate": 3.84753541153259e-06,
      "loss": 0.8285622,
      "num_input_tokens_seen": 26667095,
      "step": 1259,
      "time_per_iteration": 2.589521646499634
    },
    {
      "auxiliary_loss_clip": 0.01217481,
      "auxiliary_loss_mlp": 0.01036359,
      "balance_loss_clip": 1.06803918,
      "balance_loss_mlp": 1.02639902,
      "epoch": 0.15150604220525463,
      "flos": 22127006465280.0,
      "grad_norm": 1.8820076689272485,
      "language_loss": 0.83314407,
      "learning_rate": 3.847236961646633e-06,
      "loss": 0.85568243,
      "num_input_tokens_seen": 26686075,
      "step": 1260,
      "time_per_iteration": 2.522284746170044
    },
    {
      "auxiliary_loss_clip": 0.01192732,
      "auxiliary_loss_mlp": 0.01050287,
      "balance_loss_clip": 1.06183481,
      "balance_loss_mlp": 1.04050541,
      "epoch": 0.1516262850958937,
      "flos": 12968708515200.0,
      "grad_norm": 2.2137828507661417,
      "language_loss": 0.7848689,
      "learning_rate": 3.846938231536296e-06,
      "loss": 0.80729902,
      "num_input_tokens_seen": 26701695,
      "step": 1261,
      "time_per_iteration": 2.553976535797119
    },
    {
      "auxiliary_loss_clip": 0.01222371,
      "auxiliary_loss_mlp": 0.01042765,
      "balance_loss_clip": 1.06843984,
      "balance_loss_mlp": 1.03256059,
      "epoch": 0.1517465279865328,
      "flos": 21797130936960.0,
      "grad_norm": 1.6220275859618785,
      "language_loss": 0.80676317,
      "learning_rate": 3.8466392212468995e-06,
      "loss": 0.82941461,
      "num_input_tokens_seen": 26721885,
      "step": 1262,
      "time_per_iteration": 2.5212409496307373
    },
    {
      "auxiliary_loss_clip": 0.01100385,
      "auxiliary_loss_mlp": 0.01014094,
      "balance_loss_clip": 1.03325748,
      "balance_loss_mlp": 1.01130486,
      "epoch": 0.15186677087717187,
      "flos": 58174569901440.0,
      "grad_norm": 0.8194345273003114,
      "language_loss": 0.61961508,
      "learning_rate": 3.8463399308238e-06,
      "loss": 0.64075989,
      "num_input_tokens_seen": 26780990,
      "step": 1263,
      "time_per_iteration": 3.143031597137451
    },
    {
      "auxiliary_loss_clip": 0.01215643,
      "auxiliary_loss_mlp": 0.01048239,
      "balance_loss_clip": 1.06812525,
      "balance_loss_mlp": 1.03834462,
      "epoch": 0.15198701376781099,
      "flos": 32669696448000.0,
      "grad_norm": 2.861348128336932,
      "language_loss": 0.63879347,
      "learning_rate": 3.846040360312402e-06,
      "loss": 0.66143233,
      "num_input_tokens_seen": 26804250,
      "step": 1264,
      "time_per_iteration": 3.336589813232422
    },
    {
      "auxiliary_loss_clip": 0.01229117,
      "auxiliary_loss_mlp": 0.01040311,
      "balance_loss_clip": 1.06516171,
      "balance_loss_mlp": 1.03022611,
      "epoch": 0.15210725665845007,
      "flos": 28402575431040.0,
      "grad_norm": 2.1042744236454625,
      "language_loss": 0.81230032,
      "learning_rate": 3.8457405097581485e-06,
      "loss": 0.83499455,
      "num_input_tokens_seen": 26823240,
      "step": 1265,
      "time_per_iteration": 2.5521767139434814
    },
    {
      "auxiliary_loss_clip": 0.01172032,
      "auxiliary_loss_mlp": 0.01046916,
      "balance_loss_clip": 1.05521655,
      "balance_loss_mlp": 1.03674769,
      "epoch": 0.15222749954908915,
      "flos": 19938179393280.0,
      "grad_norm": 1.91911724527818,
      "language_loss": 0.78157377,
      "learning_rate": 3.8454403792065275e-06,
      "loss": 0.80376327,
      "num_input_tokens_seen": 26842060,
      "step": 1266,
      "time_per_iteration": 2.60555362701416
    },
    {
      "auxiliary_loss_clip": 0.01175756,
      "auxiliary_loss_mlp": 0.01040163,
      "balance_loss_clip": 1.06129575,
      "balance_loss_mlp": 1.03008437,
      "epoch": 0.15234774243972826,
      "flos": 21324223451520.0,
      "grad_norm": 1.994365011889893,
      "language_loss": 0.85539615,
      "learning_rate": 3.845139968703068e-06,
      "loss": 0.87755537,
      "num_input_tokens_seen": 26859580,
      "step": 1267,
      "time_per_iteration": 2.594228506088257
    },
    {
      "auxiliary_loss_clip": 0.0117031,
      "auxiliary_loss_mlp": 0.01041981,
      "balance_loss_clip": 1.05805206,
      "balance_loss_mlp": 1.03159809,
      "epoch": 0.15246798533036734,
      "flos": 25957812977280.0,
      "grad_norm": 2.0885397677778905,
      "language_loss": 0.83116251,
      "learning_rate": 3.844839278293342e-06,
      "loss": 0.85328543,
      "num_input_tokens_seen": 26880430,
      "step": 1268,
      "time_per_iteration": 3.4069008827209473
    },
    {
      "auxiliary_loss_clip": 0.01235798,
      "auxiliary_loss_mlp": 0.010436,
      "balance_loss_clip": 1.06972194,
      "balance_loss_mlp": 1.03290057,
      "epoch": 0.15258822822100643,
      "flos": 25811907932160.0,
      "grad_norm": 4.156160822937864,
      "language_loss": 0.76430798,
      "learning_rate": 3.8445383080229654e-06,
      "loss": 0.78710192,
      "num_input_tokens_seen": 26896445,
      "step": 1269,
      "time_per_iteration": 3.3065097332000732
    },
    {
      "auxiliary_loss_clip": 0.01192614,
      "auxiliary_loss_mlp": 0.01046604,
      "balance_loss_clip": 1.06062555,
      "balance_loss_mlp": 1.03634,
      "epoch": 0.1527084711116455,
      "flos": 25265455349760.0,
      "grad_norm": 3.6210443596574198,
      "language_loss": 0.73287261,
      "learning_rate": 3.844237057937593e-06,
      "loss": 0.75526476,
      "num_input_tokens_seen": 26915450,
      "step": 1270,
      "time_per_iteration": 3.4571926593780518
    },
    {
      "auxiliary_loss_clip": 0.01221118,
      "auxiliary_loss_mlp": 0.0104223,
      "balance_loss_clip": 1.06536674,
      "balance_loss_mlp": 1.03187037,
      "epoch": 0.15282871400228462,
      "flos": 29240227572480.0,
      "grad_norm": 2.1941191392480213,
      "language_loss": 0.77708435,
      "learning_rate": 3.843935528082926e-06,
      "loss": 0.7997179,
      "num_input_tokens_seen": 26936475,
      "step": 1271,
      "time_per_iteration": 2.5998668670654297
    },
    {
      "auxiliary_loss_clip": 0.01218437,
      "auxiliary_loss_mlp": 0.01036947,
      "balance_loss_clip": 1.06669927,
      "balance_loss_mlp": 1.02647495,
      "epoch": 0.1529489568929237,
      "flos": 20882952869760.0,
      "grad_norm": 1.9532299715171297,
      "language_loss": 0.84679854,
      "learning_rate": 3.843633718504704e-06,
      "loss": 0.86935234,
      "num_input_tokens_seen": 26954920,
      "step": 1272,
      "time_per_iteration": 2.5172414779663086
    },
    {
      "auxiliary_loss_clip": 0.01190148,
      "auxiliary_loss_mlp": 0.01055398,
      "balance_loss_clip": 1.0637176,
      "balance_loss_mlp": 1.045331,
      "epoch": 0.1530691997835628,
      "flos": 20083833043200.0,
      "grad_norm": 4.103450591303595,
      "language_loss": 0.8985815,
      "learning_rate": 3.843331629248715e-06,
      "loss": 0.92103702,
      "num_input_tokens_seen": 26972520,
      "step": 1273,
      "time_per_iteration": 2.593606472015381
    },
    {
      "auxiliary_loss_clip": 0.01232507,
      "auxiliary_loss_mlp": 0.01035289,
      "balance_loss_clip": 1.06970119,
      "balance_loss_mlp": 1.02572262,
      "epoch": 0.1531894426742019,
      "flos": 28759814144640.0,
      "grad_norm": 2.400743238880734,
      "language_loss": 0.76750219,
      "learning_rate": 3.843029260360782e-06,
      "loss": 0.79018021,
      "num_input_tokens_seen": 26990890,
      "step": 1274,
      "time_per_iteration": 2.5412042140960693
    },
    {
      "auxiliary_loss_clip": 0.01215027,
      "auxiliary_loss_mlp": 0.01050899,
      "balance_loss_clip": 1.06840062,
      "balance_loss_mlp": 1.04101062,
      "epoch": 0.15330968556484098,
      "flos": 22236282616320.0,
      "grad_norm": 1.8813892491913902,
      "language_loss": 0.78993058,
      "learning_rate": 3.8427266118867755e-06,
      "loss": 0.81258976,
      "num_input_tokens_seen": 27010640,
      "step": 1275,
      "time_per_iteration": 2.5872414112091064
    },
    {
      "auxiliary_loss_clip": 0.01201667,
      "auxiliary_loss_mlp": 0.01041177,
      "balance_loss_clip": 1.06608748,
      "balance_loss_mlp": 1.0305438,
      "epoch": 0.15342992845548006,
      "flos": 27527504296320.0,
      "grad_norm": 2.0130295149647757,
      "language_loss": 0.82632673,
      "learning_rate": 3.842423683872608e-06,
      "loss": 0.84875512,
      "num_input_tokens_seen": 27031215,
      "step": 1276,
      "time_per_iteration": 2.603322982788086
    },
    {
      "auxiliary_loss_clip": 0.01211349,
      "auxiliary_loss_mlp": 0.01041153,
      "balance_loss_clip": 1.062482,
      "balance_loss_mlp": 1.03077626,
      "epoch": 0.15355017134611917,
      "flos": 19609596754560.0,
      "grad_norm": 2.6238906278286493,
      "language_loss": 0.77468681,
      "learning_rate": 3.842120476364232e-06,
      "loss": 0.79721189,
      "num_input_tokens_seen": 27049665,
      "step": 1277,
      "time_per_iteration": 2.553133964538574
    },
    {
      "auxiliary_loss_clip": 0.01219452,
      "auxiliary_loss_mlp": 0.01040795,
      "balance_loss_clip": 1.06534529,
      "balance_loss_mlp": 1.03005993,
      "epoch": 0.15367041423675826,
      "flos": 18478590238080.0,
      "grad_norm": 2.0900399706312793,
      "language_loss": 0.83760548,
      "learning_rate": 3.841816989407644e-06,
      "loss": 0.86020797,
      "num_input_tokens_seen": 27065155,
      "step": 1278,
      "time_per_iteration": 2.4942965507507324
    },
    {
      "auxiliary_loss_clip": 0.01179704,
      "auxiliary_loss_mlp": 0.010428,
      "balance_loss_clip": 1.05924845,
      "balance_loss_mlp": 1.03321028,
      "epoch": 0.15379065712739734,
      "flos": 41427662342400.0,
      "grad_norm": 2.532464904028034,
      "language_loss": 0.76734936,
      "learning_rate": 3.841513223048884e-06,
      "loss": 0.78957444,
      "num_input_tokens_seen": 27085840,
      "step": 1279,
      "time_per_iteration": 2.7741076946258545
    },
    {
      "auxiliary_loss_clip": 0.01182079,
      "auxiliary_loss_mlp": 0.01042676,
      "balance_loss_clip": 1.06042171,
      "balance_loss_mlp": 1.03222156,
      "epoch": 0.15391090001803642,
      "flos": 22054215553920.0,
      "grad_norm": 2.4862515931981095,
      "language_loss": 0.78520322,
      "learning_rate": 3.841209177334031e-06,
      "loss": 0.80745077,
      "num_input_tokens_seen": 27104200,
      "step": 1280,
      "time_per_iteration": 2.615844249725342
    },
    {
      "auxiliary_loss_clip": 0.01213634,
      "auxiliary_loss_mlp": 0.01041002,
      "balance_loss_clip": 1.06682992,
      "balance_loss_mlp": 1.03094697,
      "epoch": 0.15403114290867553,
      "flos": 15450351258240.0,
      "grad_norm": 3.6544813138378918,
      "language_loss": 0.75005186,
      "learning_rate": 3.84090485230921e-06,
      "loss": 0.77259827,
      "num_input_tokens_seen": 27122440,
      "step": 1281,
      "time_per_iteration": 2.4972565174102783
    },
    {
      "auxiliary_loss_clip": 0.0122986,
      "auxiliary_loss_mlp": 0.01041908,
      "balance_loss_clip": 1.06781662,
      "balance_loss_mlp": 1.03176904,
      "epoch": 0.15415138579931462,
      "flos": 17929156826880.0,
      "grad_norm": 2.488696301385415,
      "language_loss": 0.77055323,
      "learning_rate": 3.840600248020588e-06,
      "loss": 0.79327095,
      "num_input_tokens_seen": 27139380,
      "step": 1282,
      "time_per_iteration": 2.443368911743164
    },
    {
      "auxiliary_loss_clip": 0.01205055,
      "auxiliary_loss_mlp": 0.01041244,
      "balance_loss_clip": 1.06155729,
      "balance_loss_mlp": 1.03005016,
      "epoch": 0.1542716286899537,
      "flos": 11429325296640.0,
      "grad_norm": 2.1202418620988226,
      "language_loss": 0.80228436,
      "learning_rate": 3.840295364514371e-06,
      "loss": 0.82474732,
      "num_input_tokens_seen": 27156760,
      "step": 1283,
      "time_per_iteration": 2.568103790283203
    },
    {
      "auxiliary_loss_clip": 0.01201451,
      "auxiliary_loss_mlp": 0.0104223,
      "balance_loss_clip": 1.06387055,
      "balance_loss_mlp": 1.0320859,
      "epoch": 0.1543918715805928,
      "flos": 17420338719360.0,
      "grad_norm": 2.3358260929144676,
      "language_loss": 0.79110163,
      "learning_rate": 3.83999020183681e-06,
      "loss": 0.81353843,
      "num_input_tokens_seen": 27175455,
      "step": 1284,
      "time_per_iteration": 2.5243980884552
    },
    {
      "auxiliary_loss_clip": 0.01148837,
      "auxiliary_loss_mlp": 0.01049997,
      "balance_loss_clip": 1.05503869,
      "balance_loss_mlp": 1.04004955,
      "epoch": 0.1545121144712319,
      "flos": 17786376264960.0,
      "grad_norm": 2.1372786039057843,
      "language_loss": 0.78644562,
      "learning_rate": 3.839684760034199e-06,
      "loss": 0.80843389,
      "num_input_tokens_seen": 27193660,
      "step": 1285,
      "time_per_iteration": 2.636122465133667
    },
    {
      "auxiliary_loss_clip": 0.011831,
      "auxiliary_loss_mlp": 0.01043773,
      "balance_loss_clip": 1.06483912,
      "balance_loss_mlp": 1.03377759,
      "epoch": 0.15463235736187098,
      "flos": 28220185146240.0,
      "grad_norm": 8.537811575941502,
      "language_loss": 0.65304923,
      "learning_rate": 3.8393790391528716e-06,
      "loss": 0.675318,
      "num_input_tokens_seen": 27214355,
      "step": 1286,
      "time_per_iteration": 2.6337616443634033
    },
    {
      "auxiliary_loss_clip": 0.01197015,
      "auxiliary_loss_mlp": 0.01038709,
      "balance_loss_clip": 1.06249475,
      "balance_loss_mlp": 1.0291667,
      "epoch": 0.15475260025251006,
      "flos": 22856890826880.0,
      "grad_norm": 2.1360249894860464,
      "language_loss": 0.8890475,
      "learning_rate": 3.8390730392392075e-06,
      "loss": 0.91140473,
      "num_input_tokens_seen": 27234335,
      "step": 1287,
      "time_per_iteration": 2.572516441345215
    },
    {
      "auxiliary_loss_clip": 0.01234842,
      "auxiliary_loss_mlp": 0.01043578,
      "balance_loss_clip": 1.06996894,
      "balance_loss_mlp": 1.03408897,
      "epoch": 0.15487284314314917,
      "flos": 17602872658560.0,
      "grad_norm": 2.329428949530667,
      "language_loss": 0.79137409,
      "learning_rate": 3.838766760339626e-06,
      "loss": 0.81415832,
      "num_input_tokens_seen": 27252860,
      "step": 1288,
      "time_per_iteration": 2.4679784774780273
    },
    {
      "auxiliary_loss_clip": 0.01163041,
      "auxiliary_loss_mlp": 0.01040395,
      "balance_loss_clip": 1.05738187,
      "balance_loss_mlp": 1.03034544,
      "epoch": 0.15499308603378825,
      "flos": 20082037363200.0,
      "grad_norm": 3.282836154026219,
      "language_loss": 0.79486585,
      "learning_rate": 3.838460202500587e-06,
      "loss": 0.81690013,
      "num_input_tokens_seen": 27268650,
      "step": 1289,
      "time_per_iteration": 2.5768418312072754
    },
    {
      "auxiliary_loss_clip": 0.01183733,
      "auxiliary_loss_mlp": 0.01043847,
      "balance_loss_clip": 1.06495571,
      "balance_loss_mlp": 1.03313053,
      "epoch": 0.15511332892442733,
      "flos": 15918051271680.0,
      "grad_norm": 2.439548213551908,
      "language_loss": 0.74038744,
      "learning_rate": 3.838153365768599e-06,
      "loss": 0.76266325,
      "num_input_tokens_seen": 27285160,
      "step": 1290,
      "time_per_iteration": 3.3573570251464844
    },
    {
      "auxiliary_loss_clip": 0.01185412,
      "auxiliary_loss_mlp": 0.01045768,
      "balance_loss_clip": 1.06610596,
      "balance_loss_mlp": 1.03465819,
      "epoch": 0.15523357181506645,
      "flos": 41282475569280.0,
      "grad_norm": 2.3262128462678335,
      "language_loss": 0.75381535,
      "learning_rate": 3.837846250190206e-06,
      "loss": 0.77612722,
      "num_input_tokens_seen": 27308025,
      "step": 1291,
      "time_per_iteration": 2.756483316421509
    },
    {
      "auxiliary_loss_clip": 0.01165865,
      "auxiliary_loss_mlp": 0.00762761,
      "balance_loss_clip": 1.0608511,
      "balance_loss_mlp": 1.00013447,
      "epoch": 0.15535381470570553,
      "flos": 18478769806080.0,
      "grad_norm": 2.7160933663248343,
      "language_loss": 0.77139312,
      "learning_rate": 3.837538855811998e-06,
      "loss": 0.79067945,
      "num_input_tokens_seen": 27326200,
      "step": 1292,
      "time_per_iteration": 2.599231719970703
    },
    {
      "auxiliary_loss_clip": 0.01203965,
      "auxiliary_loss_mlp": 0.01046513,
      "balance_loss_clip": 1.06472039,
      "balance_loss_mlp": 1.03618932,
      "epoch": 0.1554740575963446,
      "flos": 13918150759680.0,
      "grad_norm": 2.2043650484058883,
      "language_loss": 0.70995462,
      "learning_rate": 3.837231182680606e-06,
      "loss": 0.73245943,
      "num_input_tokens_seen": 27344165,
      "step": 1293,
      "time_per_iteration": 2.536276340484619
    },
    {
      "auxiliary_loss_clip": 0.01219528,
      "auxiliary_loss_mlp": 0.01045294,
      "balance_loss_clip": 1.06690717,
      "balance_loss_mlp": 1.03526258,
      "epoch": 0.1555943004869837,
      "flos": 20847078161280.0,
      "grad_norm": 1.617315180711769,
      "language_loss": 0.75989842,
      "learning_rate": 3.836923230842706e-06,
      "loss": 0.78254664,
      "num_input_tokens_seen": 27363280,
      "step": 1294,
      "time_per_iteration": 3.245407819747925
    },
    {
      "auxiliary_loss_clip": 0.01170933,
      "auxiliary_loss_mlp": 0.01036943,
      "balance_loss_clip": 1.05531359,
      "balance_loss_mlp": 1.02629209,
      "epoch": 0.1557145433776228,
      "flos": 22085888371200.0,
      "grad_norm": 2.523413466720098,
      "language_loss": 0.80409789,
      "learning_rate": 3.836615000345011e-06,
      "loss": 0.82617664,
      "num_input_tokens_seen": 27381460,
      "step": 1295,
      "time_per_iteration": 3.3726446628570557
    },
    {
      "auxiliary_loss_clip": 0.01230676,
      "auxiliary_loss_mlp": 0.01038058,
      "balance_loss_clip": 1.06833136,
      "balance_loss_mlp": 1.02874756,
      "epoch": 0.1558347862682619,
      "flos": 19791987039360.0,
      "grad_norm": 2.1211117119072984,
      "language_loss": 0.78068423,
      "learning_rate": 3.836306491234282e-06,
      "loss": 0.80337161,
      "num_input_tokens_seen": 27399310,
      "step": 1296,
      "time_per_iteration": 2.5048720836639404
    },
    {
      "auxiliary_loss_clip": 0.01193611,
      "auxiliary_loss_mlp": 0.01040075,
      "balance_loss_clip": 1.06492746,
      "balance_loss_mlp": 1.03024042,
      "epoch": 0.15595502915890097,
      "flos": 17237086508160.0,
      "grad_norm": 2.1679034055129125,
      "language_loss": 0.75242245,
      "learning_rate": 3.835997703557317e-06,
      "loss": 0.77475929,
      "num_input_tokens_seen": 27416050,
      "step": 1297,
      "time_per_iteration": 3.3387980461120605
    },
    {
      "auxiliary_loss_clip": 0.01170059,
      "auxiliary_loss_mlp": 0.0103422,
      "balance_loss_clip": 1.05455625,
      "balance_loss_mlp": 1.02437353,
      "epoch": 0.15607527204954008,
      "flos": 19719519350400.0,
      "grad_norm": 1.6294968758918935,
      "language_loss": 0.80057198,
      "learning_rate": 3.83568863736096e-06,
      "loss": 0.82261479,
      "num_input_tokens_seen": 27434920,
      "step": 1298,
      "time_per_iteration": 2.618720293045044
    },
    {
      "auxiliary_loss_clip": 0.01184921,
      "auxiliary_loss_mlp": 0.01038057,
      "balance_loss_clip": 1.05799699,
      "balance_loss_mlp": 1.02791262,
      "epoch": 0.15619551494017916,
      "flos": 18515650095360.0,
      "grad_norm": 2.84444293275084,
      "language_loss": 0.89225566,
      "learning_rate": 3.8353792926920975e-06,
      "loss": 0.91448545,
      "num_input_tokens_seen": 27453570,
      "step": 1299,
      "time_per_iteration": 2.6002206802368164
    },
    {
      "auxiliary_loss_clip": 0.01221339,
      "auxiliary_loss_mlp": 0.01040993,
      "balance_loss_clip": 1.06689525,
      "balance_loss_mlp": 1.03127706,
      "epoch": 0.15631575783081825,
      "flos": 19902125116800.0,
      "grad_norm": 2.09304826991198,
      "language_loss": 0.81856376,
      "learning_rate": 3.835069669597655e-06,
      "loss": 0.84118712,
      "num_input_tokens_seen": 27471960,
      "step": 1300,
      "time_per_iteration": 2.5164473056793213
    },
    {
      "auxiliary_loss_clip": 0.01220311,
      "auxiliary_loss_mlp": 0.00763862,
      "balance_loss_clip": 1.0657692,
      "balance_loss_mlp": 1.00015187,
      "epoch": 0.15643600072145733,
      "flos": 20777663128320.0,
      "grad_norm": 2.20911837898005,
      "language_loss": 0.7987113,
      "learning_rate": 3.834759768124603e-06,
      "loss": 0.81855309,
      "num_input_tokens_seen": 27490835,
      "step": 1301,
      "time_per_iteration": 2.536215305328369
    },
    {
      "auxiliary_loss_clip": 0.01188505,
      "auxiliary_loss_mlp": 0.01053361,
      "balance_loss_clip": 1.06431723,
      "balance_loss_mlp": 1.04371071,
      "epoch": 0.15655624361209644,
      "flos": 18546389159040.0,
      "grad_norm": 2.692213728442544,
      "language_loss": 0.76142442,
      "learning_rate": 3.834449588319953e-06,
      "loss": 0.7838431,
      "num_input_tokens_seen": 27508870,
      "step": 1302,
      "time_per_iteration": 2.5457448959350586
    },
    {
      "auxiliary_loss_clip": 0.01214599,
      "auxiliary_loss_mlp": 0.01051203,
      "balance_loss_clip": 1.07073689,
      "balance_loss_mlp": 1.04177356,
      "epoch": 0.15667648650273552,
      "flos": 25229544727680.0,
      "grad_norm": 1.983897442373497,
      "language_loss": 0.85384738,
      "learning_rate": 3.834139130230758e-06,
      "loss": 0.87650543,
      "num_input_tokens_seen": 27528175,
      "step": 1303,
      "time_per_iteration": 2.560774803161621
    },
    {
      "auxiliary_loss_clip": 0.01205264,
      "auxiliary_loss_mlp": 0.01035725,
      "balance_loss_clip": 1.06388581,
      "balance_loss_mlp": 1.02605665,
      "epoch": 0.1567967293933746,
      "flos": 24827093769600.0,
      "grad_norm": 1.9151240778559682,
      "language_loss": 0.81250143,
      "learning_rate": 3.833828393904117e-06,
      "loss": 0.83491135,
      "num_input_tokens_seen": 27548455,
      "step": 1304,
      "time_per_iteration": 2.574538469314575
    },
    {
      "auxiliary_loss_clip": 0.0116246,
      "auxiliary_loss_mlp": 0.01041192,
      "balance_loss_clip": 1.05579853,
      "balance_loss_mlp": 1.03192961,
      "epoch": 0.15691697228401372,
      "flos": 19164555244800.0,
      "grad_norm": 2.1007318563073385,
      "language_loss": 0.77383971,
      "learning_rate": 3.833517379387165e-06,
      "loss": 0.79587632,
      "num_input_tokens_seen": 27564910,
      "step": 1305,
      "time_per_iteration": 2.580143690109253
    },
    {
      "auxiliary_loss_clip": 0.01218174,
      "auxiliary_loss_mlp": 0.01041249,
      "balance_loss_clip": 1.06583679,
      "balance_loss_mlp": 1.03109288,
      "epoch": 0.1570372151746528,
      "flos": 24790931752320.0,
      "grad_norm": 2.1527981446876177,
      "language_loss": 0.88432157,
      "learning_rate": 3.833206086727085e-06,
      "loss": 0.90691584,
      "num_input_tokens_seen": 27584260,
      "step": 1306,
      "time_per_iteration": 2.5511558055877686
    },
    {
      "auxiliary_loss_clip": 0.01189806,
      "auxiliary_loss_mlp": 0.01039048,
      "balance_loss_clip": 1.05975461,
      "balance_loss_mlp": 1.02781224,
      "epoch": 0.15715745806529188,
      "flos": 24863650836480.0,
      "grad_norm": 2.470396706872111,
      "language_loss": 0.70315301,
      "learning_rate": 3.8328945159710994e-06,
      "loss": 0.72544152,
      "num_input_tokens_seen": 27604440,
      "step": 1307,
      "time_per_iteration": 2.6292171478271484
    },
    {
      "auxiliary_loss_clip": 0.01222725,
      "auxiliary_loss_mlp": 0.00762865,
      "balance_loss_clip": 1.0690763,
      "balance_loss_mlp": 1.00009453,
      "epoch": 0.157277700955931,
      "flos": 21872148491520.0,
      "grad_norm": 2.069449191840747,
      "language_loss": 0.88701963,
      "learning_rate": 3.832582667166473e-06,
      "loss": 0.90687549,
      "num_input_tokens_seen": 27624250,
      "step": 1308,
      "time_per_iteration": 2.5276970863342285
    },
    {
      "auxiliary_loss_clip": 0.01193974,
      "auxiliary_loss_mlp": 0.01042706,
      "balance_loss_clip": 1.05924845,
      "balance_loss_mlp": 1.03233516,
      "epoch": 0.15739794384657008,
      "flos": 24533344344960.0,
      "grad_norm": 1.726045381206105,
      "language_loss": 0.81442517,
      "learning_rate": 3.8322705403605125e-06,
      "loss": 0.83679199,
      "num_input_tokens_seen": 27644595,
      "step": 1309,
      "time_per_iteration": 2.5789806842803955
    },
    {
      "auxiliary_loss_clip": 0.01193887,
      "auxiliary_loss_mlp": 0.01032264,
      "balance_loss_clip": 1.06300426,
      "balance_loss_mlp": 1.02313268,
      "epoch": 0.15751818673720916,
      "flos": 17745329998080.0,
      "grad_norm": 2.110106844901691,
      "language_loss": 0.81422758,
      "learning_rate": 3.831958135600568e-06,
      "loss": 0.83648908,
      "num_input_tokens_seen": 27662145,
      "step": 1310,
      "time_per_iteration": 2.5102357864379883
    },
    {
      "auxiliary_loss_clip": 0.01219121,
      "auxiliary_loss_mlp": 0.01037676,
      "balance_loss_clip": 1.06795502,
      "balance_loss_mlp": 1.02794814,
      "epoch": 0.15763842962784824,
      "flos": 17858520731520.0,
      "grad_norm": 2.1907115045527674,
      "language_loss": 0.79626924,
      "learning_rate": 3.831645452934032e-06,
      "loss": 0.81883717,
      "num_input_tokens_seen": 27680575,
      "step": 1311,
      "time_per_iteration": 2.5114235877990723
    },
    {
      "auxiliary_loss_clip": 0.0123353,
      "auxiliary_loss_mlp": 0.01044008,
      "balance_loss_clip": 1.07094073,
      "balance_loss_mlp": 1.03361285,
      "epoch": 0.15775867251848735,
      "flos": 26980908059520.0,
      "grad_norm": 1.930703644158957,
      "language_loss": 0.79624182,
      "learning_rate": 3.831332492408336e-06,
      "loss": 0.81901717,
      "num_input_tokens_seen": 27701985,
      "step": 1312,
      "time_per_iteration": 2.530386447906494
    },
    {
      "auxiliary_loss_clip": 0.01192508,
      "auxiliary_loss_mlp": 0.0103326,
      "balance_loss_clip": 1.05944371,
      "balance_loss_mlp": 1.02321696,
      "epoch": 0.15787891540912644,
      "flos": 19240398812160.0,
      "grad_norm": 2.090171120956722,
      "language_loss": 0.69312155,
      "learning_rate": 3.831019254070957e-06,
      "loss": 0.71537924,
      "num_input_tokens_seen": 27719770,
      "step": 1313,
      "time_per_iteration": 2.5457286834716797
    },
    {
      "auxiliary_loss_clip": 0.01177339,
      "auxiliary_loss_mlp": 0.01044842,
      "balance_loss_clip": 1.06170464,
      "balance_loss_mlp": 1.0343399,
      "epoch": 0.15799915829976552,
      "flos": 27271102037760.0,
      "grad_norm": 2.508343507941923,
      "language_loss": 0.94584858,
      "learning_rate": 3.8307057379694135e-06,
      "loss": 0.96807045,
      "num_input_tokens_seen": 27739105,
      "step": 1314,
      "time_per_iteration": 2.6478512287139893
    },
    {
      "auxiliary_loss_clip": 0.01230092,
      "auxiliary_loss_mlp": 0.01035423,
      "balance_loss_clip": 1.06705952,
      "balance_loss_mlp": 1.02571368,
      "epoch": 0.15811940119040463,
      "flos": 20405520270720.0,
      "grad_norm": 2.237001927363376,
      "language_loss": 0.82239091,
      "learning_rate": 3.830391944151264e-06,
      "loss": 0.84504604,
      "num_input_tokens_seen": 27754985,
      "step": 1315,
      "time_per_iteration": 2.4749934673309326
    },
    {
      "auxiliary_loss_clip": 0.01202089,
      "auxiliary_loss_mlp": 0.01044725,
      "balance_loss_clip": 1.06374323,
      "balance_loss_mlp": 1.03508735,
      "epoch": 0.1582396440810437,
      "flos": 32599347661440.0,
      "grad_norm": 2.0920697648997346,
      "language_loss": 0.67369407,
      "learning_rate": 3.830077872664114e-06,
      "loss": 0.69616222,
      "num_input_tokens_seen": 27776110,
      "step": 1316,
      "time_per_iteration": 3.351430654525757
    },
    {
      "auxiliary_loss_clip": 0.01156025,
      "auxiliary_loss_mlp": 0.01037423,
      "balance_loss_clip": 1.0551765,
      "balance_loss_mlp": 1.0287025,
      "epoch": 0.1583598869716828,
      "flos": 33800559310080.0,
      "grad_norm": 2.2602731416517305,
      "language_loss": 0.72778875,
      "learning_rate": 3.829763523555604e-06,
      "loss": 0.7497232,
      "num_input_tokens_seen": 27796510,
      "step": 1317,
      "time_per_iteration": 2.746156692504883
    },
    {
      "auxiliary_loss_clip": 0.01211726,
      "auxiliary_loss_mlp": 0.01042146,
      "balance_loss_clip": 1.06829,
      "balance_loss_mlp": 1.03321087,
      "epoch": 0.15848012986232188,
      "flos": 24681332378880.0,
      "grad_norm": 2.1978226262810248,
      "language_loss": 0.77960503,
      "learning_rate": 3.829448896873423e-06,
      "loss": 0.80214369,
      "num_input_tokens_seen": 27815610,
      "step": 1318,
      "time_per_iteration": 2.543355941772461
    },
    {
      "auxiliary_loss_clip": 0.01161398,
      "auxiliary_loss_mlp": 0.00763042,
      "balance_loss_clip": 1.06197929,
      "balance_loss_mlp": 1.00004923,
      "epoch": 0.158600372752961,
      "flos": 22602068766720.0,
      "grad_norm": 1.9774672472308699,
      "language_loss": 0.79300916,
      "learning_rate": 3.829133992665299e-06,
      "loss": 0.81225359,
      "num_input_tokens_seen": 27834735,
      "step": 1319,
      "time_per_iteration": 2.6372323036193848
    },
    {
      "auxiliary_loss_clip": 0.01202681,
      "auxiliary_loss_mlp": 0.0103946,
      "balance_loss_clip": 1.06456494,
      "balance_loss_mlp": 1.02991772,
      "epoch": 0.15872061564360007,
      "flos": 27927944092800.0,
      "grad_norm": 6.130193027262617,
      "language_loss": 0.89184564,
      "learning_rate": 3.828818810979002e-06,
      "loss": 0.91426706,
      "num_input_tokens_seen": 27853065,
      "step": 1320,
      "time_per_iteration": 3.33537220954895
    },
    {
      "auxiliary_loss_clip": 0.01233636,
      "auxiliary_loss_mlp": 0.01042864,
      "balance_loss_clip": 1.07314038,
      "balance_loss_mlp": 1.03296947,
      "epoch": 0.15884085853423915,
      "flos": 23696805525120.0,
      "grad_norm": 1.9779696948097243,
      "language_loss": 0.80714428,
      "learning_rate": 3.8285033518623454e-06,
      "loss": 0.82990932,
      "num_input_tokens_seen": 27873315,
      "step": 1321,
      "time_per_iteration": 2.497628688812256
    },
    {
      "auxiliary_loss_clip": 0.01222992,
      "auxiliary_loss_mlp": 0.01039526,
      "balance_loss_clip": 1.06948614,
      "balance_loss_mlp": 1.02808166,
      "epoch": 0.15896110142487826,
      "flos": 23112359331840.0,
      "grad_norm": 5.656659443725247,
      "language_loss": 0.81327581,
      "learning_rate": 3.8281876153631845e-06,
      "loss": 0.83590102,
      "num_input_tokens_seen": 27890070,
      "step": 1322,
      "time_per_iteration": 3.2457497119903564
    },
    {
      "auxiliary_loss_clip": 0.01169409,
      "auxiliary_loss_mlp": 0.01037929,
      "balance_loss_clip": 1.05998826,
      "balance_loss_mlp": 1.02790999,
      "epoch": 0.15908134431551735,
      "flos": 14685238632960.0,
      "grad_norm": 2.008971357633955,
      "language_loss": 0.64669454,
      "learning_rate": 3.827871601529416e-06,
      "loss": 0.66876793,
      "num_input_tokens_seen": 27908590,
      "step": 1323,
      "time_per_iteration": 3.416466236114502
    },
    {
      "auxiliary_loss_clip": 0.01180867,
      "auxiliary_loss_mlp": 0.010359,
      "balance_loss_clip": 1.06068659,
      "balance_loss_mlp": 1.02632165,
      "epoch": 0.15920158720615643,
      "flos": 20193611984640.0,
      "grad_norm": 2.051079970606802,
      "language_loss": 0.80634069,
      "learning_rate": 3.827555310408979e-06,
      "loss": 0.82850838,
      "num_input_tokens_seen": 27927985,
      "step": 1324,
      "time_per_iteration": 2.588622808456421
    },
    {
      "auxiliary_loss_clip": 0.01181442,
      "auxiliary_loss_mlp": 0.01047314,
      "balance_loss_clip": 1.06330812,
      "balance_loss_mlp": 1.03767598,
      "epoch": 0.1593218300967955,
      "flos": 24826626892800.0,
      "grad_norm": 1.7489067922827104,
      "language_loss": 0.82704294,
      "learning_rate": 3.827238742049854e-06,
      "loss": 0.84933054,
      "num_input_tokens_seen": 27948280,
      "step": 1325,
      "time_per_iteration": 2.6101632118225098
    },
    {
      "auxiliary_loss_clip": 0.01229951,
      "auxiliary_loss_mlp": 0.01034994,
      "balance_loss_clip": 1.06750011,
      "balance_loss_mlp": 1.02430129,
      "epoch": 0.15944207298743462,
      "flos": 28328707111680.0,
      "grad_norm": 1.9482494700290918,
      "language_loss": 0.51733297,
      "learning_rate": 3.826921896500066e-06,
      "loss": 0.53998244,
      "num_input_tokens_seen": 27969565,
      "step": 1326,
      "time_per_iteration": 2.550607442855835
    },
    {
      "auxiliary_loss_clip": 0.01194224,
      "auxiliary_loss_mlp": 0.01043919,
      "balance_loss_clip": 1.06681967,
      "balance_loss_mlp": 1.03434622,
      "epoch": 0.1595623158780737,
      "flos": 22964838174720.0,
      "grad_norm": 1.7985937082948331,
      "language_loss": 0.77981317,
      "learning_rate": 3.826604773807678e-06,
      "loss": 0.8021946,
      "num_input_tokens_seen": 27987540,
      "step": 1327,
      "time_per_iteration": 2.6059305667877197
    },
    {
      "auxiliary_loss_clip": 0.01197115,
      "auxiliary_loss_mlp": 0.01037668,
      "balance_loss_clip": 1.0593586,
      "balance_loss_mlp": 1.02573526,
      "epoch": 0.1596825587687128,
      "flos": 19710540950400.0,
      "grad_norm": 2.8525714598516805,
      "language_loss": 0.73728055,
      "learning_rate": 3.826287374020798e-06,
      "loss": 0.7596283,
      "num_input_tokens_seen": 28002345,
      "step": 1328,
      "time_per_iteration": 2.522134780883789
    },
    {
      "auxiliary_loss_clip": 0.01236157,
      "auxiliary_loss_mlp": 0.01039269,
      "balance_loss_clip": 1.0751996,
      "balance_loss_mlp": 1.02970254,
      "epoch": 0.1598028016593519,
      "flos": 22637727993600.0,
      "grad_norm": 2.008823828749455,
      "language_loss": 0.81773835,
      "learning_rate": 3.825969697187575e-06,
      "loss": 0.84049261,
      "num_input_tokens_seen": 28021675,
      "step": 1329,
      "time_per_iteration": 2.5200088024139404
    },
    {
      "auxiliary_loss_clip": 0.01182141,
      "auxiliary_loss_mlp": 0.01035658,
      "balance_loss_clip": 1.06124008,
      "balance_loss_mlp": 1.02618074,
      "epoch": 0.15992304454999098,
      "flos": 20482908122880.0,
      "grad_norm": 2.063328543063949,
      "language_loss": 0.69497037,
      "learning_rate": 3.8256517433562015e-06,
      "loss": 0.71714836,
      "num_input_tokens_seen": 28039615,
      "step": 1330,
      "time_per_iteration": 2.564527988433838
    },
    {
      "auxiliary_loss_clip": 0.01233773,
      "auxiliary_loss_mlp": 0.01037588,
      "balance_loss_clip": 1.07253051,
      "balance_loss_mlp": 1.0281111,
      "epoch": 0.16004328744063007,
      "flos": 17676094533120.0,
      "grad_norm": 2.393507431881445,
      "language_loss": 0.91502708,
      "learning_rate": 3.82533351257491e-06,
      "loss": 0.93774068,
      "num_input_tokens_seen": 28057565,
      "step": 1331,
      "time_per_iteration": 2.486743688583374
    },
    {
      "auxiliary_loss_clip": 0.0121362,
      "auxiliary_loss_mlp": 0.01040849,
      "balance_loss_clip": 1.06803739,
      "balance_loss_mlp": 1.03137827,
      "epoch": 0.16016353033126918,
      "flos": 24098717779200.0,
      "grad_norm": 2.2955308912138874,
      "language_loss": 0.88612533,
      "learning_rate": 3.825015004891975e-06,
      "loss": 0.90867007,
      "num_input_tokens_seen": 28076305,
      "step": 1332,
      "time_per_iteration": 2.535738706588745
    },
    {
      "auxiliary_loss_clip": 0.01212612,
      "auxiliary_loss_mlp": 0.01040292,
      "balance_loss_clip": 1.06620932,
      "balance_loss_mlp": 1.03020728,
      "epoch": 0.16028377322190826,
      "flos": 27634841112960.0,
      "grad_norm": 2.0319955450616844,
      "language_loss": 0.75792092,
      "learning_rate": 3.824696220355716e-06,
      "loss": 0.78044999,
      "num_input_tokens_seen": 28097895,
      "step": 1333,
      "time_per_iteration": 2.604069232940674
    },
    {
      "auxiliary_loss_clip": 0.011967,
      "auxiliary_loss_mlp": 0.01038253,
      "balance_loss_clip": 1.06500816,
      "balance_loss_mlp": 1.02729166,
      "epoch": 0.16040401611254734,
      "flos": 20961202648320.0,
      "grad_norm": 1.7281520048897914,
      "language_loss": 0.78808647,
      "learning_rate": 3.824377159014491e-06,
      "loss": 0.81043601,
      "num_input_tokens_seen": 28118790,
      "step": 1334,
      "time_per_iteration": 2.568821907043457
    },
    {
      "auxiliary_loss_clip": 0.0121396,
      "auxiliary_loss_mlp": 0.01036874,
      "balance_loss_clip": 1.06897533,
      "balance_loss_mlp": 1.02718782,
      "epoch": 0.16052425900318643,
      "flos": 21247051080960.0,
      "grad_norm": 2.6470223979668557,
      "language_loss": 0.8485254,
      "learning_rate": 3.824057820916702e-06,
      "loss": 0.87103373,
      "num_input_tokens_seen": 28135995,
      "step": 1335,
      "time_per_iteration": 2.530348300933838
    },
    {
      "auxiliary_loss_clip": 0.012055,
      "auxiliary_loss_mlp": 0.01038584,
      "balance_loss_clip": 1.06710005,
      "balance_loss_mlp": 1.02741456,
      "epoch": 0.16064450189382554,
      "flos": 15524004096000.0,
      "grad_norm": 2.48239093332264,
      "language_loss": 0.71339142,
      "learning_rate": 3.8237382061107904e-06,
      "loss": 0.73583233,
      "num_input_tokens_seen": 28152715,
      "step": 1336,
      "time_per_iteration": 2.512709856033325
    },
    {
      "auxiliary_loss_clip": 0.01134976,
      "auxiliary_loss_mlp": 0.01042467,
      "balance_loss_clip": 1.05209231,
      "balance_loss_mlp": 1.03269768,
      "epoch": 0.16076474478446462,
      "flos": 21178497974400.0,
      "grad_norm": 1.9174083408361586,
      "language_loss": 0.78846514,
      "learning_rate": 3.823418314645243e-06,
      "loss": 0.81023955,
      "num_input_tokens_seen": 28171590,
      "step": 1337,
      "time_per_iteration": 2.6816155910491943
    },
    {
      "auxiliary_loss_clip": 0.01152735,
      "auxiliary_loss_mlp": 0.01042145,
      "balance_loss_clip": 1.05838752,
      "balance_loss_mlp": 1.03182161,
      "epoch": 0.1608849876751037,
      "flos": 18366476912640.0,
      "grad_norm": 1.860664242697388,
      "language_loss": 0.75271475,
      "learning_rate": 3.823098146568588e-06,
      "loss": 0.77466357,
      "num_input_tokens_seen": 28191295,
      "step": 1338,
      "time_per_iteration": 2.610321521759033
    },
    {
      "auxiliary_loss_clip": 0.01219367,
      "auxiliary_loss_mlp": 0.01033486,
      "balance_loss_clip": 1.07055354,
      "balance_loss_mlp": 1.02396095,
      "epoch": 0.1610052305657428,
      "flos": 29497024880640.0,
      "grad_norm": 1.9681194401659374,
      "language_loss": 0.71415246,
      "learning_rate": 3.822777701929394e-06,
      "loss": 0.73668098,
      "num_input_tokens_seen": 28213120,
      "step": 1339,
      "time_per_iteration": 2.605142116546631
    },
    {
      "auxiliary_loss_clip": 0.01203818,
      "auxiliary_loss_mlp": 0.01038975,
      "balance_loss_clip": 1.06418276,
      "balance_loss_mlp": 1.02933645,
      "epoch": 0.1611254734563819,
      "flos": 26797871329920.0,
      "grad_norm": 2.0257914292597734,
      "language_loss": 0.73490328,
      "learning_rate": 3.8224569807762714e-06,
      "loss": 0.75733113,
      "num_input_tokens_seen": 28232440,
      "step": 1340,
      "time_per_iteration": 2.5629687309265137
    },
    {
      "auxiliary_loss_clip": 0.01155353,
      "auxiliary_loss_mlp": 0.01041541,
      "balance_loss_clip": 1.05754197,
      "balance_loss_mlp": 1.0309906,
      "epoch": 0.16124571634702098,
      "flos": 22419570741120.0,
      "grad_norm": 1.841128307339568,
      "language_loss": 0.76894319,
      "learning_rate": 3.822135983157873e-06,
      "loss": 0.79091215,
      "num_input_tokens_seen": 28251715,
      "step": 1341,
      "time_per_iteration": 2.631728410720825
    },
    {
      "auxiliary_loss_clip": 0.01228669,
      "auxiliary_loss_mlp": 0.00763173,
      "balance_loss_clip": 1.06843472,
      "balance_loss_mlp": 0.99998474,
      "epoch": 0.16136595923766006,
      "flos": 10999116103680.0,
      "grad_norm": 2.8256766366828723,
      "language_loss": 0.84146512,
      "learning_rate": 3.821814709122896e-06,
      "loss": 0.8613835,
      "num_input_tokens_seen": 28269765,
      "step": 1342,
      "time_per_iteration": 3.1904115676879883
    },
    {
      "auxiliary_loss_clip": 0.01202967,
      "auxiliary_loss_mlp": 0.01037027,
      "balance_loss_clip": 1.06746721,
      "balance_loss_mlp": 1.02846754,
      "epoch": 0.16148620212829917,
      "flos": 21214983214080.0,
      "grad_norm": 2.5364497980133702,
      "language_loss": 0.84602767,
      "learning_rate": 3.821493158720076e-06,
      "loss": 0.86842763,
      "num_input_tokens_seen": 28288870,
      "step": 1343,
      "time_per_iteration": 2.558246612548828
    },
    {
      "auxiliary_loss_clip": 0.01187181,
      "auxiliary_loss_mlp": 0.01041394,
      "balance_loss_clip": 1.06021988,
      "balance_loss_mlp": 1.03086209,
      "epoch": 0.16160644501893826,
      "flos": 16758468760320.0,
      "grad_norm": 3.6188432377477477,
      "language_loss": 0.73220468,
      "learning_rate": 3.821171331998191e-06,
      "loss": 0.75449049,
      "num_input_tokens_seen": 28305400,
      "step": 1344,
      "time_per_iteration": 2.5439980030059814
    },
    {
      "auxiliary_loss_clip": 0.01104498,
      "auxiliary_loss_mlp": 0.01006691,
      "balance_loss_clip": 1.04332137,
      "balance_loss_mlp": 1.00381851,
      "epoch": 0.16172668790957734,
      "flos": 64444967308800.0,
      "grad_norm": 0.7285667685570335,
      "language_loss": 0.54465449,
      "learning_rate": 3.820849229006064e-06,
      "loss": 0.56576633,
      "num_input_tokens_seen": 28373150,
      "step": 1345,
      "time_per_iteration": 3.3187265396118164
    },
    {
      "auxiliary_loss_clip": 0.0123205,
      "auxiliary_loss_mlp": 0.01035545,
      "balance_loss_clip": 1.06879032,
      "balance_loss_mlp": 1.02598405,
      "epoch": 0.16184693080021645,
      "flos": 23257689759360.0,
      "grad_norm": 1.950969683626627,
      "language_loss": 0.70634401,
      "learning_rate": 3.8205268497925564e-06,
      "loss": 0.72902,
      "num_input_tokens_seen": 28393620,
      "step": 1346,
      "time_per_iteration": 3.3106627464294434
    },
    {
      "auxiliary_loss_clip": 0.01232794,
      "auxiliary_loss_mlp": 0.01037959,
      "balance_loss_clip": 1.07206297,
      "balance_loss_mlp": 1.0278796,
      "epoch": 0.16196717369085553,
      "flos": 17451113696640.0,
      "grad_norm": 2.7349630469156225,
      "language_loss": 0.78718638,
      "learning_rate": 3.8202041944065725e-06,
      "loss": 0.80989391,
      "num_input_tokens_seen": 28409440,
      "step": 1347,
      "time_per_iteration": 2.450683116912842
    },
    {
      "auxiliary_loss_clip": 0.01238077,
      "auxiliary_loss_mlp": 0.01046928,
      "balance_loss_clip": 1.07602477,
      "balance_loss_mlp": 1.0375998,
      "epoch": 0.16208741658149461,
      "flos": 23873377806720.0,
      "grad_norm": 2.3459986007039837,
      "language_loss": 0.74048555,
      "learning_rate": 3.819881262897061e-06,
      "loss": 0.76333559,
      "num_input_tokens_seen": 28427575,
      "step": 1348,
      "time_per_iteration": 4.026902437210083
    },
    {
      "auxiliary_loss_clip": 0.01191363,
      "auxiliary_loss_mlp": 0.01044135,
      "balance_loss_clip": 1.06866765,
      "balance_loss_mlp": 1.03296554,
      "epoch": 0.1622076594721337,
      "flos": 25884806584320.0,
      "grad_norm": 1.8713317544338712,
      "language_loss": 0.73385125,
      "learning_rate": 3.819558055313008e-06,
      "loss": 0.75620627,
      "num_input_tokens_seen": 28448260,
      "step": 1349,
      "time_per_iteration": 2.6188266277313232
    },
    {
      "auxiliary_loss_clip": 0.0122441,
      "auxiliary_loss_mlp": 0.01041002,
      "balance_loss_clip": 1.06971025,
      "balance_loss_mlp": 1.0311079,
      "epoch": 0.1623279023627728,
      "flos": 21539759011200.0,
      "grad_norm": 2.2955405247983918,
      "language_loss": 0.77274656,
      "learning_rate": 3.819234571703444e-06,
      "loss": 0.79540068,
      "num_input_tokens_seen": 28467085,
      "step": 1350,
      "time_per_iteration": 2.5220553874969482
    },
    {
      "auxiliary_loss_clip": 0.01213524,
      "auxiliary_loss_mlp": 0.01045403,
      "balance_loss_clip": 1.06794071,
      "balance_loss_mlp": 1.0351156,
      "epoch": 0.1624481452534119,
      "flos": 22085421494400.0,
      "grad_norm": 1.9760769026162388,
      "language_loss": 0.85449624,
      "learning_rate": 3.8189108121174435e-06,
      "loss": 0.87708557,
      "num_input_tokens_seen": 28486850,
      "step": 1351,
      "time_per_iteration": 2.5534543991088867
    },
    {
      "auxiliary_loss_clip": 0.01180584,
      "auxiliary_loss_mlp": 0.01038018,
      "balance_loss_clip": 1.06514311,
      "balance_loss_mlp": 1.02801609,
      "epoch": 0.16256838814405097,
      "flos": 27087490690560.0,
      "grad_norm": 1.6475827997468981,
      "language_loss": 0.83675516,
      "learning_rate": 3.818586776604118e-06,
      "loss": 0.8589412,
      "num_input_tokens_seen": 28507490,
      "step": 1352,
      "time_per_iteration": 2.6812973022460938
    },
    {
      "auxiliary_loss_clip": 0.01196077,
      "auxiliary_loss_mlp": 0.01036354,
      "balance_loss_clip": 1.06434679,
      "balance_loss_mlp": 1.02703738,
      "epoch": 0.16268863103469008,
      "flos": 20120354196480.0,
      "grad_norm": 1.9199080452929496,
      "language_loss": 0.61486042,
      "learning_rate": 3.818262465212625e-06,
      "loss": 0.63718474,
      "num_input_tokens_seen": 28527615,
      "step": 1353,
      "time_per_iteration": 2.5714099407196045
    },
    {
      "auxiliary_loss_clip": 0.01205992,
      "auxiliary_loss_mlp": 0.01040656,
      "balance_loss_clip": 1.06646371,
      "balance_loss_mlp": 1.03024364,
      "epoch": 0.16280887392532917,
      "flos": 18332792933760.0,
      "grad_norm": 2.0167963035804672,
      "language_loss": 0.77255577,
      "learning_rate": 3.817937877992161e-06,
      "loss": 0.79502225,
      "num_input_tokens_seen": 28544910,
      "step": 1354,
      "time_per_iteration": 2.5057339668273926
    },
    {
      "auxiliary_loss_clip": 0.01186515,
      "auxiliary_loss_mlp": 0.00764316,
      "balance_loss_clip": 1.06198573,
      "balance_loss_mlp": 0.9999662,
      "epoch": 0.16292911681596825,
      "flos": 11874330892800.0,
      "grad_norm": 2.639821713956828,
      "language_loss": 0.85383457,
      "learning_rate": 3.817613014991967e-06,
      "loss": 0.87334287,
      "num_input_tokens_seen": 28561050,
      "step": 1355,
      "time_per_iteration": 2.5903990268707275
    },
    {
      "auxiliary_loss_clip": 0.01168818,
      "auxiliary_loss_mlp": 0.01036005,
      "balance_loss_clip": 1.05791318,
      "balance_loss_mlp": 1.02535939,
      "epoch": 0.16304935970660733,
      "flos": 26103466627200.0,
      "grad_norm": 1.9233166147113374,
      "language_loss": 0.76836491,
      "learning_rate": 3.817287876261323e-06,
      "loss": 0.79041314,
      "num_input_tokens_seen": 28581385,
      "step": 1356,
      "time_per_iteration": 2.6245508193969727
    },
    {
      "auxiliary_loss_clip": 0.01195315,
      "auxiliary_loss_mlp": 0.01036378,
      "balance_loss_clip": 1.06434369,
      "balance_loss_mlp": 1.02601254,
      "epoch": 0.16316960259724644,
      "flos": 29351945848320.0,
      "grad_norm": 1.796914650461569,
      "language_loss": 0.79921913,
      "learning_rate": 3.816962461849553e-06,
      "loss": 0.82153606,
      "num_input_tokens_seen": 28603255,
      "step": 1357,
      "time_per_iteration": 2.6151528358459473
    },
    {
      "auxiliary_loss_clip": 0.01188264,
      "auxiliary_loss_mlp": 0.01042251,
      "balance_loss_clip": 1.06190407,
      "balance_loss_mlp": 1.03234434,
      "epoch": 0.16328984548788553,
      "flos": 20886759711360.0,
      "grad_norm": 3.0583888120868163,
      "language_loss": 0.84372187,
      "learning_rate": 3.8166367718060235e-06,
      "loss": 0.866027,
      "num_input_tokens_seen": 28623145,
      "step": 1358,
      "time_per_iteration": 2.5495662689208984
    },
    {
      "auxiliary_loss_clip": 0.01213204,
      "auxiliary_loss_mlp": 0.01040896,
      "balance_loss_clip": 1.0660491,
      "balance_loss_mlp": 1.03107333,
      "epoch": 0.1634100883785246,
      "flos": 18041090584320.0,
      "grad_norm": 2.8791618512914714,
      "language_loss": 0.76428819,
      "learning_rate": 3.816310806180139e-06,
      "loss": 0.78682911,
      "num_input_tokens_seen": 28641555,
      "step": 1359,
      "time_per_iteration": 2.523298501968384
    },
    {
      "auxiliary_loss_clip": 0.01200883,
      "auxiliary_loss_mlp": 0.01041206,
      "balance_loss_clip": 1.06839991,
      "balance_loss_mlp": 1.03045356,
      "epoch": 0.16353033126916372,
      "flos": 24572128055040.0,
      "grad_norm": 1.6719689440458074,
      "language_loss": 0.81004012,
      "learning_rate": 3.81598456502135e-06,
      "loss": 0.832461,
      "num_input_tokens_seen": 28661575,
      "step": 1360,
      "time_per_iteration": 2.569723606109619
    },
    {
      "auxiliary_loss_clip": 0.01192728,
      "auxiliary_loss_mlp": 0.01045268,
      "balance_loss_clip": 1.06389415,
      "balance_loss_mlp": 1.03531373,
      "epoch": 0.1636505741598028,
      "flos": 19892895321600.0,
      "grad_norm": 1.9012235202621883,
      "language_loss": 0.87010586,
      "learning_rate": 3.8156580483791455e-06,
      "loss": 0.8924858,
      "num_input_tokens_seen": 28676765,
      "step": 1361,
      "time_per_iteration": 2.5237433910369873
    },
    {
      "auxiliary_loss_clip": 0.01232822,
      "auxiliary_loss_mlp": 0.01033333,
      "balance_loss_clip": 1.0712769,
      "balance_loss_mlp": 1.02312231,
      "epoch": 0.16377081705044189,
      "flos": 28402611344640.0,
      "grad_norm": 2.111529834887323,
      "language_loss": 0.76671976,
      "learning_rate": 3.815331256303059e-06,
      "loss": 0.78938133,
      "num_input_tokens_seen": 28696795,
      "step": 1362,
      "time_per_iteration": 2.534902811050415
    },
    {
      "auxiliary_loss_clip": 0.01182751,
      "auxiliary_loss_mlp": 0.01034555,
      "balance_loss_clip": 1.06605947,
      "balance_loss_mlp": 1.02364719,
      "epoch": 0.163891059941081,
      "flos": 21908059113600.0,
      "grad_norm": 2.3042445604089665,
      "language_loss": 0.77210957,
      "learning_rate": 3.815004188842665e-06,
      "loss": 0.79428267,
      "num_input_tokens_seen": 28714835,
      "step": 1363,
      "time_per_iteration": 2.5863635540008545
    },
    {
      "auxiliary_loss_clip": 0.01188166,
      "auxiliary_loss_mlp": 0.01033403,
      "balance_loss_clip": 1.05729079,
      "balance_loss_mlp": 1.02362168,
      "epoch": 0.16401130283172008,
      "flos": 26797619934720.0,
      "grad_norm": 1.6598259044386634,
      "language_loss": 0.79773265,
      "learning_rate": 3.814676846047578e-06,
      "loss": 0.81994832,
      "num_input_tokens_seen": 28735710,
      "step": 1364,
      "time_per_iteration": 2.5926711559295654
    },
    {
      "auxiliary_loss_clip": 0.01213702,
      "auxiliary_loss_mlp": 0.01046916,
      "balance_loss_clip": 1.06744361,
      "balance_loss_mlp": 1.03601992,
      "epoch": 0.16413154572235916,
      "flos": 32997417160320.0,
      "grad_norm": 2.046966444360988,
      "language_loss": 0.69829565,
      "learning_rate": 3.8143492279674565e-06,
      "loss": 0.72090179,
      "num_input_tokens_seen": 28758405,
      "step": 1365,
      "time_per_iteration": 2.6342074871063232
    },
    {
      "auxiliary_loss_clip": 0.01089315,
      "auxiliary_loss_mlp": 0.01004929,
      "balance_loss_clip": 1.03356707,
      "balance_loss_mlp": 1.00216293,
      "epoch": 0.16425178861299825,
      "flos": 40113622074240.0,
      "grad_norm": 0.8360547672994669,
      "language_loss": 0.58430004,
      "learning_rate": 3.8140213346519997e-06,
      "loss": 0.60524249,
      "num_input_tokens_seen": 28809000,
      "step": 1366,
      "time_per_iteration": 2.907327651977539
    },
    {
      "auxiliary_loss_clip": 0.01173537,
      "auxiliary_loss_mlp": 0.01043055,
      "balance_loss_clip": 1.06206954,
      "balance_loss_mlp": 1.03239179,
      "epoch": 0.16437203150363736,
      "flos": 25447486498560.0,
      "grad_norm": 2.355750454497719,
      "language_loss": 0.76628548,
      "learning_rate": 3.813693166150948e-06,
      "loss": 0.78845143,
      "num_input_tokens_seen": 28829210,
      "step": 1367,
      "time_per_iteration": 2.6163647174835205
    },
    {
      "auxiliary_loss_clip": 0.011786,
      "auxiliary_loss_mlp": 0.01041505,
      "balance_loss_clip": 1.06383574,
      "balance_loss_mlp": 1.03030539,
      "epoch": 0.16449227439427644,
      "flos": 23476888506240.0,
      "grad_norm": 2.1338283625477037,
      "language_loss": 0.8558892,
      "learning_rate": 3.813364722514086e-06,
      "loss": 0.87809026,
      "num_input_tokens_seen": 28847545,
      "step": 1368,
      "time_per_iteration": 3.450256586074829
    },
    {
      "auxiliary_loss_clip": 0.01211754,
      "auxiliary_loss_mlp": 0.01028913,
      "balance_loss_clip": 1.06483424,
      "balance_loss_mlp": 1.01918602,
      "epoch": 0.16461251728491552,
      "flos": 13545217802880.0,
      "grad_norm": 1.9898887250239283,
      "language_loss": 0.80607873,
      "learning_rate": 3.8130360037912368e-06,
      "loss": 0.82848543,
      "num_input_tokens_seen": 28863990,
      "step": 1369,
      "time_per_iteration": 2.5063517093658447
    },
    {
      "auxiliary_loss_clip": 0.01210319,
      "auxiliary_loss_mlp": 0.010368,
      "balance_loss_clip": 1.0638181,
      "balance_loss_mlp": 1.02675653,
      "epoch": 0.16473276017555463,
      "flos": 23003298662400.0,
      "grad_norm": 2.3364352719887385,
      "language_loss": 0.82093334,
      "learning_rate": 3.812707010032268e-06,
      "loss": 0.84340453,
      "num_input_tokens_seen": 28883045,
      "step": 1370,
      "time_per_iteration": 2.5791122913360596
    },
    {
      "auxiliary_loss_clip": 0.0122277,
      "auxiliary_loss_mlp": 0.0103857,
      "balance_loss_clip": 1.07224357,
      "balance_loss_mlp": 1.02848446,
      "epoch": 0.16485300306619372,
      "flos": 24790680357120.0,
      "grad_norm": 2.163001124671667,
      "language_loss": 0.78942114,
      "learning_rate": 3.8123777412870863e-06,
      "loss": 0.81203449,
      "num_input_tokens_seen": 28902545,
      "step": 1371,
      "time_per_iteration": 2.5799107551574707
    },
    {
      "auxiliary_loss_clip": 0.01201885,
      "auxiliary_loss_mlp": 0.01044851,
      "balance_loss_clip": 1.06481838,
      "balance_loss_mlp": 1.0345875,
      "epoch": 0.1649732459568328,
      "flos": 21106497162240.0,
      "grad_norm": 2.2231679546362955,
      "language_loss": 0.78586143,
      "learning_rate": 3.812048197605643e-06,
      "loss": 0.80832887,
      "num_input_tokens_seen": 28921440,
      "step": 1372,
      "time_per_iteration": 3.335482358932495
    },
    {
      "auxiliary_loss_clip": 0.01212759,
      "auxiliary_loss_mlp": 0.01034129,
      "balance_loss_clip": 1.06426394,
      "balance_loss_mlp": 1.02401412,
      "epoch": 0.16509348884747188,
      "flos": 20266726118400.0,
      "grad_norm": 1.8586618209360404,
      "language_loss": 0.81092381,
      "learning_rate": 3.8117183790379277e-06,
      "loss": 0.83339268,
      "num_input_tokens_seen": 28939890,
      "step": 1373,
      "time_per_iteration": 3.218022346496582
    },
    {
      "auxiliary_loss_clip": 0.01225708,
      "auxiliary_loss_mlp": 0.01038043,
      "balance_loss_clip": 1.06643057,
      "balance_loss_mlp": 1.02845216,
      "epoch": 0.165213731738111,
      "flos": 11035493602560.0,
      "grad_norm": 3.4881536802481157,
      "language_loss": 0.94165462,
      "learning_rate": 3.811388285633976e-06,
      "loss": 0.96429217,
      "num_input_tokens_seen": 28955875,
      "step": 1374,
      "time_per_iteration": 3.2937605381011963
    },
    {
      "auxiliary_loss_clip": 0.01173615,
      "auxiliary_loss_mlp": 0.01048514,
      "balance_loss_clip": 1.06234741,
      "balance_loss_mlp": 1.03769541,
      "epoch": 0.16533397462875007,
      "flos": 29972051268480.0,
      "grad_norm": 2.1623251655045603,
      "language_loss": 0.61982131,
      "learning_rate": 3.811057917443861e-06,
      "loss": 0.64204264,
      "num_input_tokens_seen": 28975140,
      "step": 1375,
      "time_per_iteration": 2.6993937492370605
    },
    {
      "auxiliary_loss_clip": 0.01106584,
      "auxiliary_loss_mlp": 0.01003996,
      "balance_loss_clip": 1.03566253,
      "balance_loss_mlp": 1.00109899,
      "epoch": 0.16545421751938916,
      "flos": 65556763027200.0,
      "grad_norm": 0.8805192363961902,
      "language_loss": 0.68247473,
      "learning_rate": 3.8107272745177e-06,
      "loss": 0.70358062,
      "num_input_tokens_seen": 29047470,
      "step": 1376,
      "time_per_iteration": 3.2831995487213135
    },
    {
      "auxiliary_loss_clip": 0.01183185,
      "auxiliary_loss_mlp": 0.01036473,
      "balance_loss_clip": 1.06116033,
      "balance_loss_mlp": 1.02668571,
      "epoch": 0.16557446041002827,
      "flos": 22492361652480.0,
      "grad_norm": 1.723534281310649,
      "language_loss": 0.78775525,
      "learning_rate": 3.8103963569056513e-06,
      "loss": 0.80995184,
      "num_input_tokens_seen": 29066605,
      "step": 1377,
      "time_per_iteration": 2.821563959121704
    },
    {
      "auxiliary_loss_clip": 0.01192875,
      "auxiliary_loss_mlp": 0.01037328,
      "balance_loss_clip": 1.05993772,
      "balance_loss_mlp": 1.02775562,
      "epoch": 0.16569470330066735,
      "flos": 24602723464320.0,
      "grad_norm": 1.9252926707613467,
      "language_loss": 0.87956989,
      "learning_rate": 3.8100651646579146e-06,
      "loss": 0.90187192,
      "num_input_tokens_seen": 29085815,
      "step": 1378,
      "time_per_iteration": 2.5722429752349854
    },
    {
      "auxiliary_loss_clip": 0.0118977,
      "auxiliary_loss_mlp": 0.01042868,
      "balance_loss_clip": 1.05921626,
      "balance_loss_mlp": 1.03294992,
      "epoch": 0.16581494619130643,
      "flos": 15006207588480.0,
      "grad_norm": 2.7233051793348424,
      "language_loss": 0.92862546,
      "learning_rate": 3.8097336978247317e-06,
      "loss": 0.95095187,
      "num_input_tokens_seen": 29102520,
      "step": 1379,
      "time_per_iteration": 2.5165750980377197
    },
    {
      "auxiliary_loss_clip": 0.01180917,
      "auxiliary_loss_mlp": 0.01044734,
      "balance_loss_clip": 1.05985045,
      "balance_loss_mlp": 1.03387976,
      "epoch": 0.16593518908194552,
      "flos": 17420338719360.0,
      "grad_norm": 1.9878376797238948,
      "language_loss": 0.88830066,
      "learning_rate": 3.8094019564563854e-06,
      "loss": 0.91055715,
      "num_input_tokens_seen": 29119450,
      "step": 1380,
      "time_per_iteration": 2.518465280532837
    },
    {
      "auxiliary_loss_clip": 0.01230153,
      "auxiliary_loss_mlp": 0.00763229,
      "balance_loss_clip": 1.0697999,
      "balance_loss_mlp": 0.99996352,
      "epoch": 0.16605543197258463,
      "flos": 20412631163520.0,
      "grad_norm": 2.392512846522975,
      "language_loss": 0.75050139,
      "learning_rate": 3.809069940603201e-06,
      "loss": 0.77043521,
      "num_input_tokens_seen": 29137405,
      "step": 1381,
      "time_per_iteration": 2.489375352859497
    },
    {
      "auxiliary_loss_clip": 0.01186049,
      "auxiliary_loss_mlp": 0.01039166,
      "balance_loss_clip": 1.06254506,
      "balance_loss_mlp": 1.02903914,
      "epoch": 0.1661756748632237,
      "flos": 14209745368320.0,
      "grad_norm": 2.101805838205608,
      "language_loss": 0.78010279,
      "learning_rate": 3.8087376503155452e-06,
      "loss": 0.80235493,
      "num_input_tokens_seen": 29154890,
      "step": 1382,
      "time_per_iteration": 2.539548635482788
    },
    {
      "auxiliary_loss_clip": 0.01103777,
      "auxiliary_loss_mlp": 0.01004752,
      "balance_loss_clip": 1.03467107,
      "balance_loss_mlp": 1.00186729,
      "epoch": 0.1662959177538628,
      "flos": 66080877350400.0,
      "grad_norm": 0.911958710379343,
      "language_loss": 0.56211305,
      "learning_rate": 3.808405085643826e-06,
      "loss": 0.58319831,
      "num_input_tokens_seen": 29219770,
      "step": 1383,
      "time_per_iteration": 3.1782608032226562
    },
    {
      "auxiliary_loss_clip": 0.01227515,
      "auxiliary_loss_mlp": 0.00762685,
      "balance_loss_clip": 1.06877446,
      "balance_loss_mlp": 0.99994051,
      "epoch": 0.1664161606445019,
      "flos": 20740567357440.0,
      "grad_norm": 2.1472831143931908,
      "language_loss": 0.88423258,
      "learning_rate": 3.8080722466384925e-06,
      "loss": 0.90413463,
      "num_input_tokens_seen": 29237620,
      "step": 1384,
      "time_per_iteration": 2.494722366333008
    },
    {
      "auxiliary_loss_clip": 0.01225343,
      "auxiliary_loss_mlp": 0.01041487,
      "balance_loss_clip": 1.06400013,
      "balance_loss_mlp": 1.03069901,
      "epoch": 0.166536403535141,
      "flos": 25260930236160.0,
      "grad_norm": 2.359143671669053,
      "language_loss": 0.70961267,
      "learning_rate": 3.8077391333500376e-06,
      "loss": 0.73228097,
      "num_input_tokens_seen": 29256760,
      "step": 1385,
      "time_per_iteration": 2.5153942108154297
    },
    {
      "auxiliary_loss_clip": 0.01196558,
      "auxiliary_loss_mlp": 0.01038678,
      "balance_loss_clip": 1.06459522,
      "balance_loss_mlp": 1.02934992,
      "epoch": 0.16665664642578007,
      "flos": 25447450584960.0,
      "grad_norm": 1.8792113523093825,
      "language_loss": 0.76733249,
      "learning_rate": 3.8074057458289934e-06,
      "loss": 0.78968483,
      "num_input_tokens_seen": 29277450,
      "step": 1386,
      "time_per_iteration": 2.612356662750244
    },
    {
      "auxiliary_loss_clip": 0.01196415,
      "auxiliary_loss_mlp": 0.01035493,
      "balance_loss_clip": 1.06204987,
      "balance_loss_mlp": 1.02537775,
      "epoch": 0.16677688931641918,
      "flos": 22200767043840.0,
      "grad_norm": 2.0184832630058938,
      "language_loss": 0.82501101,
      "learning_rate": 3.807072084125934e-06,
      "loss": 0.84733009,
      "num_input_tokens_seen": 29299300,
      "step": 1387,
      "time_per_iteration": 2.5806500911712646
    },
    {
      "auxiliary_loss_clip": 0.01197956,
      "auxiliary_loss_mlp": 0.01038064,
      "balance_loss_clip": 1.06796527,
      "balance_loss_mlp": 1.02773404,
      "epoch": 0.16689713220705826,
      "flos": 16945958776320.0,
      "grad_norm": 3.2344479930191286,
      "language_loss": 0.80292916,
      "learning_rate": 3.806738148291477e-06,
      "loss": 0.82528937,
      "num_input_tokens_seen": 29316125,
      "step": 1388,
      "time_per_iteration": 2.5384366512298584
    },
    {
      "auxiliary_loss_clip": 0.0115533,
      "auxiliary_loss_mlp": 0.01045499,
      "balance_loss_clip": 1.05822897,
      "balance_loss_mlp": 1.03422201,
      "epoch": 0.16701737509769735,
      "flos": 36244423923840.0,
      "grad_norm": 2.51016694815606,
      "language_loss": 0.71281767,
      "learning_rate": 3.8064039383762793e-06,
      "loss": 0.73482597,
      "num_input_tokens_seen": 29338490,
      "step": 1389,
      "time_per_iteration": 2.7582147121429443
    },
    {
      "auxiliary_loss_clip": 0.01207961,
      "auxiliary_loss_mlp": 0.01043142,
      "balance_loss_clip": 1.06459939,
      "balance_loss_mlp": 1.03273535,
      "epoch": 0.16713761798833643,
      "flos": 23258659426560.0,
      "grad_norm": 2.2623424481934693,
      "language_loss": 0.76998663,
      "learning_rate": 3.8060694544310396e-06,
      "loss": 0.79249763,
      "num_input_tokens_seen": 29357000,
      "step": 1390,
      "time_per_iteration": 2.5419418811798096
    },
    {
      "auxiliary_loss_clip": 0.0122903,
      "auxiliary_loss_mlp": 0.01044588,
      "balance_loss_clip": 1.06801987,
      "balance_loss_mlp": 1.03409183,
      "epoch": 0.16725786087897554,
      "flos": 25302515207040.0,
      "grad_norm": 1.9733497488128657,
      "language_loss": 0.78361475,
      "learning_rate": 3.8057346965065006e-06,
      "loss": 0.80635095,
      "num_input_tokens_seen": 29378230,
      "step": 1391,
      "time_per_iteration": 2.5255582332611084
    },
    {
      "auxiliary_loss_clip": 0.01193287,
      "auxiliary_loss_mlp": 0.01041874,
      "balance_loss_clip": 1.06223631,
      "balance_loss_mlp": 1.03252172,
      "epoch": 0.16737810376961462,
      "flos": 31831541516160.0,
      "grad_norm": 1.7524224492814058,
      "language_loss": 0.84315723,
      "learning_rate": 3.805399664653443e-06,
      "loss": 0.86550879,
      "num_input_tokens_seen": 29400370,
      "step": 1392,
      "time_per_iteration": 2.649125099182129
    },
    {
      "auxiliary_loss_clip": 0.01229916,
      "auxiliary_loss_mlp": 0.01043868,
      "balance_loss_clip": 1.06738639,
      "balance_loss_mlp": 1.0341171,
      "epoch": 0.1674983466602537,
      "flos": 27961843553280.0,
      "grad_norm": 2.405026206722115,
      "language_loss": 0.7385056,
      "learning_rate": 3.805064358922692e-06,
      "loss": 0.76124346,
      "num_input_tokens_seen": 29418660,
      "step": 1393,
      "time_per_iteration": 2.519434690475464
    },
    {
      "auxiliary_loss_clip": 0.01212753,
      "auxiliary_loss_mlp": 0.01040363,
      "balance_loss_clip": 1.06323171,
      "balance_loss_mlp": 1.02902067,
      "epoch": 0.16761858955089282,
      "flos": 21762656858880.0,
      "grad_norm": 1.9702756158318395,
      "language_loss": 0.80901164,
      "learning_rate": 3.8047287793651136e-06,
      "loss": 0.83154279,
      "num_input_tokens_seen": 29440105,
      "step": 1394,
      "time_per_iteration": 3.3505115509033203
    },
    {
      "auxiliary_loss_clip": 0.01182287,
      "auxiliary_loss_mlp": 0.01039598,
      "balance_loss_clip": 1.06085384,
      "balance_loss_mlp": 1.02919066,
      "epoch": 0.1677388324415319,
      "flos": 23805507058560.0,
      "grad_norm": 1.8523649421610568,
      "language_loss": 0.88369763,
      "learning_rate": 3.8043929260316137e-06,
      "loss": 0.90591645,
      "num_input_tokens_seen": 29458260,
      "step": 1395,
      "time_per_iteration": 2.693598985671997
    },
    {
      "auxiliary_loss_clip": 0.01199735,
      "auxiliary_loss_mlp": 0.01047626,
      "balance_loss_clip": 1.06919909,
      "balance_loss_mlp": 1.03740335,
      "epoch": 0.16785907533217098,
      "flos": 20558859431040.0,
      "grad_norm": 1.9559252679283385,
      "language_loss": 0.83569205,
      "learning_rate": 3.8040567989731417e-06,
      "loss": 0.85816562,
      "num_input_tokens_seen": 29476205,
      "step": 1396,
      "time_per_iteration": 2.661323308944702
    },
    {
      "auxiliary_loss_clip": 0.01209435,
      "auxiliary_loss_mlp": 0.01037107,
      "balance_loss_clip": 1.06733918,
      "balance_loss_mlp": 1.02765405,
      "epoch": 0.16797931822281006,
      "flos": 15669657745920.0,
      "grad_norm": 3.143436812254544,
      "language_loss": 0.79576749,
      "learning_rate": 3.8037203982406876e-06,
      "loss": 0.81823289,
      "num_input_tokens_seen": 29494370,
      "step": 1397,
      "time_per_iteration": 2.622178077697754
    },
    {
      "auxiliary_loss_clip": 0.01228072,
      "auxiliary_loss_mlp": 0.0104093,
      "balance_loss_clip": 1.06787527,
      "balance_loss_mlp": 1.02993333,
      "epoch": 0.16809956111344918,
      "flos": 16541101607040.0,
      "grad_norm": 2.008846000897291,
      "language_loss": 0.72850817,
      "learning_rate": 3.8033837238852835e-06,
      "loss": 0.75119817,
      "num_input_tokens_seen": 29511070,
      "step": 1398,
      "time_per_iteration": 3.3236799240112305
    },
    {
      "auxiliary_loss_clip": 0.01185613,
      "auxiliary_loss_mlp": 0.01037655,
      "balance_loss_clip": 1.06066108,
      "balance_loss_mlp": 1.02784419,
      "epoch": 0.16821980400408826,
      "flos": 23258084808960.0,
      "grad_norm": 1.9786421529871654,
      "language_loss": 0.69486427,
      "learning_rate": 3.8030467759580017e-06,
      "loss": 0.71709692,
      "num_input_tokens_seen": 29531990,
      "step": 1399,
      "time_per_iteration": 2.6619603633880615
    },
    {
      "auxiliary_loss_clip": 0.01214097,
      "auxiliary_loss_mlp": 0.01043887,
      "balance_loss_clip": 1.06371641,
      "balance_loss_mlp": 1.03347969,
      "epoch": 0.16834004689472734,
      "flos": 20774754126720.0,
      "grad_norm": 1.9886428492664696,
      "language_loss": 0.86894417,
      "learning_rate": 3.802709554509958e-06,
      "loss": 0.89152408,
      "num_input_tokens_seen": 29549790,
      "step": 1400,
      "time_per_iteration": 4.297313928604126
    },
    {
      "auxiliary_loss_clip": 0.01196713,
      "auxiliary_loss_mlp": 0.01042633,
      "balance_loss_clip": 1.0615139,
      "balance_loss_mlp": 1.03290582,
      "epoch": 0.16846028978536645,
      "flos": 26687302289280.0,
      "grad_norm": 1.9223240280420086,
      "language_loss": 0.7925126,
      "learning_rate": 3.8023720595923083e-06,
      "loss": 0.814906,
      "num_input_tokens_seen": 29569045,
      "step": 1401,
      "time_per_iteration": 2.6128580570220947
    },
    {
      "auxiliary_loss_clip": 0.01161336,
      "auxiliary_loss_mlp": 0.01036099,
      "balance_loss_clip": 1.05613208,
      "balance_loss_mlp": 1.02602625,
      "epoch": 0.16858053267600553,
      "flos": 18843298980480.0,
      "grad_norm": 4.817149812004977,
      "language_loss": 0.87419111,
      "learning_rate": 3.80203429125625e-06,
      "loss": 0.89616549,
      "num_input_tokens_seen": 29587220,
      "step": 1402,
      "time_per_iteration": 2.7014243602752686
    },
    {
      "auxiliary_loss_clip": 0.01141797,
      "auxiliary_loss_mlp": 0.01040197,
      "balance_loss_clip": 1.05594945,
      "balance_loss_mlp": 1.03024912,
      "epoch": 0.16870077556664462,
      "flos": 27744548227200.0,
      "grad_norm": 1.7861258458219211,
      "language_loss": 0.70018935,
      "learning_rate": 3.8016962495530225e-06,
      "loss": 0.7220093,
      "num_input_tokens_seen": 29606410,
      "step": 1403,
      "time_per_iteration": 2.6691479682922363
    },
    {
      "auxiliary_loss_clip": 0.01228933,
      "auxiliary_loss_mlp": 0.01042224,
      "balance_loss_clip": 1.06729484,
      "balance_loss_mlp": 1.03200817,
      "epoch": 0.1688210184572837,
      "flos": 13730768484480.0,
      "grad_norm": 2.5689621608904236,
      "language_loss": 0.77089375,
      "learning_rate": 3.8013579345339063e-06,
      "loss": 0.79360533,
      "num_input_tokens_seen": 29621275,
      "step": 1404,
      "time_per_iteration": 2.454477310180664
    },
    {
      "auxiliary_loss_clip": 0.01184061,
      "auxiliary_loss_mlp": 0.01040563,
      "balance_loss_clip": 1.06234932,
      "balance_loss_mlp": 1.03009033,
      "epoch": 0.1689412613479228,
      "flos": 26468785900800.0,
      "grad_norm": 1.9988838673197138,
      "language_loss": 0.69129062,
      "learning_rate": 3.801019346250224e-06,
      "loss": 0.71353686,
      "num_input_tokens_seen": 29641420,
      "step": 1405,
      "time_per_iteration": 2.6108644008636475
    },
    {
      "auxiliary_loss_clip": 0.01208261,
      "auxiliary_loss_mlp": 0.01037853,
      "balance_loss_clip": 1.06395674,
      "balance_loss_mlp": 1.02800071,
      "epoch": 0.1690615042385619,
      "flos": 21138852337920.0,
      "grad_norm": 2.7759979757427606,
      "language_loss": 0.83822215,
      "learning_rate": 3.8006804847533395e-06,
      "loss": 0.8606832,
      "num_input_tokens_seen": 29660935,
      "step": 1406,
      "time_per_iteration": 2.540719747543335
    },
    {
      "auxiliary_loss_clip": 0.01226799,
      "auxiliary_loss_mlp": 0.01046693,
      "balance_loss_clip": 1.06707001,
      "balance_loss_mlp": 1.03673911,
      "epoch": 0.16918174712920098,
      "flos": 20849340718080.0,
      "grad_norm": 2.24411229795782,
      "language_loss": 0.85196137,
      "learning_rate": 3.8003413500946556e-06,
      "loss": 0.87469637,
      "num_input_tokens_seen": 29681045,
      "step": 1407,
      "time_per_iteration": 2.483126640319824
    },
    {
      "auxiliary_loss_clip": 0.01199506,
      "auxiliary_loss_mlp": 0.01046148,
      "balance_loss_clip": 1.06616879,
      "balance_loss_mlp": 1.03517509,
      "epoch": 0.1693019900198401,
      "flos": 16983270028800.0,
      "grad_norm": 4.14568690009498,
      "language_loss": 0.82694411,
      "learning_rate": 3.8000019423256216e-06,
      "loss": 0.84940064,
      "num_input_tokens_seen": 29698810,
      "step": 1408,
      "time_per_iteration": 2.537609577178955
    },
    {
      "auxiliary_loss_clip": 0.01183898,
      "auxiliary_loss_mlp": 0.01042414,
      "balance_loss_clip": 1.06177258,
      "balance_loss_mlp": 1.03285968,
      "epoch": 0.16942223291047917,
      "flos": 26796901662720.0,
      "grad_norm": 1.5755231158279994,
      "language_loss": 0.88008082,
      "learning_rate": 3.7996622614977234e-06,
      "loss": 0.90234393,
      "num_input_tokens_seen": 29720000,
      "step": 1409,
      "time_per_iteration": 2.595411777496338
    },
    {
      "auxiliary_loss_clip": 0.0119537,
      "auxiliary_loss_mlp": 0.01033706,
      "balance_loss_clip": 1.06551099,
      "balance_loss_mlp": 1.02402008,
      "epoch": 0.16954247580111825,
      "flos": 18583700411520.0,
      "grad_norm": 2.0098084269659267,
      "language_loss": 0.79330295,
      "learning_rate": 3.799322307662492e-06,
      "loss": 0.81559372,
      "num_input_tokens_seen": 29737820,
      "step": 1410,
      "time_per_iteration": 2.5505847930908203
    },
    {
      "auxiliary_loss_clip": 0.01169828,
      "auxiliary_loss_mlp": 0.01040654,
      "balance_loss_clip": 1.06118155,
      "balance_loss_mlp": 1.03031802,
      "epoch": 0.16966271869175734,
      "flos": 13983651210240.0,
      "grad_norm": 2.7093823340228087,
      "language_loss": 0.83474135,
      "learning_rate": 3.798982080871496e-06,
      "loss": 0.85684609,
      "num_input_tokens_seen": 29752960,
      "step": 1411,
      "time_per_iteration": 2.5773837566375732
    },
    {
      "auxiliary_loss_clip": 0.01227813,
      "auxiliary_loss_mlp": 0.01045555,
      "balance_loss_clip": 1.0677321,
      "balance_loss_mlp": 1.03511775,
      "epoch": 0.16978296158239645,
      "flos": 37487328284160.0,
      "grad_norm": 2.8513743344798113,
      "language_loss": 0.67986417,
      "learning_rate": 3.798641581176349e-06,
      "loss": 0.70259786,
      "num_input_tokens_seen": 29775240,
      "step": 1412,
      "time_per_iteration": 2.64139461517334
    },
    {
      "auxiliary_loss_clip": 0.01198402,
      "auxiliary_loss_mlp": 0.01035455,
      "balance_loss_clip": 1.06337047,
      "balance_loss_mlp": 1.02551329,
      "epoch": 0.16990320447303553,
      "flos": 28328958506880.0,
      "grad_norm": 2.1614962047542474,
      "language_loss": 0.74593604,
      "learning_rate": 3.7983008086287044e-06,
      "loss": 0.76827461,
      "num_input_tokens_seen": 29796560,
      "step": 1413,
      "time_per_iteration": 2.600578784942627
    },
    {
      "auxiliary_loss_clip": 0.01195622,
      "auxiliary_loss_mlp": 0.01036551,
      "balance_loss_clip": 1.06146479,
      "balance_loss_mlp": 1.02561367,
      "epoch": 0.1700234473636746,
      "flos": 20188189031040.0,
      "grad_norm": 5.680100536750502,
      "language_loss": 0.79291528,
      "learning_rate": 3.797959763280257e-06,
      "loss": 0.81523705,
      "num_input_tokens_seen": 29815245,
      "step": 1414,
      "time_per_iteration": 2.5528366565704346
    },
    {
      "auxiliary_loss_clip": 0.01216186,
      "auxiliary_loss_mlp": 0.01036789,
      "balance_loss_clip": 1.06760156,
      "balance_loss_mlp": 1.02672756,
      "epoch": 0.17014369025431372,
      "flos": 24858658846080.0,
      "grad_norm": 1.7953407374287453,
      "language_loss": 0.78755563,
      "learning_rate": 3.797618445182743e-06,
      "loss": 0.81008542,
      "num_input_tokens_seen": 29836640,
      "step": 1415,
      "time_per_iteration": 2.546335220336914
    },
    {
      "auxiliary_loss_clip": 0.01165839,
      "auxiliary_loss_mlp": 0.01039461,
      "balance_loss_clip": 1.05987632,
      "balance_loss_mlp": 1.02935767,
      "epoch": 0.1702639331449528,
      "flos": 16467233287680.0,
      "grad_norm": 2.2206166638963385,
      "language_loss": 0.84771717,
      "learning_rate": 3.79727685438794e-06,
      "loss": 0.86977017,
      "num_input_tokens_seen": 29850830,
      "step": 1416,
      "time_per_iteration": 2.6027839183807373
    },
    {
      "auxiliary_loss_clip": 0.01116823,
      "auxiliary_loss_mlp": 0.01008732,
      "balance_loss_clip": 1.0372237,
      "balance_loss_mlp": 1.00620449,
      "epoch": 0.1703841760355919,
      "flos": 52508870979840.0,
      "grad_norm": 0.8421112231301212,
      "language_loss": 0.61671001,
      "learning_rate": 3.796934990947667e-06,
      "loss": 0.63796556,
      "num_input_tokens_seen": 29912515,
      "step": 1417,
      "time_per_iteration": 3.1229758262634277
    },
    {
      "auxiliary_loss_clip": 0.01111701,
      "auxiliary_loss_mlp": 0.0100654,
      "balance_loss_clip": 1.03336358,
      "balance_loss_mlp": 1.00402486,
      "epoch": 0.170504418926231,
      "flos": 49370637576960.0,
      "grad_norm": 0.8769125072509799,
      "language_loss": 0.62468582,
      "learning_rate": 3.7965928549137854e-06,
      "loss": 0.64586824,
      "num_input_tokens_seen": 29969330,
      "step": 1418,
      "time_per_iteration": 3.0394320487976074
    },
    {
      "auxiliary_loss_clip": 0.01187142,
      "auxiliary_loss_mlp": 0.01040268,
      "balance_loss_clip": 1.05840611,
      "balance_loss_mlp": 1.02977133,
      "epoch": 0.17062466181687008,
      "flos": 25849219184640.0,
      "grad_norm": 2.2320632333552575,
      "language_loss": 0.77782857,
      "learning_rate": 3.7962504463381953e-06,
      "loss": 0.80010265,
      "num_input_tokens_seen": 29990820,
      "step": 1419,
      "time_per_iteration": 2.663985013961792
    },
    {
      "auxiliary_loss_clip": 0.01189577,
      "auxiliary_loss_mlp": 0.00764498,
      "balance_loss_clip": 1.06431603,
      "balance_loss_mlp": 1.0000093,
      "epoch": 0.17074490470750917,
      "flos": 20960412549120.0,
      "grad_norm": 1.741666687432193,
      "language_loss": 0.79008847,
      "learning_rate": 3.7959077652728412e-06,
      "loss": 0.80962926,
      "num_input_tokens_seen": 30009275,
      "step": 1420,
      "time_per_iteration": 3.289870023727417
    },
    {
      "auxiliary_loss_clip": 0.01199181,
      "auxiliary_loss_mlp": 0.01045866,
      "balance_loss_clip": 1.0653764,
      "balance_loss_mlp": 1.03654361,
      "epoch": 0.17086514759814825,
      "flos": 20959766104320.0,
      "grad_norm": 2.0628873933921357,
      "language_loss": 0.77483153,
      "learning_rate": 3.795564811769707e-06,
      "loss": 0.79728198,
      "num_input_tokens_seen": 30027630,
      "step": 1421,
      "time_per_iteration": 2.5581743717193604
    },
    {
      "auxiliary_loss_clip": 0.01199855,
      "auxiliary_loss_mlp": 0.01044223,
      "balance_loss_clip": 1.06912422,
      "balance_loss_mlp": 1.03415573,
      "epoch": 0.17098539048878736,
      "flos": 28474073452800.0,
      "grad_norm": 1.9174463274387357,
      "language_loss": 0.77917755,
      "learning_rate": 3.795221585880818e-06,
      "loss": 0.80161828,
      "num_input_tokens_seen": 30048310,
      "step": 1422,
      "time_per_iteration": 2.6282262802124023
    },
    {
      "auxiliary_loss_clip": 0.01190792,
      "auxiliary_loss_mlp": 0.01039924,
      "balance_loss_clip": 1.07154202,
      "balance_loss_mlp": 1.03037548,
      "epoch": 0.17110563337942644,
      "flos": 16290014561280.0,
      "grad_norm": 2.031471953422813,
      "language_loss": 0.9127214,
      "learning_rate": 3.794878087658242e-06,
      "loss": 0.93502855,
      "num_input_tokens_seen": 30066080,
      "step": 1423,
      "time_per_iteration": 2.572491407394409
    },
    {
      "auxiliary_loss_clip": 0.01215496,
      "auxiliary_loss_mlp": 0.01043525,
      "balance_loss_clip": 1.0668931,
      "balance_loss_mlp": 1.03419054,
      "epoch": 0.17122587627006552,
      "flos": 29674207693440.0,
      "grad_norm": 2.01758154185329,
      "language_loss": 0.78472501,
      "learning_rate": 3.7945343171540873e-06,
      "loss": 0.80731523,
      "num_input_tokens_seen": 30086955,
      "step": 1424,
      "time_per_iteration": 3.3473005294799805
    },
    {
      "auxiliary_loss_clip": 0.01231829,
      "auxiliary_loss_mlp": 0.01046855,
      "balance_loss_clip": 1.07065499,
      "balance_loss_mlp": 1.03688312,
      "epoch": 0.17134611916070464,
      "flos": 25338389915520.0,
      "grad_norm": 5.495954222542481,
      "language_loss": 0.78793347,
      "learning_rate": 3.7941902744205033e-06,
      "loss": 0.81072032,
      "num_input_tokens_seen": 30107990,
      "step": 1425,
      "time_per_iteration": 2.5422165393829346
    },
    {
      "auxiliary_loss_clip": 0.01202271,
      "auxiliary_loss_mlp": 0.0103683,
      "balance_loss_clip": 1.06429887,
      "balance_loss_mlp": 1.02651787,
      "epoch": 0.17146636205134372,
      "flos": 13953845900160.0,
      "grad_norm": 2.1467656665767842,
      "language_loss": 0.83779061,
      "learning_rate": 3.7938459595096817e-06,
      "loss": 0.86018163,
      "num_input_tokens_seen": 30126535,
      "step": 1426,
      "time_per_iteration": 3.271543264389038
    },
    {
      "auxiliary_loss_clip": 0.0121914,
      "auxiliary_loss_mlp": 0.0104237,
      "balance_loss_clip": 1.06786442,
      "balance_loss_mlp": 1.03215337,
      "epoch": 0.1715866049419828,
      "flos": 23915214172800.0,
      "grad_norm": 1.9110132720155548,
      "language_loss": 0.8619169,
      "learning_rate": 3.7935013724738545e-06,
      "loss": 0.88453197,
      "num_input_tokens_seen": 30147035,
      "step": 1427,
      "time_per_iteration": 3.394890546798706
    },
    {
      "auxiliary_loss_clip": 0.0120722,
      "auxiliary_loss_mlp": 0.01046881,
      "balance_loss_clip": 1.06592834,
      "balance_loss_mlp": 1.03756452,
      "epoch": 0.17170684783262188,
      "flos": 22709369669760.0,
      "grad_norm": 1.9816725958863315,
      "language_loss": 0.77821112,
      "learning_rate": 3.7931565133652945e-06,
      "loss": 0.8007521,
      "num_input_tokens_seen": 30167110,
      "step": 1428,
      "time_per_iteration": 2.5568606853485107
    },
    {
      "auxiliary_loss_clip": 0.01227504,
      "auxiliary_loss_mlp": 0.01050115,
      "balance_loss_clip": 1.06755722,
      "balance_loss_mlp": 1.03978539,
      "epoch": 0.171827090723261,
      "flos": 26613290315520.0,
      "grad_norm": 2.3128813217535398,
      "language_loss": 0.67811757,
      "learning_rate": 3.792811382236317e-06,
      "loss": 0.70089376,
      "num_input_tokens_seen": 30185620,
      "step": 1429,
      "time_per_iteration": 2.5200276374816895
    },
    {
      "auxiliary_loss_clip": 0.01215217,
      "auxiliary_loss_mlp": 0.0103827,
      "balance_loss_clip": 1.06557047,
      "balance_loss_mlp": 1.02783871,
      "epoch": 0.17194733361390008,
      "flos": 28148507556480.0,
      "grad_norm": 1.9478584090144946,
      "language_loss": 0.77962387,
      "learning_rate": 3.792465979139279e-06,
      "loss": 0.80215871,
      "num_input_tokens_seen": 30208225,
      "step": 1430,
      "time_per_iteration": 2.602949619293213
    },
    {
      "auxiliary_loss_clip": 0.01078014,
      "auxiliary_loss_mlp": 0.01044778,
      "balance_loss_clip": 1.02725673,
      "balance_loss_mlp": 1.04239368,
      "epoch": 0.17206757650453916,
      "flos": 65530689753600.0,
      "grad_norm": 0.9417768365819899,
      "language_loss": 0.6562466,
      "learning_rate": 3.792120304126576e-06,
      "loss": 0.67747456,
      "num_input_tokens_seen": 30271600,
      "step": 1431,
      "time_per_iteration": 3.192549467086792
    },
    {
      "auxiliary_loss_clip": 0.01138589,
      "auxiliary_loss_mlp": 0.01041514,
      "balance_loss_clip": 1.05443323,
      "balance_loss_mlp": 1.03136325,
      "epoch": 0.17218781939517827,
      "flos": 22273486128000.0,
      "grad_norm": 2.223884891384171,
      "language_loss": 0.83717597,
      "learning_rate": 3.791774357250649e-06,
      "loss": 0.85897702,
      "num_input_tokens_seen": 30290430,
      "step": 1432,
      "time_per_iteration": 2.700982093811035
    },
    {
      "auxiliary_loss_clip": 0.01196174,
      "auxiliary_loss_mlp": 0.01039216,
      "balance_loss_clip": 1.06511688,
      "balance_loss_mlp": 1.02865398,
      "epoch": 0.17230806228581735,
      "flos": 14137313592960.0,
      "grad_norm": 2.272785574779725,
      "language_loss": 0.7957477,
      "learning_rate": 3.7914281385639757e-06,
      "loss": 0.81810164,
      "num_input_tokens_seen": 30308305,
      "step": 1433,
      "time_per_iteration": 2.5358924865722656
    },
    {
      "auxiliary_loss_clip": 0.01210269,
      "auxiliary_loss_mlp": 0.01045633,
      "balance_loss_clip": 1.06204939,
      "balance_loss_mlp": 1.0359416,
      "epoch": 0.17242830517645644,
      "flos": 20704836303360.0,
      "grad_norm": 1.8628525507739992,
      "language_loss": 0.79627657,
      "learning_rate": 3.7910816481190784e-06,
      "loss": 0.81883556,
      "num_input_tokens_seen": 30328120,
      "step": 1434,
      "time_per_iteration": 2.542018413543701
    },
    {
      "auxiliary_loss_clip": 0.01185128,
      "auxiliary_loss_mlp": 0.01040709,
      "balance_loss_clip": 1.0599215,
      "balance_loss_mlp": 1.03061843,
      "epoch": 0.17254854806709552,
      "flos": 30774582887040.0,
      "grad_norm": 1.9639786658371834,
      "language_loss": 0.74936414,
      "learning_rate": 3.7907348859685193e-06,
      "loss": 0.77162248,
      "num_input_tokens_seen": 30349825,
      "step": 1435,
      "time_per_iteration": 2.6402735710144043
    },
    {
      "auxiliary_loss_clip": 0.01207436,
      "auxiliary_loss_mlp": 0.01038049,
      "balance_loss_clip": 1.06684279,
      "balance_loss_mlp": 1.02894175,
      "epoch": 0.17266879095773463,
      "flos": 26614726859520.0,
      "grad_norm": 2.0482864832365997,
      "language_loss": 0.80540937,
      "learning_rate": 3.790387852164902e-06,
      "loss": 0.82786429,
      "num_input_tokens_seen": 30370555,
      "step": 1436,
      "time_per_iteration": 2.5719830989837646
    },
    {
      "auxiliary_loss_clip": 0.01213251,
      "auxiliary_loss_mlp": 0.01038212,
      "balance_loss_clip": 1.06556153,
      "balance_loss_mlp": 1.02851439,
      "epoch": 0.1727890338483737,
      "flos": 20266295155200.0,
      "grad_norm": 1.7806796677299148,
      "language_loss": 0.76219273,
      "learning_rate": 3.7900405467608707e-06,
      "loss": 0.78470737,
      "num_input_tokens_seen": 30390100,
      "step": 1437,
      "time_per_iteration": 2.530714511871338
    },
    {
      "auxiliary_loss_clip": 0.01148633,
      "auxiliary_loss_mlp": 0.01051886,
      "balance_loss_clip": 1.05218899,
      "balance_loss_mlp": 1.04200363,
      "epoch": 0.1729092767390128,
      "flos": 18179812909440.0,
      "grad_norm": 5.613838741454218,
      "language_loss": 0.79602957,
      "learning_rate": 3.7896929698091114e-06,
      "loss": 0.81803477,
      "num_input_tokens_seen": 30402915,
      "step": 1438,
      "time_per_iteration": 2.569930076599121
    },
    {
      "auxiliary_loss_clip": 0.01231562,
      "auxiliary_loss_mlp": 0.01052747,
      "balance_loss_clip": 1.07121062,
      "balance_loss_mlp": 1.04247141,
      "epoch": 0.1730295196296519,
      "flos": 26759518583040.0,
      "grad_norm": 2.488771281509292,
      "language_loss": 0.67967069,
      "learning_rate": 3.7893451213623518e-06,
      "loss": 0.70251375,
      "num_input_tokens_seen": 30420145,
      "step": 1439,
      "time_per_iteration": 2.5314810276031494
    },
    {
      "auxiliary_loss_clip": 0.0121026,
      "auxiliary_loss_mlp": 0.00763969,
      "balance_loss_clip": 1.06560063,
      "balance_loss_mlp": 1.0000422,
      "epoch": 0.173149762520291,
      "flos": 23842531002240.0,
      "grad_norm": 1.9683334380393174,
      "language_loss": 0.82259959,
      "learning_rate": 3.7889970014733606e-06,
      "loss": 0.84234184,
      "num_input_tokens_seen": 30439250,
      "step": 1440,
      "time_per_iteration": 2.5377919673919678
    },
    {
      "auxiliary_loss_clip": 0.01148116,
      "auxiliary_loss_mlp": 0.010392,
      "balance_loss_clip": 1.05430257,
      "balance_loss_mlp": 1.02907908,
      "epoch": 0.17327000541093007,
      "flos": 23368186972800.0,
      "grad_norm": 1.630740678661122,
      "language_loss": 0.77576542,
      "learning_rate": 3.7886486101949463e-06,
      "loss": 0.79763854,
      "num_input_tokens_seen": 30460430,
      "step": 1441,
      "time_per_iteration": 2.664477825164795
    },
    {
      "auxiliary_loss_clip": 0.01151081,
      "auxiliary_loss_mlp": 0.01040892,
      "balance_loss_clip": 1.05370426,
      "balance_loss_mlp": 1.03158832,
      "epoch": 0.17339024830156918,
      "flos": 18221290139520.0,
      "grad_norm": 1.99201395455247,
      "language_loss": 0.88168782,
      "learning_rate": 3.7882999475799594e-06,
      "loss": 0.90360755,
      "num_input_tokens_seen": 30478465,
      "step": 1442,
      "time_per_iteration": 2.6065685749053955
    },
    {
      "auxiliary_loss_clip": 0.01150664,
      "auxiliary_loss_mlp": 0.01041143,
      "balance_loss_clip": 1.05876219,
      "balance_loss_mlp": 1.03168416,
      "epoch": 0.17351049119220827,
      "flos": 23332024955520.0,
      "grad_norm": 1.8688816848506866,
      "language_loss": 0.81599987,
      "learning_rate": 3.787951013681293e-06,
      "loss": 0.83791792,
      "num_input_tokens_seen": 30496510,
      "step": 1443,
      "time_per_iteration": 2.6398072242736816
    },
    {
      "auxiliary_loss_clip": 0.01213628,
      "auxiliary_loss_mlp": 0.01047505,
      "balance_loss_clip": 1.06558347,
      "balance_loss_mlp": 1.03681207,
      "epoch": 0.17363073408284735,
      "flos": 23803495896960.0,
      "grad_norm": 2.2567568762751056,
      "language_loss": 0.77593368,
      "learning_rate": 3.787601808551879e-06,
      "loss": 0.798545,
      "num_input_tokens_seen": 30516325,
      "step": 1444,
      "time_per_iteration": 2.5401294231414795
    },
    {
      "auxiliary_loss_clip": 0.01184676,
      "auxiliary_loss_mlp": 0.01036502,
      "balance_loss_clip": 1.06264758,
      "balance_loss_mlp": 1.02643442,
      "epoch": 0.17375097697348643,
      "flos": 18515290959360.0,
      "grad_norm": 2.659630089582604,
      "language_loss": 0.83703017,
      "learning_rate": 3.7872523322446926e-06,
      "loss": 0.85924196,
      "num_input_tokens_seen": 30535210,
      "step": 1445,
      "time_per_iteration": 2.5807924270629883
    },
    {
      "auxiliary_loss_clip": 0.01168053,
      "auxiliary_loss_mlp": 0.01038327,
      "balance_loss_clip": 1.05512249,
      "balance_loss_mlp": 1.02874875,
      "epoch": 0.17387121986412554,
      "flos": 38877897456000.0,
      "grad_norm": 2.1873450348915333,
      "language_loss": 0.60415113,
      "learning_rate": 3.7869025848127478e-06,
      "loss": 0.62621498,
      "num_input_tokens_seen": 30559405,
      "step": 1446,
      "time_per_iteration": 3.58294677734375
    },
    {
      "auxiliary_loss_clip": 0.01211267,
      "auxiliary_loss_mlp": 0.01042263,
      "balance_loss_clip": 1.06326723,
      "balance_loss_mlp": 1.03261948,
      "epoch": 0.17399146275476463,
      "flos": 20375714960640.0,
      "grad_norm": 2.6802538720167814,
      "language_loss": 0.80866027,
      "learning_rate": 3.786552566309102e-06,
      "loss": 0.83119559,
      "num_input_tokens_seen": 30577615,
      "step": 1447,
      "time_per_iteration": 2.568075656890869
    },
    {
      "auxiliary_loss_clip": 0.0119242,
      "auxiliary_loss_mlp": 0.00763495,
      "balance_loss_clip": 1.06542492,
      "balance_loss_mlp": 1.00002778,
      "epoch": 0.1741117056454037,
      "flos": 19164339763200.0,
      "grad_norm": 2.3857299760790047,
      "language_loss": 0.86283922,
      "learning_rate": 3.7862022767868517e-06,
      "loss": 0.88239837,
      "num_input_tokens_seen": 30595205,
      "step": 1448,
      "time_per_iteration": 2.521491289138794
    },
    {
      "auxiliary_loss_clip": 0.01180519,
      "auxiliary_loss_mlp": 0.01034632,
      "balance_loss_clip": 1.06756842,
      "balance_loss_mlp": 1.02511358,
      "epoch": 0.17423194853604282,
      "flos": 25374300537600.0,
      "grad_norm": 2.017867728641486,
      "language_loss": 0.84379548,
      "learning_rate": 3.7858517162991367e-06,
      "loss": 0.86594701,
      "num_input_tokens_seen": 30615280,
      "step": 1449,
      "time_per_iteration": 2.6106793880462646
    },
    {
      "auxiliary_loss_clip": 0.01184088,
      "auxiliary_loss_mlp": 0.01050976,
      "balance_loss_clip": 1.06151223,
      "balance_loss_mlp": 1.04139709,
      "epoch": 0.1743521914266819,
      "flos": 25191874339200.0,
      "grad_norm": 2.5815007274916697,
      "language_loss": 0.60931665,
      "learning_rate": 3.7855008848991363e-06,
      "loss": 0.63166732,
      "num_input_tokens_seen": 30633485,
      "step": 1450,
      "time_per_iteration": 3.3537607192993164
    },
    {
      "auxiliary_loss_clip": 0.01196121,
      "auxiliary_loss_mlp": 0.01043746,
      "balance_loss_clip": 1.0669024,
      "balance_loss_mlp": 1.03482926,
      "epoch": 0.17447243431732098,
      "flos": 25666577504640.0,
      "grad_norm": 1.775737250947227,
      "language_loss": 0.77384877,
      "learning_rate": 3.7851497826400714e-06,
      "loss": 0.79624748,
      "num_input_tokens_seen": 30653625,
      "step": 1451,
      "time_per_iteration": 3.3626925945281982
    },
    {
      "auxiliary_loss_clip": 0.01227942,
      "auxiliary_loss_mlp": 0.0105996,
      "balance_loss_clip": 1.06911218,
      "balance_loss_mlp": 1.04846883,
      "epoch": 0.17459267720796007,
      "flos": 36281950657920.0,
      "grad_norm": 2.512445677498836,
      "language_loss": 0.76062685,
      "learning_rate": 3.7847984095752034e-06,
      "loss": 0.78350586,
      "num_input_tokens_seen": 30677080,
      "step": 1452,
      "time_per_iteration": 3.6212472915649414
    },
    {
      "auxiliary_loss_clip": 0.01224737,
      "auxiliary_loss_mlp": 0.01034226,
      "balance_loss_clip": 1.06602681,
      "balance_loss_mlp": 1.02526116,
      "epoch": 0.17471292009859918,
      "flos": 20011113959040.0,
      "grad_norm": 2.0480958816963857,
      "language_loss": 0.8033272,
      "learning_rate": 3.784446765757836e-06,
      "loss": 0.82591677,
      "num_input_tokens_seen": 30695725,
      "step": 1453,
      "time_per_iteration": 2.4838550090789795
    },
    {
      "auxiliary_loss_clip": 0.01164127,
      "auxiliary_loss_mlp": 0.01041891,
      "balance_loss_clip": 1.05817986,
      "balance_loss_mlp": 1.0311625,
      "epoch": 0.17483316298923826,
      "flos": 27819242559360.0,
      "grad_norm": 2.1671794210877744,
      "language_loss": 0.77732468,
      "learning_rate": 3.7840948512413133e-06,
      "loss": 0.79938483,
      "num_input_tokens_seen": 30713310,
      "step": 1454,
      "time_per_iteration": 2.629128932952881
    },
    {
      "auxiliary_loss_clip": 0.01176014,
      "auxiliary_loss_mlp": 0.01039123,
      "balance_loss_clip": 1.06337237,
      "balance_loss_mlp": 1.02888334,
      "epoch": 0.17495340587987734,
      "flos": 44017934791680.0,
      "grad_norm": 1.8242041563377056,
      "language_loss": 0.78671551,
      "learning_rate": 3.7837426660790196e-06,
      "loss": 0.80886686,
      "num_input_tokens_seen": 30734725,
      "step": 1455,
      "time_per_iteration": 2.7803046703338623
    },
    {
      "auxiliary_loss_clip": 0.01223971,
      "auxiliary_loss_mlp": 0.01032303,
      "balance_loss_clip": 1.0662154,
      "balance_loss_mlp": 1.02323723,
      "epoch": 0.17507364877051645,
      "flos": 20885825957760.0,
      "grad_norm": 2.506330515155809,
      "language_loss": 0.81767577,
      "learning_rate": 3.783390210324382e-06,
      "loss": 0.84023851,
      "num_input_tokens_seen": 30754450,
      "step": 1456,
      "time_per_iteration": 2.5097544193267822
    },
    {
      "auxiliary_loss_clip": 0.0117941,
      "auxiliary_loss_mlp": 0.01033982,
      "balance_loss_clip": 1.06398773,
      "balance_loss_mlp": 1.02508879,
      "epoch": 0.17519389166115554,
      "flos": 24717602136960.0,
      "grad_norm": 4.541607797428815,
      "language_loss": 0.72494256,
      "learning_rate": 3.7830374840308676e-06,
      "loss": 0.74707651,
      "num_input_tokens_seen": 30774605,
      "step": 1457,
      "time_per_iteration": 2.614821195602417
    },
    {
      "auxiliary_loss_clip": 0.01207825,
      "auxiliary_loss_mlp": 0.01036513,
      "balance_loss_clip": 1.06496644,
      "balance_loss_mlp": 1.02661264,
      "epoch": 0.17531413455179462,
      "flos": 23798144770560.0,
      "grad_norm": 2.774140771543342,
      "language_loss": 0.82623708,
      "learning_rate": 3.7826844872519842e-06,
      "loss": 0.84868038,
      "num_input_tokens_seen": 30792460,
      "step": 1458,
      "time_per_iteration": 2.5492103099823
    },
    {
      "auxiliary_loss_clip": 0.0119401,
      "auxiliary_loss_mlp": 0.01040569,
      "balance_loss_clip": 1.06554222,
      "balance_loss_mlp": 1.03137839,
      "epoch": 0.1754343774424337,
      "flos": 24572379450240.0,
      "grad_norm": 1.9781146423826275,
      "language_loss": 0.72576058,
      "learning_rate": 3.782331220041282e-06,
      "loss": 0.74810636,
      "num_input_tokens_seen": 30812525,
      "step": 1459,
      "time_per_iteration": 2.616131067276001
    },
    {
      "auxiliary_loss_clip": 0.01186352,
      "auxiliary_loss_mlp": 0.0103519,
      "balance_loss_clip": 1.06024098,
      "balance_loss_mlp": 1.02498531,
      "epoch": 0.17555462033307281,
      "flos": 18114599767680.0,
      "grad_norm": 2.2115048429092226,
      "language_loss": 0.82877642,
      "learning_rate": 3.7819776824523504e-06,
      "loss": 0.85099185,
      "num_input_tokens_seen": 30830390,
      "step": 1460,
      "time_per_iteration": 2.566915988922119
    },
    {
      "auxiliary_loss_clip": 0.01204441,
      "auxiliary_loss_mlp": 0.01043132,
      "balance_loss_clip": 1.06451309,
      "balance_loss_mlp": 1.03329134,
      "epoch": 0.1756748632237119,
      "flos": 28366018364160.0,
      "grad_norm": 2.020261316261261,
      "language_loss": 0.84018487,
      "learning_rate": 3.7816238745388213e-06,
      "loss": 0.86266065,
      "num_input_tokens_seen": 30849935,
      "step": 1461,
      "time_per_iteration": 2.6337151527404785
    },
    {
      "auxiliary_loss_clip": 0.01201901,
      "auxiliary_loss_mlp": 0.01032436,
      "balance_loss_clip": 1.06285667,
      "balance_loss_mlp": 1.02323329,
      "epoch": 0.17579510611435098,
      "flos": 25732939881600.0,
      "grad_norm": 1.8890123011359579,
      "language_loss": 0.86768591,
      "learning_rate": 3.781269796354367e-06,
      "loss": 0.89002919,
      "num_input_tokens_seen": 30869555,
      "step": 1462,
      "time_per_iteration": 2.590184211730957
    },
    {
      "auxiliary_loss_clip": 0.01200371,
      "auxiliary_loss_mlp": 0.01039979,
      "balance_loss_clip": 1.06680512,
      "balance_loss_mlp": 1.03076422,
      "epoch": 0.1759153490049901,
      "flos": 18588081870720.0,
      "grad_norm": 1.6644148923140722,
      "language_loss": 0.86085731,
      "learning_rate": 3.7809154479527006e-06,
      "loss": 0.88326085,
      "num_input_tokens_seen": 30888760,
      "step": 1463,
      "time_per_iteration": 2.5499353408813477
    },
    {
      "auxiliary_loss_clip": 0.01175723,
      "auxiliary_loss_mlp": 0.01040279,
      "balance_loss_clip": 1.06283247,
      "balance_loss_mlp": 1.03052735,
      "epoch": 0.17603559189562917,
      "flos": 18619323724800.0,
      "grad_norm": 2.1238055719626483,
      "language_loss": 0.84270471,
      "learning_rate": 3.780560829387577e-06,
      "loss": 0.86486471,
      "num_input_tokens_seen": 30907260,
      "step": 1464,
      "time_per_iteration": 2.562060832977295
    },
    {
      "auxiliary_loss_clip": 0.01103392,
      "auxiliary_loss_mlp": 0.01011353,
      "balance_loss_clip": 1.02628541,
      "balance_loss_mlp": 1.00893331,
      "epoch": 0.17615583478626826,
      "flos": 60530775373440.0,
      "grad_norm": 0.8618925605951729,
      "language_loss": 0.57929695,
      "learning_rate": 3.7802059407127915e-06,
      "loss": 0.60044432,
      "num_input_tokens_seen": 30965810,
      "step": 1465,
      "time_per_iteration": 3.153050184249878
    },
    {
      "auxiliary_loss_clip": 0.01187407,
      "auxiliary_loss_mlp": 0.01036181,
      "balance_loss_clip": 1.05984592,
      "balance_loss_mlp": 1.02588749,
      "epoch": 0.17627607767690734,
      "flos": 23616221362560.0,
      "grad_norm": 2.157152820883273,
      "language_loss": 0.86169446,
      "learning_rate": 3.7798507819821797e-06,
      "loss": 0.88393039,
      "num_input_tokens_seen": 30982935,
      "step": 1466,
      "time_per_iteration": 2.5706067085266113
    },
    {
      "auxiliary_loss_clip": 0.01172193,
      "auxiliary_loss_mlp": 0.01042413,
      "balance_loss_clip": 1.06073713,
      "balance_loss_mlp": 1.03213763,
      "epoch": 0.17639632056754645,
      "flos": 17639070589440.0,
      "grad_norm": 2.341536771038795,
      "language_loss": 0.78631604,
      "learning_rate": 3.7794953532496197e-06,
      "loss": 0.80846208,
      "num_input_tokens_seen": 30998840,
      "step": 1467,
      "time_per_iteration": 2.5720338821411133
    },
    {
      "auxiliary_loss_clip": 0.01064566,
      "auxiliary_loss_mlp": 0.00753602,
      "balance_loss_clip": 1.02635026,
      "balance_loss_mlp": 1.00001144,
      "epoch": 0.17651656345818553,
      "flos": 57932604910080.0,
      "grad_norm": 0.9194190154793992,
      "language_loss": 0.57952601,
      "learning_rate": 3.7791396545690295e-06,
      "loss": 0.59770769,
      "num_input_tokens_seen": 31060075,
      "step": 1468,
      "time_per_iteration": 3.215304374694824
    },
    {
      "auxiliary_loss_clip": 0.0121417,
      "auxiliary_loss_mlp": 0.01036199,
      "balance_loss_clip": 1.06852186,
      "balance_loss_mlp": 1.02654862,
      "epoch": 0.17663680634882462,
      "flos": 22929502170240.0,
      "grad_norm": 3.4612559885687415,
      "language_loss": 0.80816615,
      "learning_rate": 3.7787836859943685e-06,
      "loss": 0.83066988,
      "num_input_tokens_seen": 31078800,
      "step": 1469,
      "time_per_iteration": 2.550433397293091
    },
    {
      "auxiliary_loss_clip": 0.01209267,
      "auxiliary_loss_mlp": 0.01049246,
      "balance_loss_clip": 1.06373811,
      "balance_loss_mlp": 1.03942275,
      "epoch": 0.17675704923946373,
      "flos": 22637979388800.0,
      "grad_norm": 3.9915065294864065,
      "language_loss": 0.79063171,
      "learning_rate": 3.7784274475796363e-06,
      "loss": 0.81321687,
      "num_input_tokens_seen": 31097430,
      "step": 1470,
      "time_per_iteration": 2.540336847305298
    },
    {
      "auxiliary_loss_clip": 0.01177436,
      "auxiliary_loss_mlp": 0.01033518,
      "balance_loss_clip": 1.05815172,
      "balance_loss_mlp": 1.02304602,
      "epoch": 0.1768772921301028,
      "flos": 27126525795840.0,
      "grad_norm": 2.990498070170467,
      "language_loss": 0.76226056,
      "learning_rate": 3.7780709393788745e-06,
      "loss": 0.78437012,
      "num_input_tokens_seen": 31117905,
      "step": 1471,
      "time_per_iteration": 2.6271116733551025
    },
    {
      "auxiliary_loss_clip": 0.01222779,
      "auxiliary_loss_mlp": 0.01043078,
      "balance_loss_clip": 1.06526446,
      "balance_loss_mlp": 1.03414881,
      "epoch": 0.1769975350207419,
      "flos": 19172133014400.0,
      "grad_norm": 2.1725868379031446,
      "language_loss": 0.75394273,
      "learning_rate": 3.777714161446165e-06,
      "loss": 0.77660131,
      "num_input_tokens_seen": 31137610,
      "step": 1472,
      "time_per_iteration": 3.3371667861938477
    },
    {
      "auxiliary_loss_clip": 0.01212767,
      "auxiliary_loss_mlp": 0.01041326,
      "balance_loss_clip": 1.06754482,
      "balance_loss_mlp": 1.03155088,
      "epoch": 0.177117777911381,
      "flos": 36134932291200.0,
      "grad_norm": 2.115346935327689,
      "language_loss": 0.69375223,
      "learning_rate": 3.7773571138356304e-06,
      "loss": 0.71629322,
      "num_input_tokens_seen": 31157780,
      "step": 1473,
      "time_per_iteration": 2.6712160110473633
    },
    {
      "auxiliary_loss_clip": 0.01152481,
      "auxiliary_loss_mlp": 0.01041919,
      "balance_loss_clip": 1.05960047,
      "balance_loss_mlp": 1.03297806,
      "epoch": 0.17723802080202009,
      "flos": 22090593052800.0,
      "grad_norm": 2.2057653189350854,
      "language_loss": 0.88907099,
      "learning_rate": 3.776999796601435e-06,
      "loss": 0.91101503,
      "num_input_tokens_seen": 31176540,
      "step": 1474,
      "time_per_iteration": 2.626495599746704
    },
    {
      "auxiliary_loss_clip": 0.01214531,
      "auxiliary_loss_mlp": 0.01038482,
      "balance_loss_clip": 1.06537962,
      "balance_loss_mlp": 1.02868295,
      "epoch": 0.17735826369265917,
      "flos": 30222671437440.0,
      "grad_norm": 1.89763792791142,
      "language_loss": 0.73025763,
      "learning_rate": 3.776642209797783e-06,
      "loss": 0.75278771,
      "num_input_tokens_seen": 31198370,
      "step": 1475,
      "time_per_iteration": 2.607489824295044
    },
    {
      "auxiliary_loss_clip": 0.01201611,
      "auxiliary_loss_mlp": 0.01053952,
      "balance_loss_clip": 1.06168199,
      "balance_loss_mlp": 1.04440284,
      "epoch": 0.17747850658329825,
      "flos": 21397588980480.0,
      "grad_norm": 2.693476577264495,
      "language_loss": 0.77957904,
      "learning_rate": 3.7762843534789205e-06,
      "loss": 0.80213463,
      "num_input_tokens_seen": 31217120,
      "step": 1476,
      "time_per_iteration": 3.2833545207977295
    },
    {
      "auxiliary_loss_clip": 0.01197829,
      "auxiliary_loss_mlp": 0.01039851,
      "balance_loss_clip": 1.0608567,
      "balance_loss_mlp": 1.03008747,
      "epoch": 0.17759874947393736,
      "flos": 16983341856000.0,
      "grad_norm": 2.2303346811322275,
      "language_loss": 0.87966663,
      "learning_rate": 3.7759262276991343e-06,
      "loss": 0.90204346,
      "num_input_tokens_seen": 31234730,
      "step": 1477,
      "time_per_iteration": 3.336195945739746
    },
    {
      "auxiliary_loss_clip": 0.01197545,
      "auxiliary_loss_mlp": 0.01042816,
      "balance_loss_clip": 1.06301951,
      "balance_loss_mlp": 1.03257012,
      "epoch": 0.17771899236457644,
      "flos": 11546107390080.0,
      "grad_norm": 2.166519149451295,
      "language_loss": 0.80644631,
      "learning_rate": 3.7755678325127506e-06,
      "loss": 0.82884991,
      "num_input_tokens_seen": 31252410,
      "step": 1478,
      "time_per_iteration": 2.52659010887146
    },
    {
      "auxiliary_loss_clip": 0.01158322,
      "auxiliary_loss_mlp": 0.01045372,
      "balance_loss_clip": 1.05849004,
      "balance_loss_mlp": 1.03544235,
      "epoch": 0.17783923525521553,
      "flos": 18807747494400.0,
      "grad_norm": 1.8503159986133724,
      "language_loss": 0.75788003,
      "learning_rate": 3.7752091679741393e-06,
      "loss": 0.779917,
      "num_input_tokens_seen": 31270200,
      "step": 1479,
      "time_per_iteration": 3.4348692893981934
    },
    {
      "auxiliary_loss_clip": 0.01207998,
      "auxiliary_loss_mlp": 0.01037485,
      "balance_loss_clip": 1.06570053,
      "balance_loss_mlp": 1.02799606,
      "epoch": 0.17795947814585464,
      "flos": 30408365773440.0,
      "grad_norm": 3.856463005557222,
      "language_loss": 0.77929425,
      "learning_rate": 3.774850234137708e-06,
      "loss": 0.80174911,
      "num_input_tokens_seen": 31287495,
      "step": 1480,
      "time_per_iteration": 2.57098126411438
    },
    {
      "auxiliary_loss_clip": 0.01205831,
      "auxiliary_loss_mlp": 0.01041965,
      "balance_loss_clip": 1.06184053,
      "balance_loss_mlp": 1.032094,
      "epoch": 0.17807972103649372,
      "flos": 24389055411840.0,
      "grad_norm": 3.3822472585350902,
      "language_loss": 0.83195019,
      "learning_rate": 3.7744910310579076e-06,
      "loss": 0.85442823,
      "num_input_tokens_seen": 31306420,
      "step": 1481,
      "time_per_iteration": 2.56130051612854
    },
    {
      "auxiliary_loss_clip": 0.01221547,
      "auxiliary_loss_mlp": 0.01043092,
      "balance_loss_clip": 1.06543684,
      "balance_loss_mlp": 1.03344762,
      "epoch": 0.1781999639271328,
      "flos": 20301559332480.0,
      "grad_norm": 1.8085056243548028,
      "language_loss": 0.84915358,
      "learning_rate": 3.774131558789229e-06,
      "loss": 0.87179995,
      "num_input_tokens_seen": 31325750,
      "step": 1482,
      "time_per_iteration": 2.483839988708496
    },
    {
      "auxiliary_loss_clip": 0.01223994,
      "auxiliary_loss_mlp": 0.00762632,
      "balance_loss_clip": 1.06625187,
      "balance_loss_mlp": 0.99996281,
      "epoch": 0.1783202068177719,
      "flos": 15924479806080.0,
      "grad_norm": 2.4074267337522506,
      "language_loss": 0.69889843,
      "learning_rate": 3.773771817386203e-06,
      "loss": 0.71876478,
      "num_input_tokens_seen": 31343080,
      "step": 1483,
      "time_per_iteration": 2.5520598888397217
    },
    {
      "auxiliary_loss_clip": 0.01190213,
      "auxiliary_loss_mlp": 0.0103421,
      "balance_loss_clip": 1.06000447,
      "balance_loss_mlp": 1.02501297,
      "epoch": 0.178440449708411,
      "flos": 20631758083200.0,
      "grad_norm": 1.6288762733605522,
      "language_loss": 0.79136693,
      "learning_rate": 3.773411806903403e-06,
      "loss": 0.81361115,
      "num_input_tokens_seen": 31362160,
      "step": 1484,
      "time_per_iteration": 2.6342556476593018
    },
    {
      "auxiliary_loss_clip": 0.01149716,
      "auxiliary_loss_mlp": 0.01043557,
      "balance_loss_clip": 1.05514646,
      "balance_loss_mlp": 1.03369856,
      "epoch": 0.17856069259905008,
      "flos": 21686059105920.0,
      "grad_norm": 2.0195564743043395,
      "language_loss": 0.9458971,
      "learning_rate": 3.7730515273954415e-06,
      "loss": 0.96782982,
      "num_input_tokens_seen": 31380770,
      "step": 1485,
      "time_per_iteration": 2.645796298980713
    },
    {
      "auxiliary_loss_clip": 0.01224266,
      "auxiliary_loss_mlp": 0.01033903,
      "balance_loss_clip": 1.067029,
      "balance_loss_mlp": 1.02380037,
      "epoch": 0.17868093548968916,
      "flos": 26572962320640.0,
      "grad_norm": 1.8074744471311217,
      "language_loss": 0.84915018,
      "learning_rate": 3.772690978916973e-06,
      "loss": 0.87173188,
      "num_input_tokens_seen": 31400525,
      "step": 1486,
      "time_per_iteration": 2.535668134689331
    },
    {
      "auxiliary_loss_clip": 0.01210346,
      "auxiliary_loss_mlp": 0.01040013,
      "balance_loss_clip": 1.06784523,
      "balance_loss_mlp": 1.02982628,
      "epoch": 0.17880117838032827,
      "flos": 18581006891520.0,
      "grad_norm": 2.231442128606665,
      "language_loss": 0.868469,
      "learning_rate": 3.772330161522693e-06,
      "loss": 0.89097255,
      "num_input_tokens_seen": 31418435,
      "step": 1487,
      "time_per_iteration": 2.517756938934326
    },
    {
      "auxiliary_loss_clip": 0.01191931,
      "auxiliary_loss_mlp": 0.01037909,
      "balance_loss_clip": 1.0664463,
      "balance_loss_mlp": 1.02858138,
      "epoch": 0.17892142127096736,
      "flos": 26541217676160.0,
      "grad_norm": 2.1712225305019013,
      "language_loss": 0.79618561,
      "learning_rate": 3.7719690752673365e-06,
      "loss": 0.81848401,
      "num_input_tokens_seen": 31439230,
      "step": 1488,
      "time_per_iteration": 2.573653221130371
    },
    {
      "auxiliary_loss_clip": 0.0118418,
      "auxiliary_loss_mlp": 0.01049426,
      "balance_loss_clip": 1.06431782,
      "balance_loss_mlp": 1.03981161,
      "epoch": 0.17904166416160644,
      "flos": 23872623621120.0,
      "grad_norm": 1.9099062603064685,
      "language_loss": 0.78287435,
      "learning_rate": 3.7716077202056796e-06,
      "loss": 0.80521035,
      "num_input_tokens_seen": 31457705,
      "step": 1489,
      "time_per_iteration": 2.605098009109497
    },
    {
      "auxiliary_loss_clip": 0.01179062,
      "auxiliary_loss_mlp": 0.01047408,
      "balance_loss_clip": 1.06022799,
      "balance_loss_mlp": 1.03654206,
      "epoch": 0.17916190705224552,
      "flos": 19134426712320.0,
      "grad_norm": 2.1607969766930877,
      "language_loss": 0.93610096,
      "learning_rate": 3.7712460963925404e-06,
      "loss": 0.95836568,
      "num_input_tokens_seen": 31473645,
      "step": 1490,
      "time_per_iteration": 2.5133471488952637
    },
    {
      "auxiliary_loss_clip": 0.01185472,
      "auxiliary_loss_mlp": 0.01035773,
      "balance_loss_clip": 1.06016922,
      "balance_loss_mlp": 1.02681994,
      "epoch": 0.17928214994288463,
      "flos": 25152120961920.0,
      "grad_norm": 1.8168001452556795,
      "language_loss": 0.7536341,
      "learning_rate": 3.7708842038827775e-06,
      "loss": 0.7758466,
      "num_input_tokens_seen": 31492605,
      "step": 1491,
      "time_per_iteration": 2.5829224586486816
    },
    {
      "auxiliary_loss_clip": 0.01207396,
      "auxiliary_loss_mlp": 0.01050889,
      "balance_loss_clip": 1.06069052,
      "balance_loss_mlp": 1.04129839,
      "epoch": 0.17940239283352372,
      "flos": 22384629786240.0,
      "grad_norm": 1.7985117972124915,
      "language_loss": 0.85829532,
      "learning_rate": 3.770522042731288e-06,
      "loss": 0.88087809,
      "num_input_tokens_seen": 31514500,
      "step": 1492,
      "time_per_iteration": 2.554351329803467
    },
    {
      "auxiliary_loss_clip": 0.01158754,
      "auxiliary_loss_mlp": 0.01044088,
      "balance_loss_clip": 1.06052935,
      "balance_loss_mlp": 1.03391981,
      "epoch": 0.1795226357241628,
      "flos": 23178685795200.0,
      "grad_norm": 1.930896392562346,
      "language_loss": 0.87572682,
      "learning_rate": 3.7701596129930122e-06,
      "loss": 0.89775521,
      "num_input_tokens_seen": 31533225,
      "step": 1493,
      "time_per_iteration": 2.6225645542144775
    },
    {
      "auxiliary_loss_clip": 0.01188602,
      "auxiliary_loss_mlp": 0.01029268,
      "balance_loss_clip": 1.06381631,
      "balance_loss_mlp": 1.01934385,
      "epoch": 0.1796428786148019,
      "flos": 22090413484800.0,
      "grad_norm": 4.144401204839204,
      "language_loss": 0.73243254,
      "learning_rate": 3.7697969147229315e-06,
      "loss": 0.75461125,
      "num_input_tokens_seen": 31551385,
      "step": 1494,
      "time_per_iteration": 2.599379062652588
    },
    {
      "auxiliary_loss_clip": 0.01209953,
      "auxiliary_loss_mlp": 0.01043239,
      "balance_loss_clip": 1.06490481,
      "balance_loss_mlp": 1.03379166,
      "epoch": 0.179763121505441,
      "flos": 21324618501120.0,
      "grad_norm": 1.969454641914461,
      "language_loss": 0.85120654,
      "learning_rate": 3.7694339479760647e-06,
      "loss": 0.87373853,
      "num_input_tokens_seen": 31570415,
      "step": 1495,
      "time_per_iteration": 2.514608144760132
    },
    {
      "auxiliary_loss_clip": 0.0108858,
      "auxiliary_loss_mlp": 0.01003784,
      "balance_loss_clip": 1.02506685,
      "balance_loss_mlp": 1.00150728,
      "epoch": 0.17988336439608008,
      "flos": 68161864815360.0,
      "grad_norm": 0.7716434244681164,
      "language_loss": 0.57329476,
      "learning_rate": 3.769070712807476e-06,
      "loss": 0.59421837,
      "num_input_tokens_seen": 31632445,
      "step": 1496,
      "time_per_iteration": 3.210659980773926
    },
    {
      "auxiliary_loss_clip": 0.0113849,
      "auxiliary_loss_mlp": 0.01035832,
      "balance_loss_clip": 1.0561254,
      "balance_loss_mlp": 1.0256933,
      "epoch": 0.18000360728671919,
      "flos": 21945047143680.0,
      "grad_norm": 8.409180134690091,
      "language_loss": 0.7832309,
      "learning_rate": 3.768707209272266e-06,
      "loss": 0.80497408,
      "num_input_tokens_seen": 31652575,
      "step": 1497,
      "time_per_iteration": 2.6351358890533447
    },
    {
      "auxiliary_loss_clip": 0.01192077,
      "auxiliary_loss_mlp": 0.01043779,
      "balance_loss_clip": 1.06321609,
      "balance_loss_mlp": 1.03396845,
      "epoch": 0.18012385017735827,
      "flos": 18986330937600.0,
      "grad_norm": 2.257416628677278,
      "language_loss": 0.76798677,
      "learning_rate": 3.768343437425579e-06,
      "loss": 0.79034531,
      "num_input_tokens_seen": 31671145,
      "step": 1498,
      "time_per_iteration": 3.2797653675079346
    },
    {
      "auxiliary_loss_clip": 0.01122288,
      "auxiliary_loss_mlp": 0.01036653,
      "balance_loss_clip": 1.05213952,
      "balance_loss_mlp": 1.02694941,
      "epoch": 0.18024409306799735,
      "flos": 19748103598080.0,
      "grad_norm": 2.976526158329585,
      "language_loss": 0.86206388,
      "learning_rate": 3.7679793973225987e-06,
      "loss": 0.88365334,
      "num_input_tokens_seen": 31686955,
      "step": 1499,
      "time_per_iteration": 2.7085530757904053
    },
    {
      "auxiliary_loss_clip": 0.01059131,
      "auxiliary_loss_mlp": 0.01007084,
      "balance_loss_clip": 1.02096677,
      "balance_loss_mlp": 1.004879,
      "epoch": 0.18036433595863643,
      "flos": 67227183060480.0,
      "grad_norm": 0.8536866501646673,
      "language_loss": 0.61661541,
      "learning_rate": 3.767615089018549e-06,
      "loss": 0.6372776,
      "num_input_tokens_seen": 31749300,
      "step": 1500,
      "time_per_iteration": 3.2229766845703125
    },
    {
      "auxiliary_loss_clip": 0.01187858,
      "auxiliary_loss_mlp": 0.01038291,
      "balance_loss_clip": 1.06063616,
      "balance_loss_mlp": 1.02766371,
      "epoch": 0.18048457884927555,
      "flos": 18181464935040.0,
      "grad_norm": 2.0250824065102058,
      "language_loss": 0.86185533,
      "learning_rate": 3.7672505125686966e-06,
      "loss": 0.88411683,
      "num_input_tokens_seen": 31765665,
      "step": 1501,
      "time_per_iteration": 2.569427251815796
    },
    {
      "auxiliary_loss_clip": 0.01163784,
      "auxiliary_loss_mlp": 0.01041233,
      "balance_loss_clip": 1.05891323,
      "balance_loss_mlp": 1.03064132,
      "epoch": 0.18060482173991463,
      "flos": 15813767111040.0,
      "grad_norm": 3.9946685672686373,
      "language_loss": 0.84002924,
      "learning_rate": 3.7668856680283455e-06,
      "loss": 0.86207938,
      "num_input_tokens_seen": 31782690,
      "step": 1502,
      "time_per_iteration": 3.3340561389923096
    },
    {
      "auxiliary_loss_clip": 0.01200961,
      "auxiliary_loss_mlp": 0.01046974,
      "balance_loss_clip": 1.06473649,
      "balance_loss_mlp": 1.03644788,
      "epoch": 0.1807250646305537,
      "flos": 18587399512320.0,
      "grad_norm": 1.9180714958242873,
      "language_loss": 0.82524037,
      "learning_rate": 3.7665205554528437e-06,
      "loss": 0.84771967,
      "num_input_tokens_seen": 31802045,
      "step": 1503,
      "time_per_iteration": 3.227564573287964
    },
    {
      "auxiliary_loss_clip": 0.01195427,
      "auxiliary_loss_mlp": 0.01034344,
      "balance_loss_clip": 1.06666553,
      "balance_loss_mlp": 1.02470553,
      "epoch": 0.18084530752119282,
      "flos": 23149131880320.0,
      "grad_norm": 1.6546176147307523,
      "language_loss": 0.74496472,
      "learning_rate": 3.7661551748975782e-06,
      "loss": 0.76726246,
      "num_input_tokens_seen": 31820220,
      "step": 1504,
      "time_per_iteration": 3.391793966293335
    },
    {
      "auxiliary_loss_clip": 0.01088856,
      "auxiliary_loss_mlp": 0.01003506,
      "balance_loss_clip": 1.02463984,
      "balance_loss_mlp": 1.00107408,
      "epoch": 0.1809655504118319,
      "flos": 59803153568640.0,
      "grad_norm": 0.8156460389296397,
      "language_loss": 0.60500389,
      "learning_rate": 3.7657895264179772e-06,
      "loss": 0.62592751,
      "num_input_tokens_seen": 31876195,
      "step": 1505,
      "time_per_iteration": 3.1301932334899902
    },
    {
      "auxiliary_loss_clip": 0.01183661,
      "auxiliary_loss_mlp": 0.01033698,
      "balance_loss_clip": 1.05916739,
      "balance_loss_mlp": 1.02395225,
      "epoch": 0.181085793302471,
      "flos": 44201941188480.0,
      "grad_norm": 1.8596296540972728,
      "language_loss": 0.74606764,
      "learning_rate": 3.765423610069509e-06,
      "loss": 0.76824117,
      "num_input_tokens_seen": 31901585,
      "step": 1506,
      "time_per_iteration": 2.74784255027771
    },
    {
      "auxiliary_loss_clip": 0.01195788,
      "auxiliary_loss_mlp": 0.01038822,
      "balance_loss_clip": 1.06475949,
      "balance_loss_mlp": 1.02884972,
      "epoch": 0.18120603619311007,
      "flos": 34898384638080.0,
      "grad_norm": 1.8851615574828986,
      "language_loss": 0.72109091,
      "learning_rate": 3.765057425907683e-06,
      "loss": 0.74343699,
      "num_input_tokens_seen": 31923045,
      "step": 1507,
      "time_per_iteration": 2.6937994956970215
    },
    {
      "auxiliary_loss_clip": 0.01211167,
      "auxiliary_loss_mlp": 0.01036532,
      "balance_loss_clip": 1.06312633,
      "balance_loss_mlp": 1.02674508,
      "epoch": 0.18132627908374918,
      "flos": 21506757390720.0,
      "grad_norm": 1.768929514658489,
      "language_loss": 0.78538179,
      "learning_rate": 3.764690973988048e-06,
      "loss": 0.80785882,
      "num_input_tokens_seen": 31943385,
      "step": 1508,
      "time_per_iteration": 2.5595650672912598
    },
    {
      "auxiliary_loss_clip": 0.01182481,
      "auxiliary_loss_mlp": 0.01042709,
      "balance_loss_clip": 1.06036294,
      "balance_loss_mlp": 1.0328207,
      "epoch": 0.18144652197438826,
      "flos": 29057693633280.0,
      "grad_norm": 1.8455794211593006,
      "language_loss": 0.74236906,
      "learning_rate": 3.7643242543661967e-06,
      "loss": 0.76462096,
      "num_input_tokens_seen": 31966045,
      "step": 1509,
      "time_per_iteration": 2.6198956966400146
    },
    {
      "auxiliary_loss_clip": 0.01077482,
      "auxiliary_loss_mlp": 0.01004015,
      "balance_loss_clip": 1.01978374,
      "balance_loss_mlp": 1.00165451,
      "epoch": 0.18156676486502735,
      "flos": 68675064382080.0,
      "grad_norm": 0.8141100265282946,
      "language_loss": 0.6047442,
      "learning_rate": 3.7639572670977573e-06,
      "loss": 0.62555915,
      "num_input_tokens_seen": 32021540,
      "step": 1510,
      "time_per_iteration": 3.034628391265869
    },
    {
      "auxiliary_loss_clip": 0.0118307,
      "auxiliary_loss_mlp": 0.01032463,
      "balance_loss_clip": 1.06088781,
      "balance_loss_mlp": 1.02371883,
      "epoch": 0.18168700775566646,
      "flos": 26471515334400.0,
      "grad_norm": 1.689231618986345,
      "language_loss": 0.76439929,
      "learning_rate": 3.7635900122384042e-06,
      "loss": 0.78655457,
      "num_input_tokens_seen": 32044535,
      "step": 1511,
      "time_per_iteration": 2.633913040161133
    },
    {
      "auxiliary_loss_clip": 0.0119277,
      "auxiliary_loss_mlp": 0.01039108,
      "balance_loss_clip": 1.06035399,
      "balance_loss_mlp": 1.02766371,
      "epoch": 0.18180725064630554,
      "flos": 15005668884480.0,
      "grad_norm": 2.11101222314304,
      "language_loss": 0.86849564,
      "learning_rate": 3.7632224898438477e-06,
      "loss": 0.89081442,
      "num_input_tokens_seen": 32061010,
      "step": 1512,
      "time_per_iteration": 2.5246124267578125
    },
    {
      "auxiliary_loss_clip": 0.01183613,
      "auxiliary_loss_mlp": 0.01037638,
      "balance_loss_clip": 1.06146169,
      "balance_loss_mlp": 1.02763009,
      "epoch": 0.18192749353694462,
      "flos": 19682387665920.0,
      "grad_norm": 1.6838685803242521,
      "language_loss": 0.79145873,
      "learning_rate": 3.762854699969842e-06,
      "loss": 0.81367123,
      "num_input_tokens_seen": 32081520,
      "step": 1513,
      "time_per_iteration": 2.6028246879577637
    },
    {
      "auxiliary_loss_clip": 0.01204483,
      "auxiliary_loss_mlp": 0.01034155,
      "balance_loss_clip": 1.06427622,
      "balance_loss_mlp": 1.02461851,
      "epoch": 0.1820477364275837,
      "flos": 20702717400960.0,
      "grad_norm": 1.9062881956836504,
      "language_loss": 0.7312277,
      "learning_rate": 3.762486642672179e-06,
      "loss": 0.75361407,
      "num_input_tokens_seen": 32098460,
      "step": 1514,
      "time_per_iteration": 2.5115625858306885
    },
    {
      "auxiliary_loss_clip": 0.01197231,
      "auxiliary_loss_mlp": 0.01040359,
      "balance_loss_clip": 1.06478834,
      "balance_loss_mlp": 1.03078651,
      "epoch": 0.18216797931822282,
      "flos": 17128708197120.0,
      "grad_norm": 2.102554337710655,
      "language_loss": 0.86775666,
      "learning_rate": 3.7621183180066946e-06,
      "loss": 0.89013255,
      "num_input_tokens_seen": 32116420,
      "step": 1515,
      "time_per_iteration": 2.523895025253296
    },
    {
      "auxiliary_loss_clip": 0.01189582,
      "auxiliary_loss_mlp": 0.01038367,
      "balance_loss_clip": 1.05765152,
      "balance_loss_mlp": 1.02834129,
      "epoch": 0.1822882222088619,
      "flos": 29242561956480.0,
      "grad_norm": 1.968434127860691,
      "language_loss": 0.73876262,
      "learning_rate": 3.7617497260292625e-06,
      "loss": 0.76104206,
      "num_input_tokens_seen": 32138475,
      "step": 1516,
      "time_per_iteration": 2.6273505687713623
    },
    {
      "auxiliary_loss_clip": 0.0118839,
      "auxiliary_loss_mlp": 0.01037132,
      "balance_loss_clip": 1.06220722,
      "balance_loss_mlp": 1.02676654,
      "epoch": 0.18240846509950098,
      "flos": 17702739446400.0,
      "grad_norm": 3.349364183620209,
      "language_loss": 0.78685212,
      "learning_rate": 3.7613808667957967e-06,
      "loss": 0.8091073,
      "num_input_tokens_seen": 32151165,
      "step": 1517,
      "time_per_iteration": 2.5250847339630127
    },
    {
      "auxiliary_loss_clip": 0.01193989,
      "auxiliary_loss_mlp": 0.01043487,
      "balance_loss_clip": 1.0603497,
      "balance_loss_mlp": 1.03366995,
      "epoch": 0.1825287079901401,
      "flos": 14790025584000.0,
      "grad_norm": 1.974432659882828,
      "language_loss": 0.90683413,
      "learning_rate": 3.7610117403622547e-06,
      "loss": 0.92920893,
      "num_input_tokens_seen": 32167725,
      "step": 1518,
      "time_per_iteration": 2.523420572280884
    },
    {
      "auxiliary_loss_clip": 0.01174113,
      "auxiliary_loss_mlp": 0.01040583,
      "balance_loss_clip": 1.05876625,
      "balance_loss_mlp": 1.0306468,
      "epoch": 0.18264895088077918,
      "flos": 21946232292480.0,
      "grad_norm": 1.6944604763319746,
      "language_loss": 0.90227997,
      "learning_rate": 3.7606423467846313e-06,
      "loss": 0.92442691,
      "num_input_tokens_seen": 32187330,
      "step": 1519,
      "time_per_iteration": 2.6038784980773926
    },
    {
      "auxiliary_loss_clip": 0.01188438,
      "auxiliary_loss_mlp": 0.01050841,
      "balance_loss_clip": 1.06547582,
      "balance_loss_mlp": 1.04107177,
      "epoch": 0.18276919377141826,
      "flos": 20886759711360.0,
      "grad_norm": 1.4754682258010787,
      "language_loss": 0.79688585,
      "learning_rate": 3.760272686118964e-06,
      "loss": 0.8192786,
      "num_input_tokens_seen": 32205550,
      "step": 1520,
      "time_per_iteration": 2.5905654430389404
    },
    {
      "auxiliary_loss_clip": 0.01199473,
      "auxiliary_loss_mlp": 0.01033783,
      "balance_loss_clip": 1.06486034,
      "balance_loss_mlp": 1.02400219,
      "epoch": 0.18288943666205737,
      "flos": 21469877101440.0,
      "grad_norm": 2.119909026274675,
      "language_loss": 0.92671323,
      "learning_rate": 3.7599027584213297e-06,
      "loss": 0.94904578,
      "num_input_tokens_seen": 32224430,
      "step": 1521,
      "time_per_iteration": 2.569788932800293
    },
    {
      "auxiliary_loss_clip": 0.01211728,
      "auxiliary_loss_mlp": 0.01036578,
      "balance_loss_clip": 1.06329942,
      "balance_loss_mlp": 1.02660036,
      "epoch": 0.18300967955269645,
      "flos": 21539363961600.0,
      "grad_norm": 2.1363640443559104,
      "language_loss": 0.78239214,
      "learning_rate": 3.7595325637478465e-06,
      "loss": 0.80487525,
      "num_input_tokens_seen": 32242455,
      "step": 1522,
      "time_per_iteration": 2.509934425354004
    },
    {
      "auxiliary_loss_clip": 0.01183247,
      "auxiliary_loss_mlp": 0.01041711,
      "balance_loss_clip": 1.06197846,
      "balance_loss_mlp": 1.03139985,
      "epoch": 0.18312992244333554,
      "flos": 28876237102080.0,
      "grad_norm": 1.897685973617354,
      "language_loss": 0.81743228,
      "learning_rate": 3.7591621021546723e-06,
      "loss": 0.83968186,
      "num_input_tokens_seen": 32264450,
      "step": 1523,
      "time_per_iteration": 2.5924739837646484
    },
    {
      "auxiliary_loss_clip": 0.01203565,
      "auxiliary_loss_mlp": 0.01038506,
      "balance_loss_clip": 1.06330025,
      "balance_loss_mlp": 1.02834344,
      "epoch": 0.18325016533397462,
      "flos": 20120102801280.0,
      "grad_norm": 1.8704142237463572,
      "language_loss": 0.81470537,
      "learning_rate": 3.7587913736980062e-06,
      "loss": 0.83712602,
      "num_input_tokens_seen": 32284090,
      "step": 1524,
      "time_per_iteration": 3.221287727355957
    },
    {
      "auxiliary_loss_clip": 0.01129305,
      "auxiliary_loss_mlp": 0.01047868,
      "balance_loss_clip": 1.05044973,
      "balance_loss_mlp": 1.03841496,
      "epoch": 0.18337040822461373,
      "flos": 23329187781120.0,
      "grad_norm": 1.6767655551456115,
      "language_loss": 0.84322566,
      "learning_rate": 3.7584203784340865e-06,
      "loss": 0.86499739,
      "num_input_tokens_seen": 32303260,
      "step": 1525,
      "time_per_iteration": 2.6494548320770264
    },
    {
      "auxiliary_loss_clip": 0.01187608,
      "auxiliary_loss_mlp": 0.01035737,
      "balance_loss_clip": 1.0586679,
      "balance_loss_mlp": 1.02618194,
      "epoch": 0.1834906511152528,
      "flos": 25009555881600.0,
      "grad_norm": 2.5006777771099045,
      "language_loss": 0.85549092,
      "learning_rate": 3.7580491164191938e-06,
      "loss": 0.87772441,
      "num_input_tokens_seen": 32321570,
      "step": 1526,
      "time_per_iteration": 2.5863144397735596
    },
    {
      "auxiliary_loss_clip": 0.01099542,
      "auxiliary_loss_mlp": 0.01005181,
      "balance_loss_clip": 1.023561,
      "balance_loss_mlp": 1.00288057,
      "epoch": 0.1836108940058919,
      "flos": 67251493589760.0,
      "grad_norm": 0.7490254039796733,
      "language_loss": 0.61291361,
      "learning_rate": 3.757677587709648e-06,
      "loss": 0.63396084,
      "num_input_tokens_seen": 32384835,
      "step": 1527,
      "time_per_iteration": 3.9287827014923096
    },
    {
      "auxiliary_loss_clip": 0.01175726,
      "auxiliary_loss_mlp": 0.01035236,
      "balance_loss_clip": 1.06310415,
      "balance_loss_mlp": 1.02561557,
      "epoch": 0.183731136896531,
      "flos": 25738721971200.0,
      "grad_norm": 1.901934923493248,
      "language_loss": 0.75813723,
      "learning_rate": 3.7573057923618095e-06,
      "loss": 0.78024685,
      "num_input_tokens_seen": 32404930,
      "step": 1528,
      "time_per_iteration": 2.614283323287964
    },
    {
      "auxiliary_loss_clip": 0.01159313,
      "auxiliary_loss_mlp": 0.01037457,
      "balance_loss_clip": 1.05477118,
      "balance_loss_mlp": 1.02697217,
      "epoch": 0.1838513797871701,
      "flos": 20449403712000.0,
      "grad_norm": 3.0487601404800104,
      "language_loss": 0.7423656,
      "learning_rate": 3.7569337304320793e-06,
      "loss": 0.76433331,
      "num_input_tokens_seen": 32424515,
      "step": 1529,
      "time_per_iteration": 3.3069894313812256
    },
    {
      "auxiliary_loss_clip": 0.0108509,
      "auxiliary_loss_mlp": 0.01002326,
      "balance_loss_clip": 1.0211314,
      "balance_loss_mlp": 0.99998969,
      "epoch": 0.18397162267780917,
      "flos": 68565141786240.0,
      "grad_norm": 0.8470666919323194,
      "language_loss": 0.64457798,
      "learning_rate": 3.756561401976899e-06,
      "loss": 0.66545212,
      "num_input_tokens_seen": 32484220,
      "step": 1530,
      "time_per_iteration": 3.020946979522705
    },
    {
      "auxiliary_loss_clip": 0.01225184,
      "auxiliary_loss_mlp": 0.01037763,
      "balance_loss_clip": 1.06730294,
      "balance_loss_mlp": 1.02790415,
      "epoch": 0.18409186556844825,
      "flos": 31941104976000.0,
      "grad_norm": 1.7752049267531378,
      "language_loss": 0.82390738,
      "learning_rate": 3.7561888070527514e-06,
      "loss": 0.84653687,
      "num_input_tokens_seen": 32506260,
      "step": 1531,
      "time_per_iteration": 3.3986682891845703
    },
    {
      "auxiliary_loss_clip": 0.01161928,
      "auxiliary_loss_mlp": 0.00763049,
      "balance_loss_clip": 1.05631495,
      "balance_loss_mlp": 0.99995065,
      "epoch": 0.18421210845908736,
      "flos": 20120533764480.0,
      "grad_norm": 2.2375225032963004,
      "language_loss": 0.79808092,
      "learning_rate": 3.7558159457161577e-06,
      "loss": 0.81733072,
      "num_input_tokens_seen": 32524225,
      "step": 1532,
      "time_per_iteration": 2.5843939781188965
    },
    {
      "auxiliary_loss_clip": 0.01191487,
      "auxiliary_loss_mlp": 0.00762893,
      "balance_loss_clip": 1.06011915,
      "balance_loss_mlp": 0.9999702,
      "epoch": 0.18433235134972645,
      "flos": 23110491824640.0,
      "grad_norm": 2.3630891632324933,
      "language_loss": 0.77532244,
      "learning_rate": 3.755442818023681e-06,
      "loss": 0.7948662,
      "num_input_tokens_seen": 32543850,
      "step": 1533,
      "time_per_iteration": 2.5639564990997314
    },
    {
      "auxiliary_loss_clip": 0.01178188,
      "auxiliary_loss_mlp": 0.0103062,
      "balance_loss_clip": 1.06029916,
      "balance_loss_mlp": 1.02135181,
      "epoch": 0.18445259424036553,
      "flos": 18291351617280.0,
      "grad_norm": 1.9760335825044806,
      "language_loss": 0.76010692,
      "learning_rate": 3.7550694240319246e-06,
      "loss": 0.78219497,
      "num_input_tokens_seen": 32561725,
      "step": 1534,
      "time_per_iteration": 2.5595877170562744
    },
    {
      "auxiliary_loss_clip": 0.01212728,
      "auxiliary_loss_mlp": 0.01041286,
      "balance_loss_clip": 1.06418991,
      "balance_loss_mlp": 1.03177357,
      "epoch": 0.18457283713100464,
      "flos": 21324079797120.0,
      "grad_norm": 2.4658305033267327,
      "language_loss": 0.77134097,
      "learning_rate": 3.7546957637975326e-06,
      "loss": 0.79388118,
      "num_input_tokens_seen": 32579135,
      "step": 1535,
      "time_per_iteration": 2.5183582305908203
    },
    {
      "auxiliary_loss_clip": 0.01137556,
      "auxiliary_loss_mlp": 0.01035613,
      "balance_loss_clip": 1.04933047,
      "balance_loss_mlp": 1.02677941,
      "epoch": 0.18469308002164372,
      "flos": 20375679047040.0,
      "grad_norm": 1.5505820138035824,
      "language_loss": 0.74185795,
      "learning_rate": 3.7543218373771873e-06,
      "loss": 0.76358962,
      "num_input_tokens_seen": 32598460,
      "step": 1536,
      "time_per_iteration": 2.6275808811187744
    },
    {
      "auxiliary_loss_clip": 0.0113907,
      "auxiliary_loss_mlp": 0.00762731,
      "balance_loss_clip": 1.05277157,
      "balance_loss_mlp": 0.99995345,
      "epoch": 0.1848133229122828,
      "flos": 26435892021120.0,
      "grad_norm": 1.413192405025168,
      "language_loss": 0.77757478,
      "learning_rate": 3.753947644827615e-06,
      "loss": 0.79659283,
      "num_input_tokens_seen": 32621920,
      "step": 1537,
      "time_per_iteration": 2.773341178894043
    },
    {
      "auxiliary_loss_clip": 0.01084548,
      "auxiliary_loss_mlp": 0.01012139,
      "balance_loss_clip": 1.01953483,
      "balance_loss_mlp": 1.00973129,
      "epoch": 0.1849335658029219,
      "flos": 70547447612160.0,
      "grad_norm": 0.9394750042745883,
      "language_loss": 0.57167655,
      "learning_rate": 3.753573186205579e-06,
      "loss": 0.59264344,
      "num_input_tokens_seen": 32690040,
      "step": 1538,
      "time_per_iteration": 3.288184642791748
    },
    {
      "auxiliary_loss_clip": 0.01177926,
      "auxiliary_loss_mlp": 0.00762962,
      "balance_loss_clip": 1.05674195,
      "balance_loss_mlp": 0.99995553,
      "epoch": 0.185053808693561,
      "flos": 17384140788480.0,
      "grad_norm": 2.137564284253662,
      "language_loss": 0.7758376,
      "learning_rate": 3.753198461567885e-06,
      "loss": 0.79524648,
      "num_input_tokens_seen": 32707285,
      "step": 1539,
      "time_per_iteration": 2.536231517791748
    },
    {
      "auxiliary_loss_clip": 0.01170512,
      "auxiliary_loss_mlp": 0.01035419,
      "balance_loss_clip": 1.06242597,
      "balance_loss_mlp": 1.0265801,
      "epoch": 0.18517405158420008,
      "flos": 28986159697920.0,
      "grad_norm": 1.8125412293795107,
      "language_loss": 0.91884267,
      "learning_rate": 3.7528234709713783e-06,
      "loss": 0.94090199,
      "num_input_tokens_seen": 32730030,
      "step": 1540,
      "time_per_iteration": 2.633136749267578
    },
    {
      "auxiliary_loss_clip": 0.01211803,
      "auxiliary_loss_mlp": 0.01049627,
      "balance_loss_clip": 1.06628299,
      "balance_loss_mlp": 1.04032278,
      "epoch": 0.18529429447483917,
      "flos": 26794962328320.0,
      "grad_norm": 1.9768859262158636,
      "language_loss": 0.84305435,
      "learning_rate": 3.7524482144729447e-06,
      "loss": 0.86566865,
      "num_input_tokens_seen": 32749485,
      "step": 1541,
      "time_per_iteration": 2.5639114379882812
    },
    {
      "auxiliary_loss_clip": 0.01169858,
      "auxiliary_loss_mlp": 0.01037198,
      "balance_loss_clip": 1.05620933,
      "balance_loss_mlp": 1.02676129,
      "epoch": 0.18541453736547828,
      "flos": 13581595301760.0,
      "grad_norm": 2.162070760397459,
      "language_loss": 0.83690727,
      "learning_rate": 3.7520726921295106e-06,
      "loss": 0.85897779,
      "num_input_tokens_seen": 32766205,
      "step": 1542,
      "time_per_iteration": 2.5539703369140625
    },
    {
      "auxiliary_loss_clip": 0.01200362,
      "auxiliary_loss_mlp": 0.01045691,
      "balance_loss_clip": 1.05969083,
      "balance_loss_mlp": 1.0360235,
      "epoch": 0.18553478025611736,
      "flos": 24025424077440.0,
      "grad_norm": 2.0135258610314253,
      "language_loss": 0.7261343,
      "learning_rate": 3.751696903998042e-06,
      "loss": 0.74859488,
      "num_input_tokens_seen": 32784840,
      "step": 1543,
      "time_per_iteration": 2.545659065246582
    },
    {
      "auxiliary_loss_clip": 0.01205326,
      "auxiliary_loss_mlp": 0.01037053,
      "balance_loss_clip": 1.06652558,
      "balance_loss_mlp": 1.02792168,
      "epoch": 0.18565502314675644,
      "flos": 25885165720320.0,
      "grad_norm": 3.4351690661654124,
      "language_loss": 0.69872129,
      "learning_rate": 3.7513208501355456e-06,
      "loss": 0.72114515,
      "num_input_tokens_seen": 32805945,
      "step": 1544,
      "time_per_iteration": 2.581122398376465
    },
    {
      "auxiliary_loss_clip": 0.01191431,
      "auxiliary_loss_mlp": 0.01034131,
      "balance_loss_clip": 1.06305015,
      "balance_loss_mlp": 1.0249635,
      "epoch": 0.18577526603739553,
      "flos": 19610063631360.0,
      "grad_norm": 2.532714989420842,
      "language_loss": 0.83470249,
      "learning_rate": 3.750944530599069e-06,
      "loss": 0.85695809,
      "num_input_tokens_seen": 32825515,
      "step": 1545,
      "time_per_iteration": 2.592742443084717
    },
    {
      "auxiliary_loss_clip": 0.01214324,
      "auxiliary_loss_mlp": 0.01032609,
      "balance_loss_clip": 1.06543922,
      "balance_loss_mlp": 1.02312613,
      "epoch": 0.18589550892803464,
      "flos": 18474891137280.0,
      "grad_norm": 2.1361612923652844,
      "language_loss": 0.80833191,
      "learning_rate": 3.7505679454456992e-06,
      "loss": 0.83080125,
      "num_input_tokens_seen": 32842125,
      "step": 1546,
      "time_per_iteration": 2.510756015777588
    },
    {
      "auxiliary_loss_clip": 0.0112536,
      "auxiliary_loss_mlp": 0.01037643,
      "balance_loss_clip": 1.05226469,
      "balance_loss_mlp": 1.02854133,
      "epoch": 0.18601575181867372,
      "flos": 23549966726400.0,
      "grad_norm": 2.0448043156578812,
      "language_loss": 0.70070255,
      "learning_rate": 3.750191094732564e-06,
      "loss": 0.72233248,
      "num_input_tokens_seen": 32862990,
      "step": 1547,
      "time_per_iteration": 2.7696404457092285
    },
    {
      "auxiliary_loss_clip": 0.01122448,
      "auxiliary_loss_mlp": 0.00763267,
      "balance_loss_clip": 1.04884839,
      "balance_loss_mlp": 0.9999705,
      "epoch": 0.1861359947093128,
      "flos": 26360192108160.0,
      "grad_norm": 1.7003317984376691,
      "language_loss": 0.75442553,
      "learning_rate": 3.7498139785168313e-06,
      "loss": 0.77328265,
      "num_input_tokens_seen": 32883595,
      "step": 1548,
      "time_per_iteration": 2.8635051250457764
    },
    {
      "auxiliary_loss_clip": 0.01205898,
      "auxiliary_loss_mlp": 0.01036649,
      "balance_loss_clip": 1.06422997,
      "balance_loss_mlp": 1.02762485,
      "epoch": 0.1862562375999519,
      "flos": 23331198942720.0,
      "grad_norm": 1.7207088838599416,
      "language_loss": 0.77259612,
      "learning_rate": 3.749436596855709e-06,
      "loss": 0.79502153,
      "num_input_tokens_seen": 32902895,
      "step": 1549,
      "time_per_iteration": 2.5355703830718994
    },
    {
      "auxiliary_loss_clip": 0.0119779,
      "auxiliary_loss_mlp": 0.0104666,
      "balance_loss_clip": 1.06027389,
      "balance_loss_mlp": 1.0356096,
      "epoch": 0.186376480490591,
      "flos": 16648222942080.0,
      "grad_norm": 1.9263211145734314,
      "language_loss": 0.90676194,
      "learning_rate": 3.749058949806446e-06,
      "loss": 0.92920649,
      "num_input_tokens_seen": 32919620,
      "step": 1550,
      "time_per_iteration": 3.3030893802642822
    },
    {
      "auxiliary_loss_clip": 0.01209049,
      "auxiliary_loss_mlp": 0.01042724,
      "balance_loss_clip": 1.06452656,
      "balance_loss_mlp": 1.0334909,
      "epoch": 0.18649672338123008,
      "flos": 21468656039040.0,
      "grad_norm": 1.6321679991220814,
      "language_loss": 0.84358507,
      "learning_rate": 3.748681037426331e-06,
      "loss": 0.86610281,
      "num_input_tokens_seen": 32938830,
      "step": 1551,
      "time_per_iteration": 2.5320985317230225
    },
    {
      "auxiliary_loss_clip": 0.01223667,
      "auxiliary_loss_mlp": 0.01043876,
      "balance_loss_clip": 1.0667634,
      "balance_loss_mlp": 1.03407145,
      "epoch": 0.1866169662718692,
      "flos": 12312728386560.0,
      "grad_norm": 2.1290873349337214,
      "language_loss": 0.91829687,
      "learning_rate": 3.7483028597726936e-06,
      "loss": 0.94097227,
      "num_input_tokens_seen": 32955600,
      "step": 1552,
      "time_per_iteration": 2.4589030742645264
    },
    {
      "auxiliary_loss_clip": 0.01176953,
      "auxiliary_loss_mlp": 0.01050509,
      "balance_loss_clip": 1.06197023,
      "balance_loss_mlp": 1.03953612,
      "epoch": 0.18673720916250827,
      "flos": 23581280407680.0,
      "grad_norm": 1.7638450267452004,
      "language_loss": 0.62522751,
      "learning_rate": 3.7479244169029017e-06,
      "loss": 0.64750218,
      "num_input_tokens_seen": 32975390,
      "step": 1553,
      "time_per_iteration": 3.5767829418182373
    },
    {
      "auxiliary_loss_clip": 0.01206272,
      "auxiliary_loss_mlp": 0.01031111,
      "balance_loss_clip": 1.05982089,
      "balance_loss_mlp": 1.02143741,
      "epoch": 0.18685745205314735,
      "flos": 19718370115200.0,
      "grad_norm": 2.576689265557877,
      "language_loss": 0.73838198,
      "learning_rate": 3.7475457088743658e-06,
      "loss": 0.76075578,
      "num_input_tokens_seen": 32992640,
      "step": 1554,
      "time_per_iteration": 2.511605739593506
    },
    {
      "auxiliary_loss_clip": 0.01181872,
      "auxiliary_loss_mlp": 0.0103614,
      "balance_loss_clip": 1.05984354,
      "balance_loss_mlp": 1.02638245,
      "epoch": 0.18697769494378644,
      "flos": 34204123589760.0,
      "grad_norm": 1.9737628870263553,
      "language_loss": 0.74543142,
      "learning_rate": 3.7471667357445348e-06,
      "loss": 0.7676115,
      "num_input_tokens_seen": 33012470,
      "step": 1555,
      "time_per_iteration": 3.3819191455841064
    },
    {
      "auxiliary_loss_clip": 0.01144858,
      "auxiliary_loss_mlp": 0.01032203,
      "balance_loss_clip": 1.05571735,
      "balance_loss_mlp": 1.02294624,
      "epoch": 0.18709793783442555,
      "flos": 34241327101440.0,
      "grad_norm": 2.105336051580151,
      "language_loss": 0.72356141,
      "learning_rate": 3.7467874975709e-06,
      "loss": 0.745332,
      "num_input_tokens_seen": 33033275,
      "step": 1556,
      "time_per_iteration": 3.59055757522583
    },
    {
      "auxiliary_loss_clip": 0.01213722,
      "auxiliary_loss_mlp": 0.01038859,
      "balance_loss_clip": 1.06765676,
      "balance_loss_mlp": 1.02947164,
      "epoch": 0.18721818072506463,
      "flos": 40734550529280.0,
      "grad_norm": 2.316404847050491,
      "language_loss": 0.78320998,
      "learning_rate": 3.7464079944109904e-06,
      "loss": 0.80573583,
      "num_input_tokens_seen": 33055135,
      "step": 1557,
      "time_per_iteration": 2.6955080032348633
    },
    {
      "auxiliary_loss_clip": 0.01174933,
      "auxiliary_loss_mlp": 0.01036035,
      "balance_loss_clip": 1.05731547,
      "balance_loss_mlp": 1.02617645,
      "epoch": 0.18733842361570371,
      "flos": 22157386392960.0,
      "grad_norm": 1.817672367832006,
      "language_loss": 0.77580929,
      "learning_rate": 3.746028226322376e-06,
      "loss": 0.79791898,
      "num_input_tokens_seen": 33071015,
      "step": 1558,
      "time_per_iteration": 2.5665252208709717
    },
    {
      "auxiliary_loss_clip": 0.01187956,
      "auxiliary_loss_mlp": 0.01035751,
      "balance_loss_clip": 1.06078529,
      "balance_loss_mlp": 1.02747834,
      "epoch": 0.18745866650634282,
      "flos": 18914940656640.0,
      "grad_norm": 1.9069420471941214,
      "language_loss": 0.75631493,
      "learning_rate": 3.745648193362669e-06,
      "loss": 0.77855206,
      "num_input_tokens_seen": 33090370,
      "step": 1559,
      "time_per_iteration": 2.5902605056762695
    },
    {
      "auxiliary_loss_clip": 0.01192695,
      "auxiliary_loss_mlp": 0.01038514,
      "balance_loss_clip": 1.06400871,
      "balance_loss_mlp": 1.0293709,
      "epoch": 0.1875789093969819,
      "flos": 19314626267520.0,
      "grad_norm": 2.1280167972778536,
      "language_loss": 0.71972919,
      "learning_rate": 3.745267895589518e-06,
      "loss": 0.74204123,
      "num_input_tokens_seen": 33108910,
      "step": 1560,
      "time_per_iteration": 2.5557291507720947
    },
    {
      "auxiliary_loss_clip": 0.01189336,
      "auxiliary_loss_mlp": 0.01036951,
      "balance_loss_clip": 1.06040108,
      "balance_loss_mlp": 1.02689564,
      "epoch": 0.187699152287621,
      "flos": 17018965169280.0,
      "grad_norm": 2.833461108890595,
      "language_loss": 0.82187605,
      "learning_rate": 3.7448873330606154e-06,
      "loss": 0.84413892,
      "num_input_tokens_seen": 33126680,
      "step": 1561,
      "time_per_iteration": 2.5465331077575684
    },
    {
      "auxiliary_loss_clip": 0.01169454,
      "auxiliary_loss_mlp": 0.01043611,
      "balance_loss_clip": 1.06006181,
      "balance_loss_mlp": 1.03392577,
      "epoch": 0.18781939517826007,
      "flos": 22346384780160.0,
      "grad_norm": 2.043775681049307,
      "language_loss": 0.87289125,
      "learning_rate": 3.7445065058336914e-06,
      "loss": 0.89502192,
      "num_input_tokens_seen": 33145550,
      "step": 1562,
      "time_per_iteration": 2.573452949523926
    },
    {
      "auxiliary_loss_clip": 0.0114474,
      "auxiliary_loss_mlp": 0.01035922,
      "balance_loss_clip": 1.05005646,
      "balance_loss_mlp": 1.02606368,
      "epoch": 0.18793963806889918,
      "flos": 14611478054400.0,
      "grad_norm": 1.9512708451196548,
      "language_loss": 0.86468744,
      "learning_rate": 3.7441254139665176e-06,
      "loss": 0.8864941,
      "num_input_tokens_seen": 33161735,
      "step": 1563,
      "time_per_iteration": 2.582707405090332
    },
    {
      "auxiliary_loss_clip": 0.01222995,
      "auxiliary_loss_mlp": 0.01041988,
      "balance_loss_clip": 1.06849694,
      "balance_loss_mlp": 1.03278482,
      "epoch": 0.18805988095953827,
      "flos": 17457075354240.0,
      "grad_norm": 1.6870894677479416,
      "language_loss": 0.82578194,
      "learning_rate": 3.743744057516905e-06,
      "loss": 0.84843183,
      "num_input_tokens_seen": 33179795,
      "step": 1564,
      "time_per_iteration": 2.4786689281463623
    },
    {
      "auxiliary_loss_clip": 0.01159889,
      "auxiliary_loss_mlp": 0.01050165,
      "balance_loss_clip": 1.05807924,
      "balance_loss_mlp": 1.03954375,
      "epoch": 0.18818012385017735,
      "flos": 15043877976960.0,
      "grad_norm": 2.915923157980633,
      "language_loss": 0.88036609,
      "learning_rate": 3.743362436542706e-06,
      "loss": 0.90246665,
      "num_input_tokens_seen": 33194485,
      "step": 1565,
      "time_per_iteration": 2.594353675842285
    },
    {
      "auxiliary_loss_clip": 0.01220944,
      "auxiliary_loss_mlp": 0.01045611,
      "balance_loss_clip": 1.06598806,
      "balance_loss_mlp": 1.03630042,
      "epoch": 0.18830036674081646,
      "flos": 47551975136640.0,
      "grad_norm": 1.7679283692267709,
      "language_loss": 0.76665384,
      "learning_rate": 3.7429805511018115e-06,
      "loss": 0.7893194,
      "num_input_tokens_seen": 33216145,
      "step": 1566,
      "time_per_iteration": 2.751323699951172
    },
    {
      "auxiliary_loss_clip": 0.01174665,
      "auxiliary_loss_mlp": 0.00763703,
      "balance_loss_clip": 1.06283772,
      "balance_loss_mlp": 0.99996382,
      "epoch": 0.18842060963145554,
      "flos": 30044626698240.0,
      "grad_norm": 1.6098371619079042,
      "language_loss": 0.7797001,
      "learning_rate": 3.7425984012521524e-06,
      "loss": 0.79908371,
      "num_input_tokens_seen": 33236345,
      "step": 1567,
      "time_per_iteration": 2.658365249633789
    },
    {
      "auxiliary_loss_clip": 0.01071946,
      "auxiliary_loss_mlp": 0.00753401,
      "balance_loss_clip": 1.02488101,
      "balance_loss_mlp": 1.00011468,
      "epoch": 0.18854085252209463,
      "flos": 70318372625280.0,
      "grad_norm": 0.7397093051307336,
      "language_loss": 0.60428119,
      "learning_rate": 3.7422159870517025e-06,
      "loss": 0.62253463,
      "num_input_tokens_seen": 33301600,
      "step": 1568,
      "time_per_iteration": 3.187368631362915
    },
    {
      "auxiliary_loss_clip": 0.01191388,
      "auxiliary_loss_mlp": 0.01040071,
      "balance_loss_clip": 1.06223154,
      "balance_loss_mlp": 1.03099895,
      "epoch": 0.1886610954127337,
      "flos": 21289318410240.0,
      "grad_norm": 1.6218824778168317,
      "language_loss": 0.79163098,
      "learning_rate": 3.7418333085584717e-06,
      "loss": 0.81394553,
      "num_input_tokens_seen": 33322785,
      "step": 1569,
      "time_per_iteration": 2.6207797527313232
    },
    {
      "auxiliary_loss_clip": 0.01180528,
      "auxiliary_loss_mlp": 0.01043096,
      "balance_loss_clip": 1.06318498,
      "balance_loss_mlp": 1.03335059,
      "epoch": 0.18878133830337282,
      "flos": 17266819991040.0,
      "grad_norm": 2.895595849773554,
      "language_loss": 0.90939426,
      "learning_rate": 3.7414503658305128e-06,
      "loss": 0.93163049,
      "num_input_tokens_seen": 33340020,
      "step": 1570,
      "time_per_iteration": 2.5801126956939697
    },
    {
      "auxiliary_loss_clip": 0.01168922,
      "auxiliary_loss_mlp": 0.01043105,
      "balance_loss_clip": 1.05825305,
      "balance_loss_mlp": 1.03319883,
      "epoch": 0.1889015811940119,
      "flos": 25775207210880.0,
      "grad_norm": 2.0154541266340518,
      "language_loss": 0.77537584,
      "learning_rate": 3.7410671589259185e-06,
      "loss": 0.79749608,
      "num_input_tokens_seen": 33358620,
      "step": 1571,
      "time_per_iteration": 2.660343647003174
    },
    {
      "auxiliary_loss_clip": 0.012247,
      "auxiliary_loss_mlp": 0.01040396,
      "balance_loss_clip": 1.06999385,
      "balance_loss_mlp": 1.03039408,
      "epoch": 0.18902182408465099,
      "flos": 21032197879680.0,
      "grad_norm": 1.9460538247130146,
      "language_loss": 0.79485863,
      "learning_rate": 3.7406836879028205e-06,
      "loss": 0.81750959,
      "num_input_tokens_seen": 33378845,
      "step": 1572,
      "time_per_iteration": 2.5058295726776123
    },
    {
      "auxiliary_loss_clip": 0.01205059,
      "auxiliary_loss_mlp": 0.01047407,
      "balance_loss_clip": 1.06544745,
      "balance_loss_mlp": 1.03811479,
      "epoch": 0.1891420669752901,
      "flos": 22272121411200.0,
      "grad_norm": 1.8514262518484188,
      "language_loss": 0.76701748,
      "learning_rate": 3.7402999528193907e-06,
      "loss": 0.7895422,
      "num_input_tokens_seen": 33398345,
      "step": 1573,
      "time_per_iteration": 2.5200045108795166
    },
    {
      "auxiliary_loss_clip": 0.01164583,
      "auxiliary_loss_mlp": 0.0076385,
      "balance_loss_clip": 1.06117094,
      "balance_loss_mlp": 0.99997497,
      "epoch": 0.18926230986592918,
      "flos": 22017802141440.0,
      "grad_norm": 2.225848764168749,
      "language_loss": 0.85295284,
      "learning_rate": 3.739915953733842e-06,
      "loss": 0.87223715,
      "num_input_tokens_seen": 33416390,
      "step": 1574,
      "time_per_iteration": 2.595884323120117
    },
    {
      "auxiliary_loss_clip": 0.0122293,
      "auxiliary_loss_mlp": 0.01034792,
      "balance_loss_clip": 1.06788445,
      "balance_loss_mlp": 1.0255233,
      "epoch": 0.18938255275656826,
      "flos": 24462672336000.0,
      "grad_norm": 1.5271483554484147,
      "language_loss": 0.81873727,
      "learning_rate": 3.7395316907044264e-06,
      "loss": 0.84131444,
      "num_input_tokens_seen": 33437175,
      "step": 1575,
      "time_per_iteration": 2.541637897491455
    },
    {
      "auxiliary_loss_clip": 0.01209155,
      "auxiliary_loss_mlp": 0.01048727,
      "balance_loss_clip": 1.06492496,
      "balance_loss_mlp": 1.03816485,
      "epoch": 0.18950279564720737,
      "flos": 24427049022720.0,
      "grad_norm": 1.6606104031430846,
      "language_loss": 0.79969609,
      "learning_rate": 3.7391471637894364e-06,
      "loss": 0.82227492,
      "num_input_tokens_seen": 33459440,
      "step": 1576,
      "time_per_iteration": 3.332841157913208
    },
    {
      "auxiliary_loss_clip": 0.01176275,
      "auxiliary_loss_mlp": 0.01037652,
      "balance_loss_clip": 1.05751526,
      "balance_loss_mlp": 1.02794242,
      "epoch": 0.18962303853784646,
      "flos": 19756291898880.0,
      "grad_norm": 1.7207416401348117,
      "language_loss": 0.85116196,
      "learning_rate": 3.738762373047205e-06,
      "loss": 0.87330127,
      "num_input_tokens_seen": 33479360,
      "step": 1577,
      "time_per_iteration": 2.6020920276641846
    },
    {
      "auxiliary_loss_clip": 0.01181895,
      "auxiliary_loss_mlp": 0.01039342,
      "balance_loss_clip": 1.06253266,
      "balance_loss_mlp": 1.03021073,
      "epoch": 0.18974328142848554,
      "flos": 21032054225280.0,
      "grad_norm": 1.5993358643671012,
      "language_loss": 0.83223462,
      "learning_rate": 3.738377318536103e-06,
      "loss": 0.85444707,
      "num_input_tokens_seen": 33499245,
      "step": 1578,
      "time_per_iteration": 3.3471055030822754
    },
    {
      "auxiliary_loss_clip": 0.01218784,
      "auxiliary_loss_mlp": 0.01035563,
      "balance_loss_clip": 1.06743145,
      "balance_loss_mlp": 1.02675939,
      "epoch": 0.18986352431912462,
      "flos": 12966122736000.0,
      "grad_norm": 2.2916791920256476,
      "language_loss": 0.70852053,
      "learning_rate": 3.7379920003145447e-06,
      "loss": 0.73106396,
      "num_input_tokens_seen": 33513520,
      "step": 1579,
      "time_per_iteration": 2.463214874267578
    },
    {
      "auxiliary_loss_clip": 0.0118767,
      "auxiliary_loss_mlp": 0.01034849,
      "balance_loss_clip": 1.06668365,
      "balance_loss_mlp": 1.02434111,
      "epoch": 0.18998376720976373,
      "flos": 23767908497280.0,
      "grad_norm": 1.6783597772581498,
      "language_loss": 0.83887434,
      "learning_rate": 3.7376064184409817e-06,
      "loss": 0.86109948,
      "num_input_tokens_seen": 33533100,
      "step": 1580,
      "time_per_iteration": 2.5786995887756348
    },
    {
      "auxiliary_loss_clip": 0.01189477,
      "auxiliary_loss_mlp": 0.01043785,
      "balance_loss_clip": 1.06370533,
      "balance_loss_mlp": 1.03353858,
      "epoch": 0.19010401010040281,
      "flos": 22966023323520.0,
      "grad_norm": 1.4417709694603642,
      "language_loss": 0.87085807,
      "learning_rate": 3.7372205729739063e-06,
      "loss": 0.89319062,
      "num_input_tokens_seen": 33554915,
      "step": 1581,
      "time_per_iteration": 3.31112003326416
    },
    {
      "auxiliary_loss_clip": 0.01211809,
      "auxiliary_loss_mlp": 0.01044859,
      "balance_loss_clip": 1.06601357,
      "balance_loss_mlp": 1.03453541,
      "epoch": 0.1902242529910419,
      "flos": 19135647774720.0,
      "grad_norm": 2.5928263941316683,
      "language_loss": 0.71894276,
      "learning_rate": 3.7368344639718514e-06,
      "loss": 0.7415095,
      "num_input_tokens_seen": 33572850,
      "step": 1582,
      "time_per_iteration": 3.2351958751678467
    },
    {
      "auxiliary_loss_clip": 0.01210686,
      "auxiliary_loss_mlp": 0.01034794,
      "balance_loss_clip": 1.06549561,
      "balance_loss_mlp": 1.02542377,
      "epoch": 0.190344495881681,
      "flos": 25483935824640.0,
      "grad_norm": 1.6735957400698258,
      "language_loss": 0.80493921,
      "learning_rate": 3.7364480914933895e-06,
      "loss": 0.82739401,
      "num_input_tokens_seen": 33593090,
      "step": 1583,
      "time_per_iteration": 2.5721964836120605
    },
    {
      "auxiliary_loss_clip": 0.01161165,
      "auxiliary_loss_mlp": 0.00763551,
      "balance_loss_clip": 1.0606916,
      "balance_loss_mlp": 0.99995327,
      "epoch": 0.1904647387723201,
      "flos": 26792843425920.0,
      "grad_norm": 2.252631233473881,
      "language_loss": 0.81006652,
      "learning_rate": 3.7360614555971325e-06,
      "loss": 0.82931364,
      "num_input_tokens_seen": 33612745,
      "step": 1584,
      "time_per_iteration": 2.674231767654419
    },
    {
      "auxiliary_loss_clip": 0.01207129,
      "auxiliary_loss_mlp": 0.00762995,
      "balance_loss_clip": 1.06598055,
      "balance_loss_mlp": 0.99995089,
      "epoch": 0.19058498166295917,
      "flos": 23987753688960.0,
      "grad_norm": 1.9683190947197498,
      "language_loss": 0.85154605,
      "learning_rate": 3.735674556341733e-06,
      "loss": 0.87124729,
      "num_input_tokens_seen": 33632360,
      "step": 1585,
      "time_per_iteration": 2.5384464263916016
    },
    {
      "auxiliary_loss_clip": 0.01190474,
      "auxiliary_loss_mlp": 0.01039261,
      "balance_loss_clip": 1.06607461,
      "balance_loss_mlp": 1.02946222,
      "epoch": 0.19070522455359826,
      "flos": 28293299280000.0,
      "grad_norm": 1.8808456708347476,
      "language_loss": 0.82036543,
      "learning_rate": 3.7352873937858835e-06,
      "loss": 0.84266281,
      "num_input_tokens_seen": 33653895,
      "step": 1586,
      "time_per_iteration": 2.6135456562042236
    },
    {
      "auxiliary_loss_clip": 0.01172176,
      "auxiliary_loss_mlp": 0.00763291,
      "balance_loss_clip": 1.06119657,
      "balance_loss_mlp": 0.99996519,
      "epoch": 0.19082546744423737,
      "flos": 25660220797440.0,
      "grad_norm": 2.109215747995312,
      "language_loss": 0.71816301,
      "learning_rate": 3.734899967988316e-06,
      "loss": 0.73751771,
      "num_input_tokens_seen": 33672075,
      "step": 1587,
      "time_per_iteration": 2.6193926334381104
    },
    {
      "auxiliary_loss_clip": 0.01169248,
      "auxiliary_loss_mlp": 0.0103551,
      "balance_loss_clip": 1.05968571,
      "balance_loss_mlp": 1.02700996,
      "epoch": 0.19094571033487645,
      "flos": 19719483436800.0,
      "grad_norm": 1.8633460370060302,
      "language_loss": 0.83956748,
      "learning_rate": 3.7345122790078026e-06,
      "loss": 0.86161506,
      "num_input_tokens_seen": 33689640,
      "step": 1588,
      "time_per_iteration": 2.562445640563965
    },
    {
      "auxiliary_loss_clip": 0.01208098,
      "auxiliary_loss_mlp": 0.01048044,
      "balance_loss_clip": 1.0677706,
      "balance_loss_mlp": 1.0379169,
      "epoch": 0.19106595322551553,
      "flos": 21616320850560.0,
      "grad_norm": 3.3597897858932337,
      "language_loss": 0.92385173,
      "learning_rate": 3.7341243269031556e-06,
      "loss": 0.94641316,
      "num_input_tokens_seen": 33708630,
      "step": 1589,
      "time_per_iteration": 2.521916389465332
    },
    {
      "auxiliary_loss_clip": 0.0118601,
      "auxiliary_loss_mlp": 0.0104087,
      "balance_loss_clip": 1.06382263,
      "balance_loss_mlp": 1.03172624,
      "epoch": 0.19118619611615464,
      "flos": 29896890059520.0,
      "grad_norm": 1.9878148621655847,
      "language_loss": 0.77184314,
      "learning_rate": 3.7337361117332275e-06,
      "loss": 0.79411197,
      "num_input_tokens_seen": 33730370,
      "step": 1590,
      "time_per_iteration": 2.62974214553833
    },
    {
      "auxiliary_loss_clip": 0.01182797,
      "auxiliary_loss_mlp": 0.01042776,
      "balance_loss_clip": 1.06324339,
      "balance_loss_mlp": 1.03247666,
      "epoch": 0.19130643900679373,
      "flos": 17273428093440.0,
      "grad_norm": 2.767126878092092,
      "language_loss": 0.77270925,
      "learning_rate": 3.7333476335569087e-06,
      "loss": 0.79496503,
      "num_input_tokens_seen": 33748370,
      "step": 1591,
      "time_per_iteration": 2.567790985107422
    },
    {
      "auxiliary_loss_clip": 0.01194914,
      "auxiliary_loss_mlp": 0.01044931,
      "balance_loss_clip": 1.06686544,
      "balance_loss_mlp": 1.03414226,
      "epoch": 0.1914266818974328,
      "flos": 24826339584000.0,
      "grad_norm": 2.3652462505806624,
      "language_loss": 0.67202127,
      "learning_rate": 3.7329588924331325e-06,
      "loss": 0.69441968,
      "num_input_tokens_seen": 33769575,
      "step": 1592,
      "time_per_iteration": 2.593810558319092
    },
    {
      "auxiliary_loss_clip": 0.01169589,
      "auxiliary_loss_mlp": 0.0103522,
      "balance_loss_clip": 1.05703378,
      "balance_loss_mlp": 1.02609468,
      "epoch": 0.1915469247880719,
      "flos": 18952467390720.0,
      "grad_norm": 1.848251233538044,
      "language_loss": 0.82832348,
      "learning_rate": 3.732569888420871e-06,
      "loss": 0.8503716,
      "num_input_tokens_seen": 33789110,
      "step": 1593,
      "time_per_iteration": 2.5826942920684814
    },
    {
      "auxiliary_loss_clip": 0.01225083,
      "auxiliary_loss_mlp": 0.01040581,
      "balance_loss_clip": 1.06719661,
      "balance_loss_mlp": 1.03027487,
      "epoch": 0.191667167678711,
      "flos": 21032952065280.0,
      "grad_norm": 3.0539324790950633,
      "language_loss": 0.82551259,
      "learning_rate": 3.732180621579134e-06,
      "loss": 0.84816921,
      "num_input_tokens_seen": 33808325,
      "step": 1594,
      "time_per_iteration": 2.5077199935913086
    },
    {
      "auxiliary_loss_clip": 0.01189182,
      "auxiliary_loss_mlp": 0.01039352,
      "balance_loss_clip": 1.06585526,
      "balance_loss_mlp": 1.02927291,
      "epoch": 0.1917874105693501,
      "flos": 34237663914240.0,
      "grad_norm": 2.152136303767593,
      "language_loss": 0.8138991,
      "learning_rate": 3.7317910919669745e-06,
      "loss": 0.83618438,
      "num_input_tokens_seen": 33829520,
      "step": 1595,
      "time_per_iteration": 2.680912494659424
    },
    {
      "auxiliary_loss_clip": 0.01210084,
      "auxiliary_loss_mlp": 0.01039479,
      "balance_loss_clip": 1.07068038,
      "balance_loss_mlp": 1.02973986,
      "epoch": 0.19190765345998917,
      "flos": 23550613171200.0,
      "grad_norm": 2.668108021249357,
      "language_loss": 0.76453,
      "learning_rate": 3.7314012996434826e-06,
      "loss": 0.78702569,
      "num_input_tokens_seen": 33848250,
      "step": 1596,
      "time_per_iteration": 2.5401010513305664
    },
    {
      "auxiliary_loss_clip": 0.01193579,
      "auxiliary_loss_mlp": 0.010314,
      "balance_loss_clip": 1.06635487,
      "balance_loss_mlp": 1.02194691,
      "epoch": 0.19202789635062828,
      "flos": 19861330245120.0,
      "grad_norm": 2.1155611690859444,
      "language_loss": 0.80858648,
      "learning_rate": 3.7310112446677907e-06,
      "loss": 0.83083636,
      "num_input_tokens_seen": 33866160,
      "step": 1597,
      "time_per_iteration": 2.530593156814575
    },
    {
      "auxiliary_loss_clip": 0.01225033,
      "auxiliary_loss_mlp": 0.0104155,
      "balance_loss_clip": 1.06996918,
      "balance_loss_mlp": 1.03172743,
      "epoch": 0.19214813924126736,
      "flos": 20922957642240.0,
      "grad_norm": 2.175317961755692,
      "language_loss": 0.68722361,
      "learning_rate": 3.7306209270990695e-06,
      "loss": 0.70988941,
      "num_input_tokens_seen": 33884165,
      "step": 1598,
      "time_per_iteration": 2.4971094131469727
    },
    {
      "auxiliary_loss_clip": 0.01193571,
      "auxiliary_loss_mlp": 0.01039289,
      "balance_loss_clip": 1.06478572,
      "balance_loss_mlp": 1.02985954,
      "epoch": 0.19226838213190645,
      "flos": 26359725231360.0,
      "grad_norm": 1.8955949831120291,
      "language_loss": 0.86957085,
      "learning_rate": 3.7302303469965292e-06,
      "loss": 0.89189947,
      "num_input_tokens_seen": 33903705,
      "step": 1599,
      "time_per_iteration": 2.5892434120178223
    },
    {
      "auxiliary_loss_clip": 0.01209281,
      "auxiliary_loss_mlp": 0.01033866,
      "balance_loss_clip": 1.06941295,
      "balance_loss_mlp": 1.02390575,
      "epoch": 0.19238862502254553,
      "flos": 20850525866880.0,
      "grad_norm": 2.124731459402865,
      "language_loss": 0.70592016,
      "learning_rate": 3.7298395044194206e-06,
      "loss": 0.72835159,
      "num_input_tokens_seen": 33922515,
      "step": 1600,
      "time_per_iteration": 2.528857707977295
    },
    {
      "auxiliary_loss_clip": 0.01226768,
      "auxiliary_loss_mlp": 0.01036843,
      "balance_loss_clip": 1.07209706,
      "balance_loss_mlp": 1.02739549,
      "epoch": 0.19250886791318464,
      "flos": 21726063878400.0,
      "grad_norm": 1.785306764877352,
      "language_loss": 0.93947339,
      "learning_rate": 3.7294483994270356e-06,
      "loss": 0.96210945,
      "num_input_tokens_seen": 33940840,
      "step": 1601,
      "time_per_iteration": 2.49971342086792
    },
    {
      "auxiliary_loss_clip": 0.01157755,
      "auxiliary_loss_mlp": 0.01042009,
      "balance_loss_clip": 1.06000626,
      "balance_loss_mlp": 1.03283024,
      "epoch": 0.19262911080382372,
      "flos": 23367827836800.0,
      "grad_norm": 2.1367982376329238,
      "language_loss": 0.77881706,
      "learning_rate": 3.7290570320787033e-06,
      "loss": 0.80081475,
      "num_input_tokens_seen": 33960420,
      "step": 1602,
      "time_per_iteration": 3.3707187175750732
    },
    {
      "auxiliary_loss_clip": 0.01205087,
      "auxiliary_loss_mlp": 0.01042187,
      "balance_loss_clip": 1.06622124,
      "balance_loss_mlp": 1.0320065,
      "epoch": 0.1927493536944628,
      "flos": 21943502858880.0,
      "grad_norm": 2.066283884138619,
      "language_loss": 0.71182287,
      "learning_rate": 3.728665402433793e-06,
      "loss": 0.73429561,
      "num_input_tokens_seen": 33978990,
      "step": 1603,
      "time_per_iteration": 2.5080907344818115
    },
    {
      "auxiliary_loss_clip": 0.01190602,
      "auxiliary_loss_mlp": 0.0103903,
      "balance_loss_clip": 1.06432211,
      "balance_loss_mlp": 1.0298332,
      "epoch": 0.19286959658510192,
      "flos": 16545590807040.0,
      "grad_norm": 2.430601814237927,
      "language_loss": 0.85823929,
      "learning_rate": 3.7282735105517164e-06,
      "loss": 0.8805356,
      "num_input_tokens_seen": 33997115,
      "step": 1604,
      "time_per_iteration": 3.318866491317749
    },
    {
      "auxiliary_loss_clip": 0.01170947,
      "auxiliary_loss_mlp": 0.01044963,
      "balance_loss_clip": 1.06134415,
      "balance_loss_mlp": 1.0350033,
      "epoch": 0.192989839475741,
      "flos": 21616967295360.0,
      "grad_norm": 2.2769409088068504,
      "language_loss": 0.67114073,
      "learning_rate": 3.727881356491922e-06,
      "loss": 0.69329983,
      "num_input_tokens_seen": 34015525,
      "step": 1605,
      "time_per_iteration": 2.6759657859802246
    },
    {
      "auxiliary_loss_clip": 0.01225077,
      "auxiliary_loss_mlp": 0.01033853,
      "balance_loss_clip": 1.07316947,
      "balance_loss_mlp": 1.02528811,
      "epoch": 0.19311008236638008,
      "flos": 19281516906240.0,
      "grad_norm": 1.9055238019129455,
      "language_loss": 0.75721776,
      "learning_rate": 3.7274889403139002e-06,
      "loss": 0.77980709,
      "num_input_tokens_seen": 34033150,
      "step": 1606,
      "time_per_iteration": 2.4606754779815674
    },
    {
      "auxiliary_loss_clip": 0.0115798,
      "auxiliary_loss_mlp": 0.01031068,
      "balance_loss_clip": 1.06199849,
      "balance_loss_mlp": 1.0219599,
      "epoch": 0.1932303252570192,
      "flos": 28652369587200.0,
      "grad_norm": 2.897975487281842,
      "language_loss": 0.78316689,
      "learning_rate": 3.727096262077179e-06,
      "loss": 0.80505735,
      "num_input_tokens_seen": 34052145,
      "step": 1607,
      "time_per_iteration": 3.421605110168457
    },
    {
      "auxiliary_loss_clip": 0.01207986,
      "auxiliary_loss_mlp": 0.0103925,
      "balance_loss_clip": 1.06710792,
      "balance_loss_mlp": 1.03016675,
      "epoch": 0.19335056814765827,
      "flos": 18368990864640.0,
      "grad_norm": 1.8330893357982525,
      "language_loss": 0.85523307,
      "learning_rate": 3.7267033218413285e-06,
      "loss": 0.87770545,
      "num_input_tokens_seen": 34069940,
      "step": 1608,
      "time_per_iteration": 3.309817314147949
    },
    {
      "auxiliary_loss_clip": 0.01147369,
      "auxiliary_loss_mlp": 0.01040209,
      "balance_loss_clip": 1.05417132,
      "balance_loss_mlp": 1.03060102,
      "epoch": 0.19347081103829736,
      "flos": 13260877741440.0,
      "grad_norm": 2.1407826113892283,
      "language_loss": 0.8104462,
      "learning_rate": 3.726310119665957e-06,
      "loss": 0.832322,
      "num_input_tokens_seen": 34086275,
      "step": 1609,
      "time_per_iteration": 2.6206576824188232
    },
    {
      "auxiliary_loss_clip": 0.01208705,
      "auxiliary_loss_mlp": 0.01034858,
      "balance_loss_clip": 1.0672524,
      "balance_loss_mlp": 1.02573848,
      "epoch": 0.19359105392893644,
      "flos": 20300122788480.0,
      "grad_norm": 1.94994621344314,
      "language_loss": 0.85252362,
      "learning_rate": 3.725916655610713e-06,
      "loss": 0.87495923,
      "num_input_tokens_seen": 34105605,
      "step": 1610,
      "time_per_iteration": 2.5416109561920166
    },
    {
      "auxiliary_loss_clip": 0.01180643,
      "auxiliary_loss_mlp": 0.01039714,
      "balance_loss_clip": 1.06076884,
      "balance_loss_mlp": 1.03018367,
      "epoch": 0.19371129681957555,
      "flos": 20484596062080.0,
      "grad_norm": 4.936361726402711,
      "language_loss": 0.75070357,
      "learning_rate": 3.725522929735284e-06,
      "loss": 0.7729072,
      "num_input_tokens_seen": 34122540,
      "step": 1611,
      "time_per_iteration": 2.5264620780944824
    },
    {
      "auxiliary_loss_clip": 0.01194711,
      "auxiliary_loss_mlp": 0.01042943,
      "balance_loss_clip": 1.06017208,
      "balance_loss_mlp": 1.03363323,
      "epoch": 0.19383153971021463,
      "flos": 30445497457920.0,
      "grad_norm": 1.9971649939096234,
      "language_loss": 0.74061704,
      "learning_rate": 3.725128942099399e-06,
      "loss": 0.76299357,
      "num_input_tokens_seen": 34142940,
      "step": 1612,
      "time_per_iteration": 2.6491646766662598
    },
    {
      "auxiliary_loss_clip": 0.0118019,
      "auxiliary_loss_mlp": 0.01037872,
      "balance_loss_clip": 1.05943346,
      "balance_loss_mlp": 1.02820396,
      "epoch": 0.19395178260085372,
      "flos": 24569937325440.0,
      "grad_norm": 1.6282666223723283,
      "language_loss": 0.8012684,
      "learning_rate": 3.7247346927628245e-06,
      "loss": 0.82344902,
      "num_input_tokens_seen": 34162875,
      "step": 1613,
      "time_per_iteration": 2.576286792755127
    },
    {
      "auxiliary_loss_clip": 0.01193712,
      "auxiliary_loss_mlp": 0.00763102,
      "balance_loss_clip": 1.06629205,
      "balance_loss_mlp": 0.99999571,
      "epoch": 0.19407202549149283,
      "flos": 28950608211840.0,
      "grad_norm": 1.7497663935025447,
      "language_loss": 0.79053062,
      "learning_rate": 3.7243401817853694e-06,
      "loss": 0.81009877,
      "num_input_tokens_seen": 34183565,
      "step": 1614,
      "time_per_iteration": 2.6022913455963135
    },
    {
      "auxiliary_loss_clip": 0.01199165,
      "auxiliary_loss_mlp": 0.0103511,
      "balance_loss_clip": 1.06402922,
      "balance_loss_mlp": 1.0266099,
      "epoch": 0.1941922683821319,
      "flos": 18004497603840.0,
      "grad_norm": 1.8315426792994132,
      "language_loss": 0.7178247,
      "learning_rate": 3.723945409226879e-06,
      "loss": 0.74016744,
      "num_input_tokens_seen": 34202055,
      "step": 1615,
      "time_per_iteration": 2.489342212677002
    },
    {
      "auxiliary_loss_clip": 0.0120247,
      "auxiliary_loss_mlp": 0.01037313,
      "balance_loss_clip": 1.06162024,
      "balance_loss_mlp": 1.02859926,
      "epoch": 0.194312511272771,
      "flos": 9720337034880.0,
      "grad_norm": 2.2395806243731657,
      "language_loss": 0.80309802,
      "learning_rate": 3.723550375147241e-06,
      "loss": 0.82549584,
      "num_input_tokens_seen": 34216830,
      "step": 1616,
      "time_per_iteration": 2.595553398132324
    },
    {
      "auxiliary_loss_clip": 0.01164006,
      "auxiliary_loss_mlp": 0.01031516,
      "balance_loss_clip": 1.05850804,
      "balance_loss_mlp": 1.02150893,
      "epoch": 0.19443275416341008,
      "flos": 27016208150400.0,
      "grad_norm": 2.1438228125620475,
      "language_loss": 0.79932648,
      "learning_rate": 3.7231550796063816e-06,
      "loss": 0.82128167,
      "num_input_tokens_seen": 34236840,
      "step": 1617,
      "time_per_iteration": 2.6086628437042236
    },
    {
      "auxiliary_loss_clip": 0.01199231,
      "auxiliary_loss_mlp": 0.01039163,
      "balance_loss_clip": 1.0676837,
      "balance_loss_mlp": 1.02903616,
      "epoch": 0.1945529970540492,
      "flos": 15846625077120.0,
      "grad_norm": 2.124657731583981,
      "language_loss": 0.64883018,
      "learning_rate": 3.722759522664266e-06,
      "loss": 0.6712141,
      "num_input_tokens_seen": 34254140,
      "step": 1618,
      "time_per_iteration": 2.5219247341156006
    },
    {
      "auxiliary_loss_clip": 0.01165326,
      "auxiliary_loss_mlp": 0.01035637,
      "balance_loss_clip": 1.05960512,
      "balance_loss_mlp": 1.02610648,
      "epoch": 0.19467323994468827,
      "flos": 19314985403520.0,
      "grad_norm": 2.445517349566283,
      "language_loss": 0.81907588,
      "learning_rate": 3.7223637043809016e-06,
      "loss": 0.84108555,
      "num_input_tokens_seen": 34273120,
      "step": 1619,
      "time_per_iteration": 2.5876033306121826
    },
    {
      "auxiliary_loss_clip": 0.01181123,
      "auxiliary_loss_mlp": 0.01042436,
      "balance_loss_clip": 1.06553149,
      "balance_loss_mlp": 1.03360248,
      "epoch": 0.19479348283532735,
      "flos": 24133227770880.0,
      "grad_norm": 1.6982010173416828,
      "language_loss": 0.86391312,
      "learning_rate": 3.7219676248163322e-06,
      "loss": 0.88614875,
      "num_input_tokens_seen": 34290285,
      "step": 1620,
      "time_per_iteration": 2.59574818611145
    },
    {
      "auxiliary_loss_clip": 0.01211218,
      "auxiliary_loss_mlp": 0.0103989,
      "balance_loss_clip": 1.06813896,
      "balance_loss_mlp": 1.03050792,
      "epoch": 0.19491372572596646,
      "flos": 25775638174080.0,
      "grad_norm": 1.7038799929322694,
      "language_loss": 0.93443179,
      "learning_rate": 3.721571284030643e-06,
      "loss": 0.95694292,
      "num_input_tokens_seen": 34310095,
      "step": 1621,
      "time_per_iteration": 2.547821044921875
    },
    {
      "auxiliary_loss_clip": 0.01211485,
      "auxiliary_loss_mlp": 0.01040123,
      "balance_loss_clip": 1.06786156,
      "balance_loss_mlp": 1.03079462,
      "epoch": 0.19503396861660555,
      "flos": 19645220067840.0,
      "grad_norm": 2.502105022175523,
      "language_loss": 0.79054672,
      "learning_rate": 3.7211746820839587e-06,
      "loss": 0.81306279,
      "num_input_tokens_seen": 34327190,
      "step": 1622,
      "time_per_iteration": 2.510995864868164
    },
    {
      "auxiliary_loss_clip": 0.0111523,
      "auxiliary_loss_mlp": 0.01030042,
      "balance_loss_clip": 1.05225563,
      "balance_loss_mlp": 1.02107739,
      "epoch": 0.19515421150724463,
      "flos": 21033023892480.0,
      "grad_norm": 1.66729912829583,
      "language_loss": 0.80894327,
      "learning_rate": 3.7207778190364437e-06,
      "loss": 0.83039594,
      "num_input_tokens_seen": 34345615,
      "step": 1623,
      "time_per_iteration": 2.661425828933716
    },
    {
      "auxiliary_loss_clip": 0.01127833,
      "auxiliary_loss_mlp": 0.01043969,
      "balance_loss_clip": 1.05255485,
      "balance_loss_mlp": 1.03468239,
      "epoch": 0.1952744543978837,
      "flos": 32961255143040.0,
      "grad_norm": 1.516021560858203,
      "language_loss": 0.73783696,
      "learning_rate": 3.720380694948302e-06,
      "loss": 0.75955504,
      "num_input_tokens_seen": 34368500,
      "step": 1624,
      "time_per_iteration": 2.754500389099121
    },
    {
      "auxiliary_loss_clip": 0.01070753,
      "auxiliary_loss_mlp": 0.01003295,
      "balance_loss_clip": 1.02800202,
      "balance_loss_mlp": 1.00142312,
      "epoch": 0.19539469728852282,
      "flos": 64044312030720.0,
      "grad_norm": 1.0450052395083482,
      "language_loss": 0.71250159,
      "learning_rate": 3.719983309879777e-06,
      "loss": 0.73324203,
      "num_input_tokens_seen": 34428280,
      "step": 1625,
      "time_per_iteration": 3.178575038909912
    },
    {
      "auxiliary_loss_clip": 0.01164572,
      "auxiliary_loss_mlp": 0.01046766,
      "balance_loss_clip": 1.05908418,
      "balance_loss_mlp": 1.0369668,
      "epoch": 0.1955149401791619,
      "flos": 13370908078080.0,
      "grad_norm": 1.7297579136428711,
      "language_loss": 0.77489305,
      "learning_rate": 3.719585663891151e-06,
      "loss": 0.79700649,
      "num_input_tokens_seen": 34445815,
      "step": 1626,
      "time_per_iteration": 2.5778322219848633
    },
    {
      "auxiliary_loss_clip": 0.01150409,
      "auxiliary_loss_mlp": 0.01040869,
      "balance_loss_clip": 1.05883801,
      "balance_loss_mlp": 1.03130853,
      "epoch": 0.195635183069801,
      "flos": 18728887184640.0,
      "grad_norm": 2.06231860278537,
      "language_loss": 0.78809005,
      "learning_rate": 3.719187757042747e-06,
      "loss": 0.8100028,
      "num_input_tokens_seen": 34463635,
      "step": 1627,
      "time_per_iteration": 2.5879223346710205
    },
    {
      "auxiliary_loss_clip": 0.01087725,
      "auxiliary_loss_mlp": 0.01004702,
      "balance_loss_clip": 1.02703094,
      "balance_loss_mlp": 1.00281811,
      "epoch": 0.1957554259604401,
      "flos": 69313952615040.0,
      "grad_norm": 0.7413128730078512,
      "language_loss": 0.55001438,
      "learning_rate": 3.7187895893949275e-06,
      "loss": 0.57093859,
      "num_input_tokens_seen": 34530105,
      "step": 1628,
      "time_per_iteration": 3.9408671855926514
    },
    {
      "auxiliary_loss_clip": 0.01143294,
      "auxiliary_loss_mlp": 0.01034143,
      "balance_loss_clip": 1.05248523,
      "balance_loss_mlp": 1.02475488,
      "epoch": 0.19587566885107918,
      "flos": 21069257736960.0,
      "grad_norm": 2.2904547130428776,
      "language_loss": 0.76186401,
      "learning_rate": 3.7183911610080937e-06,
      "loss": 0.78363836,
      "num_input_tokens_seen": 34546970,
      "step": 1629,
      "time_per_iteration": 2.638697385787964
    },
    {
      "auxiliary_loss_clip": 0.0117936,
      "auxiliary_loss_mlp": 0.01038143,
      "balance_loss_clip": 1.06250405,
      "balance_loss_mlp": 1.02808189,
      "epoch": 0.19599591174171827,
      "flos": 22194661731840.0,
      "grad_norm": 3.0567512892015296,
      "language_loss": 0.74920225,
      "learning_rate": 3.7179924719426872e-06,
      "loss": 0.77137721,
      "num_input_tokens_seen": 34564865,
      "step": 1630,
      "time_per_iteration": 3.418473482131958
    },
    {
      "auxiliary_loss_clip": 0.01211234,
      "auxiliary_loss_mlp": 0.01037349,
      "balance_loss_clip": 1.06916034,
      "balance_loss_mlp": 1.02675676,
      "epoch": 0.19611615463235738,
      "flos": 23768375374080.0,
      "grad_norm": 2.653027540645547,
      "language_loss": 0.76022184,
      "learning_rate": 3.7175935222591885e-06,
      "loss": 0.78270763,
      "num_input_tokens_seen": 34584165,
      "step": 1631,
      "time_per_iteration": 2.5463669300079346
    },
    {
      "auxiliary_loss_clip": 0.01196606,
      "auxiliary_loss_mlp": 0.01040632,
      "balance_loss_clip": 1.06804824,
      "balance_loss_mlp": 1.03114915,
      "epoch": 0.19623639752299646,
      "flos": 28618218731520.0,
      "grad_norm": 1.87296166965338,
      "language_loss": 0.74371636,
      "learning_rate": 3.717194312018118e-06,
      "loss": 0.76608878,
      "num_input_tokens_seen": 34603150,
      "step": 1632,
      "time_per_iteration": 2.6122450828552246
    },
    {
      "auxiliary_loss_clip": 0.01207141,
      "auxiliary_loss_mlp": 0.0103979,
      "balance_loss_clip": 1.06560934,
      "balance_loss_mlp": 1.03019929,
      "epoch": 0.19635664041363554,
      "flos": 21032700670080.0,
      "grad_norm": 1.9460597192089308,
      "language_loss": 0.75805688,
      "learning_rate": 3.716794841280036e-06,
      "loss": 0.78052628,
      "num_input_tokens_seen": 34621855,
      "step": 1633,
      "time_per_iteration": 3.272639274597168
    },
    {
      "auxiliary_loss_clip": 0.01210701,
      "auxiliary_loss_mlp": 0.01035661,
      "balance_loss_clip": 1.06453967,
      "balance_loss_mlp": 1.02564764,
      "epoch": 0.19647688330427462,
      "flos": 18879748306560.0,
      "grad_norm": 1.93153429877565,
      "language_loss": 0.7748543,
      "learning_rate": 3.7163951101055407e-06,
      "loss": 0.79731792,
      "num_input_tokens_seen": 34639915,
      "step": 1634,
      "time_per_iteration": 3.2295241355895996
    },
    {
      "auxiliary_loss_clip": 0.01184948,
      "auxiliary_loss_mlp": 0.01042427,
      "balance_loss_clip": 1.06239843,
      "balance_loss_mlp": 1.0324192,
      "epoch": 0.19659712619491373,
      "flos": 24242503921920.0,
      "grad_norm": 1.8455574489839723,
      "language_loss": 0.7878921,
      "learning_rate": 3.715995118555273e-06,
      "loss": 0.81016588,
      "num_input_tokens_seen": 34659890,
      "step": 1635,
      "time_per_iteration": 2.5671169757843018
    },
    {
      "auxiliary_loss_clip": 0.01154313,
      "auxiliary_loss_mlp": 0.01046557,
      "balance_loss_clip": 1.05686176,
      "balance_loss_mlp": 1.03705561,
      "epoch": 0.19671736908555282,
      "flos": 24717422568960.0,
      "grad_norm": 2.01976674002462,
      "language_loss": 0.86075938,
      "learning_rate": 3.71559486668991e-06,
      "loss": 0.88276809,
      "num_input_tokens_seen": 34678750,
      "step": 1636,
      "time_per_iteration": 2.6302835941314697
    },
    {
      "auxiliary_loss_clip": 0.0120871,
      "auxiliary_loss_mlp": 0.00762849,
      "balance_loss_clip": 1.06486833,
      "balance_loss_mlp": 1.00003982,
      "epoch": 0.1968376119761919,
      "flos": 23842279607040.0,
      "grad_norm": 1.8319770376628974,
      "language_loss": 0.7733829,
      "learning_rate": 3.715194354570169e-06,
      "loss": 0.79309857,
      "num_input_tokens_seen": 34698755,
      "step": 1637,
      "time_per_iteration": 2.5438408851623535
    },
    {
      "auxiliary_loss_clip": 0.01208172,
      "auxiliary_loss_mlp": 0.01038372,
      "balance_loss_clip": 1.06956387,
      "balance_loss_mlp": 1.02988458,
      "epoch": 0.196957854866831,
      "flos": 18113917409280.0,
      "grad_norm": 1.829312273207489,
      "language_loss": 0.83151591,
      "learning_rate": 3.714793582256809e-06,
      "loss": 0.85398138,
      "num_input_tokens_seen": 34715820,
      "step": 1638,
      "time_per_iteration": 2.4935898780822754
    },
    {
      "auxiliary_loss_clip": 0.01222556,
      "auxiliary_loss_mlp": 0.01039566,
      "balance_loss_clip": 1.06954718,
      "balance_loss_mlp": 1.03054237,
      "epoch": 0.1970780977574701,
      "flos": 21653129312640.0,
      "grad_norm": 2.4409974170492883,
      "language_loss": 0.84953928,
      "learning_rate": 3.7143925498106253e-06,
      "loss": 0.87216055,
      "num_input_tokens_seen": 34734360,
      "step": 1639,
      "time_per_iteration": 2.482412815093994
    },
    {
      "auxiliary_loss_clip": 0.01189857,
      "auxiliary_loss_mlp": 0.01039213,
      "balance_loss_clip": 1.0598489,
      "balance_loss_mlp": 1.02907407,
      "epoch": 0.19719834064810918,
      "flos": 20811813984000.0,
      "grad_norm": 1.9602916486412927,
      "language_loss": 0.7920441,
      "learning_rate": 3.7139912572924558e-06,
      "loss": 0.81433481,
      "num_input_tokens_seen": 34753390,
      "step": 1640,
      "time_per_iteration": 2.569084882736206
    },
    {
      "auxiliary_loss_clip": 0.01201437,
      "auxiliary_loss_mlp": 0.01033835,
      "balance_loss_clip": 1.0610249,
      "balance_loss_mlp": 1.02381516,
      "epoch": 0.19731858353874826,
      "flos": 23434800744960.0,
      "grad_norm": 2.8047809100343,
      "language_loss": 0.80473369,
      "learning_rate": 3.7135897047631744e-06,
      "loss": 0.82708633,
      "num_input_tokens_seen": 34771275,
      "step": 1641,
      "time_per_iteration": 2.527642250061035
    },
    {
      "auxiliary_loss_clip": 0.01194726,
      "auxiliary_loss_mlp": 0.0103712,
      "balance_loss_clip": 1.06532276,
      "balance_loss_mlp": 1.02816117,
      "epoch": 0.19743882642938737,
      "flos": 23988184652160.0,
      "grad_norm": 2.1323922182323445,
      "language_loss": 0.7601124,
      "learning_rate": 3.713187892283698e-06,
      "loss": 0.78243089,
      "num_input_tokens_seen": 34790885,
      "step": 1642,
      "time_per_iteration": 2.5713937282562256
    },
    {
      "auxiliary_loss_clip": 0.01159388,
      "auxiliary_loss_mlp": 0.0103736,
      "balance_loss_clip": 1.05697334,
      "balance_loss_mlp": 1.02758515,
      "epoch": 0.19755906932002645,
      "flos": 15004340081280.0,
      "grad_norm": 2.1669940921877746,
      "language_loss": 0.87160707,
      "learning_rate": 3.71278581991498e-06,
      "loss": 0.8935746,
      "num_input_tokens_seen": 34806745,
      "step": 1643,
      "time_per_iteration": 2.57444429397583
    },
    {
      "auxiliary_loss_clip": 0.01177839,
      "auxiliary_loss_mlp": 0.00763737,
      "balance_loss_clip": 1.06376219,
      "balance_loss_mlp": 1.00004983,
      "epoch": 0.19767931221066554,
      "flos": 19494466686720.0,
      "grad_norm": 2.1251505445119108,
      "language_loss": 0.78275228,
      "learning_rate": 3.712383487718015e-06,
      "loss": 0.80216801,
      "num_input_tokens_seen": 34824985,
      "step": 1644,
      "time_per_iteration": 2.578662633895874
    },
    {
      "auxiliary_loss_clip": 0.0113545,
      "auxiliary_loss_mlp": 0.01039547,
      "balance_loss_clip": 1.05492628,
      "balance_loss_mlp": 1.0305531,
      "epoch": 0.19779955510130465,
      "flos": 25737895958400.0,
      "grad_norm": 1.8439776060040405,
      "language_loss": 0.86640251,
      "learning_rate": 3.7119808957538365e-06,
      "loss": 0.88815248,
      "num_input_tokens_seen": 34843980,
      "step": 1645,
      "time_per_iteration": 2.647707223892212
    },
    {
      "auxiliary_loss_clip": 0.01183152,
      "auxiliary_loss_mlp": 0.01044005,
      "balance_loss_clip": 1.05762625,
      "balance_loss_mlp": 1.03386641,
      "epoch": 0.19791979799194373,
      "flos": 20777699041920.0,
      "grad_norm": 1.9501377555327886,
      "language_loss": 0.80147415,
      "learning_rate": 3.711578044083517e-06,
      "loss": 0.82374573,
      "num_input_tokens_seen": 34860780,
      "step": 1646,
      "time_per_iteration": 2.547541856765747
    },
    {
      "auxiliary_loss_clip": 0.01190153,
      "auxiliary_loss_mlp": 0.01037249,
      "balance_loss_clip": 1.06017685,
      "balance_loss_mlp": 1.02786708,
      "epoch": 0.1980400408825828,
      "flos": 25589010084480.0,
      "grad_norm": 1.6616707598656286,
      "language_loss": 0.7458334,
      "learning_rate": 3.7111749327681698e-06,
      "loss": 0.76810741,
      "num_input_tokens_seen": 34880815,
      "step": 1647,
      "time_per_iteration": 2.5879123210906982
    },
    {
      "auxiliary_loss_clip": 0.01206919,
      "auxiliary_loss_mlp": 0.01032961,
      "balance_loss_clip": 1.06561279,
      "balance_loss_mlp": 1.02335286,
      "epoch": 0.1981602837732219,
      "flos": 23513840622720.0,
      "grad_norm": 2.4490809756660132,
      "language_loss": 0.85995662,
      "learning_rate": 3.7107715618689455e-06,
      "loss": 0.88235539,
      "num_input_tokens_seen": 34899790,
      "step": 1648,
      "time_per_iteration": 2.542264461517334
    },
    {
      "auxiliary_loss_clip": 0.01201459,
      "auxiliary_loss_mlp": 0.01037928,
      "balance_loss_clip": 1.06430554,
      "balance_loss_mlp": 1.02852809,
      "epoch": 0.198280526663861,
      "flos": 23185365724800.0,
      "grad_norm": 1.5080095576724653,
      "language_loss": 0.83385187,
      "learning_rate": 3.710367931447035e-06,
      "loss": 0.85624576,
      "num_input_tokens_seen": 34921570,
      "step": 1649,
      "time_per_iteration": 2.5481879711151123
    },
    {
      "auxiliary_loss_clip": 0.01209605,
      "auxiliary_loss_mlp": 0.01045607,
      "balance_loss_clip": 1.06373739,
      "balance_loss_mlp": 1.03575397,
      "epoch": 0.1984007695545001,
      "flos": 21689470897920.0,
      "grad_norm": 3.453751346777361,
      "language_loss": 0.86955124,
      "learning_rate": 3.70996404156367e-06,
      "loss": 0.89210337,
      "num_input_tokens_seen": 34941205,
      "step": 1650,
      "time_per_iteration": 2.5325958728790283
    },
    {
      "auxiliary_loss_clip": 0.01142327,
      "auxiliary_loss_mlp": 0.01048036,
      "balance_loss_clip": 1.05143237,
      "balance_loss_mlp": 1.03858829,
      "epoch": 0.19852101244513917,
      "flos": 36064008887040.0,
      "grad_norm": 1.6790164296390193,
      "language_loss": 0.72803462,
      "learning_rate": 3.7095598922801187e-06,
      "loss": 0.74993825,
      "num_input_tokens_seen": 34963280,
      "step": 1651,
      "time_per_iteration": 2.7295308113098145
    },
    {
      "auxiliary_loss_clip": 0.01221923,
      "auxiliary_loss_mlp": 0.01036122,
      "balance_loss_clip": 1.06751752,
      "balance_loss_mlp": 1.02765775,
      "epoch": 0.19864125533577828,
      "flos": 23105894883840.0,
      "grad_norm": 3.150521696053154,
      "language_loss": 0.76295793,
      "learning_rate": 3.7091554836576914e-06,
      "loss": 0.78553832,
      "num_input_tokens_seen": 34979955,
      "step": 1652,
      "time_per_iteration": 2.487957239151001
    },
    {
      "auxiliary_loss_clip": 0.01202418,
      "auxiliary_loss_mlp": 0.00762309,
      "balance_loss_clip": 1.06686866,
      "balance_loss_mlp": 1.00005031,
      "epoch": 0.19876149822641737,
      "flos": 24608505553920.0,
      "grad_norm": 1.6855345552631666,
      "language_loss": 0.82677126,
      "learning_rate": 3.708750815757736e-06,
      "loss": 0.8464185,
      "num_input_tokens_seen": 35000725,
      "step": 1653,
      "time_per_iteration": 2.5743284225463867
    },
    {
      "auxiliary_loss_clip": 0.01202944,
      "auxiliary_loss_mlp": 0.01042441,
      "balance_loss_clip": 1.06324387,
      "balance_loss_mlp": 1.03341103,
      "epoch": 0.19888174111705645,
      "flos": 32196645308160.0,
      "grad_norm": 2.1469407760084045,
      "language_loss": 0.73211455,
      "learning_rate": 3.7083458886416407e-06,
      "loss": 0.7545684,
      "num_input_tokens_seen": 35019920,
      "step": 1654,
      "time_per_iteration": 3.2785933017730713
    },
    {
      "auxiliary_loss_clip": 0.01145236,
      "auxiliary_loss_mlp": 0.01035049,
      "balance_loss_clip": 1.05686259,
      "balance_loss_mlp": 1.0260253,
      "epoch": 0.19900198400769553,
      "flos": 24608469640320.0,
      "grad_norm": 2.352900313313421,
      "language_loss": 0.88229173,
      "learning_rate": 3.707940702370832e-06,
      "loss": 0.90409458,
      "num_input_tokens_seen": 35040765,
      "step": 1655,
      "time_per_iteration": 2.6687936782836914
    },
    {
      "auxiliary_loss_clip": 0.01091479,
      "auxiliary_loss_mlp": 0.01002835,
      "balance_loss_clip": 1.0230577,
      "balance_loss_mlp": 1.00076103,
      "epoch": 0.19912222689833464,
      "flos": 67915805673600.0,
      "grad_norm": 0.7577168363036942,
      "language_loss": 0.582515,
      "learning_rate": 3.707535257006777e-06,
      "loss": 0.60345817,
      "num_input_tokens_seen": 35106390,
      "step": 1656,
      "time_per_iteration": 3.9134440422058105
    },
    {
      "auxiliary_loss_clip": 0.0118635,
      "auxiliary_loss_mlp": 0.010381,
      "balance_loss_clip": 1.05930853,
      "balance_loss_mlp": 1.02892661,
      "epoch": 0.19924246978897373,
      "flos": 15742340916480.0,
      "grad_norm": 2.083139257330628,
      "language_loss": 0.88520992,
      "learning_rate": 3.707129552610981e-06,
      "loss": 0.90745437,
      "num_input_tokens_seen": 35125040,
      "step": 1657,
      "time_per_iteration": 2.520728349685669
    },
    {
      "auxiliary_loss_clip": 0.01182783,
      "auxiliary_loss_mlp": 0.01040741,
      "balance_loss_clip": 1.06299591,
      "balance_loss_mlp": 1.03122842,
      "epoch": 0.1993627126796128,
      "flos": 17566566986880.0,
      "grad_norm": 1.8843867576530986,
      "language_loss": 0.7371524,
      "learning_rate": 3.70672358924499e-06,
      "loss": 0.75938761,
      "num_input_tokens_seen": 35144280,
      "step": 1658,
      "time_per_iteration": 2.5416126251220703
    },
    {
      "auxiliary_loss_clip": 0.01173483,
      "auxiliary_loss_mlp": 0.0103659,
      "balance_loss_clip": 1.06268525,
      "balance_loss_mlp": 1.02751243,
      "epoch": 0.19948295557025192,
      "flos": 40843826680320.0,
      "grad_norm": 2.0047277355743516,
      "language_loss": 0.78372967,
      "learning_rate": 3.706317366970386e-06,
      "loss": 0.80583042,
      "num_input_tokens_seen": 35165280,
      "step": 1659,
      "time_per_iteration": 3.4792778491973877
    },
    {
      "auxiliary_loss_clip": 0.01220803,
      "auxiliary_loss_mlp": 0.00763386,
      "balance_loss_clip": 1.06524277,
      "balance_loss_mlp": 1.00005984,
      "epoch": 0.199603198460891,
      "flos": 25082418620160.0,
      "grad_norm": 2.1702711357606814,
      "language_loss": 0.83417308,
      "learning_rate": 3.705910885848795e-06,
      "loss": 0.85401499,
      "num_input_tokens_seen": 35183655,
      "step": 1660,
      "time_per_iteration": 3.332139492034912
    },
    {
      "auxiliary_loss_clip": 0.01202558,
      "auxiliary_loss_mlp": 0.01036797,
      "balance_loss_clip": 1.06446171,
      "balance_loss_mlp": 1.02748656,
      "epoch": 0.19972344135153008,
      "flos": 20084120352000.0,
      "grad_norm": 2.012091285507101,
      "language_loss": 0.84191287,
      "learning_rate": 3.705504145941879e-06,
      "loss": 0.86430639,
      "num_input_tokens_seen": 35201825,
      "step": 1661,
      "time_per_iteration": 2.498910903930664
    },
    {
      "auxiliary_loss_clip": 0.01217105,
      "auxiliary_loss_mlp": 0.01041612,
      "balance_loss_clip": 1.06409335,
      "balance_loss_mlp": 1.03311491,
      "epoch": 0.1998436842421692,
      "flos": 23727472761600.0,
      "grad_norm": 1.7879300457031901,
      "language_loss": 0.79102981,
      "learning_rate": 3.7050971473113403e-06,
      "loss": 0.81361699,
      "num_input_tokens_seen": 35221600,
      "step": 1662,
      "time_per_iteration": 2.5092973709106445
    },
    {
      "auxiliary_loss_clip": 0.01199158,
      "auxiliary_loss_mlp": 0.00762087,
      "balance_loss_clip": 1.06132412,
      "balance_loss_mlp": 1.00007915,
      "epoch": 0.19996392713280828,
      "flos": 36102361633920.0,
      "grad_norm": 1.6653849155480214,
      "language_loss": 0.79995596,
      "learning_rate": 3.7046898900189196e-06,
      "loss": 0.8195684,
      "num_input_tokens_seen": 35245935,
      "step": 1663,
      "time_per_iteration": 2.7298309803009033
    },
    {
      "auxiliary_loss_clip": 0.01179859,
      "auxiliary_loss_mlp": 0.01044334,
      "balance_loss_clip": 1.06248331,
      "balance_loss_mlp": 1.03528631,
      "epoch": 0.20008417002344736,
      "flos": 23657662679040.0,
      "grad_norm": 1.8664789692871777,
      "language_loss": 0.82883871,
      "learning_rate": 3.704282374126398e-06,
      "loss": 0.85108066,
      "num_input_tokens_seen": 35265615,
      "step": 1664,
      "time_per_iteration": 2.6399664878845215
    },
    {
      "auxiliary_loss_clip": 0.01172609,
      "auxiliary_loss_mlp": 0.0104253,
      "balance_loss_clip": 1.05944324,
      "balance_loss_mlp": 1.03327298,
      "epoch": 0.20020441291408644,
      "flos": 21872076664320.0,
      "grad_norm": 1.7981491045014408,
      "language_loss": 0.87421274,
      "learning_rate": 3.7038745996955954e-06,
      "loss": 0.89636415,
      "num_input_tokens_seen": 35284960,
      "step": 1665,
      "time_per_iteration": 2.6242945194244385
    },
    {
      "auxiliary_loss_clip": 0.0117703,
      "auxiliary_loss_mlp": 0.0103785,
      "balance_loss_clip": 1.05757391,
      "balance_loss_mlp": 1.02793801,
      "epoch": 0.20032465580472555,
      "flos": 23179691376000.0,
      "grad_norm": 3.194060676773115,
      "language_loss": 0.71920693,
      "learning_rate": 3.703466566788371e-06,
      "loss": 0.74135572,
      "num_input_tokens_seen": 35304090,
      "step": 1666,
      "time_per_iteration": 2.580848217010498
    },
    {
      "auxiliary_loss_clip": 0.01179305,
      "auxiliary_loss_mlp": 0.01038413,
      "balance_loss_clip": 1.06178761,
      "balance_loss_mlp": 1.02901912,
      "epoch": 0.20044489869536464,
      "flos": 23873521461120.0,
      "grad_norm": 2.045467283379868,
      "language_loss": 0.74783653,
      "learning_rate": 3.703058275466622e-06,
      "loss": 0.77001369,
      "num_input_tokens_seen": 35323325,
      "step": 1667,
      "time_per_iteration": 2.6313648223876953
    },
    {
      "auxiliary_loss_clip": 0.01186387,
      "auxiliary_loss_mlp": 0.0103034,
      "balance_loss_clip": 1.06189454,
      "balance_loss_mlp": 1.022192,
      "epoch": 0.20056514158600372,
      "flos": 21945226711680.0,
      "grad_norm": 1.850712046730482,
      "language_loss": 0.77665269,
      "learning_rate": 3.7026497257922877e-06,
      "loss": 0.79881996,
      "num_input_tokens_seen": 35343635,
      "step": 1668,
      "time_per_iteration": 2.585545301437378
    },
    {
      "auxiliary_loss_clip": 0.01151382,
      "auxiliary_loss_mlp": 0.01037342,
      "balance_loss_clip": 1.0541929,
      "balance_loss_mlp": 1.02823484,
      "epoch": 0.20068538447664283,
      "flos": 23879159896320.0,
      "grad_norm": 1.599503241159991,
      "language_loss": 0.85068649,
      "learning_rate": 3.7022409178273436e-06,
      "loss": 0.87257373,
      "num_input_tokens_seen": 35364615,
      "step": 1669,
      "time_per_iteration": 2.6325223445892334
    },
    {
      "auxiliary_loss_clip": 0.01201321,
      "auxiliary_loss_mlp": 0.01036071,
      "balance_loss_clip": 1.06611466,
      "balance_loss_mlp": 1.02748168,
      "epoch": 0.2008056273672819,
      "flos": 18442823270400.0,
      "grad_norm": 1.9264878397785965,
      "language_loss": 0.78469884,
      "learning_rate": 3.7018318516338054e-06,
      "loss": 0.80707276,
      "num_input_tokens_seen": 35383775,
      "step": 1670,
      "time_per_iteration": 2.499721050262451
    },
    {
      "auxiliary_loss_clip": 0.01207686,
      "auxiliary_loss_mlp": 0.01036362,
      "balance_loss_clip": 1.06490457,
      "balance_loss_mlp": 1.02661657,
      "epoch": 0.200925870257921,
      "flos": 23659530186240.0,
      "grad_norm": 2.2642014318600787,
      "language_loss": 0.81522083,
      "learning_rate": 3.7014225272737284e-06,
      "loss": 0.83766127,
      "num_input_tokens_seen": 35403000,
      "step": 1671,
      "time_per_iteration": 2.5328280925750732
    },
    {
      "auxiliary_loss_clip": 0.01195347,
      "auxiliary_loss_mlp": 0.01034361,
      "balance_loss_clip": 1.06195986,
      "balance_loss_mlp": 1.02537215,
      "epoch": 0.20104611314856008,
      "flos": 16217115909120.0,
      "grad_norm": 2.3783087943848202,
      "language_loss": 0.74130112,
      "learning_rate": 3.701012944809207e-06,
      "loss": 0.7635982,
      "num_input_tokens_seen": 35420115,
      "step": 1672,
      "time_per_iteration": 2.4908220767974854
    },
    {
      "auxiliary_loss_clip": 0.0119199,
      "auxiliary_loss_mlp": 0.00762498,
      "balance_loss_clip": 1.06557107,
      "balance_loss_mlp": 1.00011671,
      "epoch": 0.2011663560391992,
      "flos": 21397373498880.0,
      "grad_norm": 1.9696943352977963,
      "language_loss": 0.78453314,
      "learning_rate": 3.700603104302374e-06,
      "loss": 0.80407798,
      "num_input_tokens_seen": 35439925,
      "step": 1673,
      "time_per_iteration": 2.5978379249572754
    },
    {
      "auxiliary_loss_clip": 0.01071792,
      "auxiliary_loss_mlp": 0.01003369,
      "balance_loss_clip": 1.03631508,
      "balance_loss_mlp": 1.00125921,
      "epoch": 0.20128659892983827,
      "flos": 62229459409920.0,
      "grad_norm": 0.896387798126695,
      "language_loss": 0.56010562,
      "learning_rate": 3.7001930058154027e-06,
      "loss": 0.58085722,
      "num_input_tokens_seen": 35504885,
      "step": 1674,
      "time_per_iteration": 3.2054691314697266
    },
    {
      "auxiliary_loss_clip": 0.01172949,
      "auxiliary_loss_mlp": 0.01045689,
      "balance_loss_clip": 1.06077921,
      "balance_loss_mlp": 1.03511548,
      "epoch": 0.20140684182047736,
      "flos": 28438737448320.0,
      "grad_norm": 2.4347814001338848,
      "language_loss": 0.79867363,
      "learning_rate": 3.6997826494105037e-06,
      "loss": 0.82085997,
      "num_input_tokens_seen": 35525330,
      "step": 1675,
      "time_per_iteration": 2.640554904937744
    },
    {
      "auxiliary_loss_clip": 0.01191701,
      "auxiliary_loss_mlp": 0.01038693,
      "balance_loss_clip": 1.06434524,
      "balance_loss_mlp": 1.03025913,
      "epoch": 0.20152708471111647,
      "flos": 28074064619520.0,
      "grad_norm": 3.1893421698910664,
      "language_loss": 0.69436586,
      "learning_rate": 3.6993720351499286e-06,
      "loss": 0.7166698,
      "num_input_tokens_seen": 35546455,
      "step": 1676,
      "time_per_iteration": 2.606886625289917
    },
    {
      "auxiliary_loss_clip": 0.01183174,
      "auxiliary_loss_mlp": 0.01033745,
      "balance_loss_clip": 1.06399202,
      "balance_loss_mlp": 1.02531099,
      "epoch": 0.20164732760175555,
      "flos": 23549751244800.0,
      "grad_norm": 2.079768944813036,
      "language_loss": 0.77067965,
      "learning_rate": 3.6989611630959666e-06,
      "loss": 0.79284883,
      "num_input_tokens_seen": 35565010,
      "step": 1677,
      "time_per_iteration": 2.5936663150787354
    },
    {
      "auxiliary_loss_clip": 0.01094814,
      "auxiliary_loss_mlp": 0.01003755,
      "balance_loss_clip": 1.02603412,
      "balance_loss_mlp": 1.00166845,
      "epoch": 0.20176757049239463,
      "flos": 71100616037760.0,
      "grad_norm": 2.2544613137243386,
      "language_loss": 0.58294541,
      "learning_rate": 3.6985500333109474e-06,
      "loss": 0.60393101,
      "num_input_tokens_seen": 35633340,
      "step": 1678,
      "time_per_iteration": 3.1958117485046387
    },
    {
      "auxiliary_loss_clip": 0.01164829,
      "auxiliary_loss_mlp": 0.01033264,
      "balance_loss_clip": 1.05878007,
      "balance_loss_mlp": 1.02409649,
      "epoch": 0.20188781338303372,
      "flos": 21430159637760.0,
      "grad_norm": 2.454360213496597,
      "language_loss": 0.76511395,
      "learning_rate": 3.6981386458572385e-06,
      "loss": 0.78709489,
      "num_input_tokens_seen": 35651315,
      "step": 1679,
      "time_per_iteration": 3.439234972000122
    },
    {
      "auxiliary_loss_clip": 0.01170543,
      "auxiliary_loss_mlp": 0.01040399,
      "balance_loss_clip": 1.05943096,
      "balance_loss_mlp": 1.03136325,
      "epoch": 0.20200805627367283,
      "flos": 11546215130880.0,
      "grad_norm": 2.2694366270189676,
      "language_loss": 0.76112741,
      "learning_rate": 3.6977270007972468e-06,
      "loss": 0.78323686,
      "num_input_tokens_seen": 35668850,
      "step": 1680,
      "time_per_iteration": 2.595019578933716
    },
    {
      "auxiliary_loss_clip": 0.01194563,
      "auxiliary_loss_mlp": 0.01033311,
      "balance_loss_clip": 1.06538737,
      "balance_loss_mlp": 1.02456117,
      "epoch": 0.2021282991643119,
      "flos": 28545391906560.0,
      "grad_norm": 2.291652527149069,
      "language_loss": 0.72189021,
      "learning_rate": 3.6973150981934196e-06,
      "loss": 0.74416888,
      "num_input_tokens_seen": 35690080,
      "step": 1681,
      "time_per_iteration": 3.3767318725585938
    },
    {
      "auxiliary_loss_clip": 0.01222159,
      "auxiliary_loss_mlp": 0.01032866,
      "balance_loss_clip": 1.06777084,
      "balance_loss_mlp": 1.02334762,
      "epoch": 0.202248542054951,
      "flos": 17923446564480.0,
      "grad_norm": 2.5265905680527463,
      "language_loss": 0.841084,
      "learning_rate": 3.6969029381082415e-06,
      "loss": 0.86363423,
      "num_input_tokens_seen": 35706075,
      "step": 1682,
      "time_per_iteration": 2.476741075515747
    },
    {
      "auxiliary_loss_clip": 0.01183947,
      "auxiliary_loss_mlp": 0.01031562,
      "balance_loss_clip": 1.06201458,
      "balance_loss_mlp": 1.02333033,
      "epoch": 0.2023687849455901,
      "flos": 19864634296320.0,
      "grad_norm": 1.9024268349402147,
      "language_loss": 0.79715955,
      "learning_rate": 3.696490520604237e-06,
      "loss": 0.8193146,
      "num_input_tokens_seen": 35724765,
      "step": 1683,
      "time_per_iteration": 2.5574145317077637
    },
    {
      "auxiliary_loss_clip": 0.01202508,
      "auxiliary_loss_mlp": 0.01038058,
      "balance_loss_clip": 1.06616604,
      "balance_loss_mlp": 1.02891493,
      "epoch": 0.20248902783622919,
      "flos": 22564721600640.0,
      "grad_norm": 1.6811910279075426,
      "language_loss": 0.8034243,
      "learning_rate": 3.696077845743968e-06,
      "loss": 0.82582998,
      "num_input_tokens_seen": 35744355,
      "step": 1684,
      "time_per_iteration": 2.5642013549804688
    },
    {
      "auxiliary_loss_clip": 0.01218386,
      "auxiliary_loss_mlp": 0.0104165,
      "balance_loss_clip": 1.06371903,
      "balance_loss_mlp": 1.03198767,
      "epoch": 0.20260927072686827,
      "flos": 22709728805760.0,
      "grad_norm": 2.7136734623903305,
      "language_loss": 0.73247087,
      "learning_rate": 3.69566491359004e-06,
      "loss": 0.75507116,
      "num_input_tokens_seen": 35761000,
      "step": 1685,
      "time_per_iteration": 3.2501134872436523
    },
    {
      "auxiliary_loss_clip": 0.01184917,
      "auxiliary_loss_mlp": 0.01040207,
      "balance_loss_clip": 1.05978477,
      "balance_loss_mlp": 1.03091502,
      "epoch": 0.20272951361750738,
      "flos": 51023998650240.0,
      "grad_norm": 2.2475290851518004,
      "language_loss": 0.69115651,
      "learning_rate": 3.695251724205092e-06,
      "loss": 0.71340775,
      "num_input_tokens_seen": 35785360,
      "step": 1686,
      "time_per_iteration": 3.58396577835083
    },
    {
      "auxiliary_loss_clip": 0.01218164,
      "auxiliary_loss_mlp": 0.01035276,
      "balance_loss_clip": 1.06554055,
      "balance_loss_mlp": 1.02623415,
      "epoch": 0.20284975650814646,
      "flos": 26578133879040.0,
      "grad_norm": 1.586489494444149,
      "language_loss": 0.86374056,
      "learning_rate": 3.6948382776518054e-06,
      "loss": 0.88627493,
      "num_input_tokens_seen": 35806065,
      "step": 1687,
      "time_per_iteration": 2.5361804962158203
    },
    {
      "auxiliary_loss_clip": 0.01178886,
      "auxiliary_loss_mlp": 0.01036376,
      "balance_loss_clip": 1.05820012,
      "balance_loss_mlp": 1.02660716,
      "epoch": 0.20296999939878554,
      "flos": 16034222833920.0,
      "grad_norm": 2.5735093798449413,
      "language_loss": 0.79383409,
      "learning_rate": 3.6944245739929e-06,
      "loss": 0.81598669,
      "num_input_tokens_seen": 35822225,
      "step": 1688,
      "time_per_iteration": 2.5313823223114014
    },
    {
      "auxiliary_loss_clip": 0.01205549,
      "auxiliary_loss_mlp": 0.01038811,
      "balance_loss_clip": 1.06417179,
      "balance_loss_mlp": 1.0289706,
      "epoch": 0.20309024228942463,
      "flos": 19203374868480.0,
      "grad_norm": 3.480369651290209,
      "language_loss": 0.72124946,
      "learning_rate": 3.6940106132911332e-06,
      "loss": 0.74369305,
      "num_input_tokens_seen": 35839410,
      "step": 1689,
      "time_per_iteration": 2.5325663089752197
    },
    {
      "auxiliary_loss_clip": 0.01201738,
      "auxiliary_loss_mlp": 0.0103291,
      "balance_loss_clip": 1.06228733,
      "balance_loss_mlp": 1.02374279,
      "epoch": 0.20321048518006374,
      "flos": 22821087945600.0,
      "grad_norm": 2.321356270082823,
      "language_loss": 0.88802993,
      "learning_rate": 3.6935963956093037e-06,
      "loss": 0.91037643,
      "num_input_tokens_seen": 35859495,
      "step": 1690,
      "time_per_iteration": 2.530231475830078
    },
    {
      "auxiliary_loss_clip": 0.01196941,
      "auxiliary_loss_mlp": 0.01042653,
      "balance_loss_clip": 1.06392837,
      "balance_loss_mlp": 1.03347969,
      "epoch": 0.20333072807070282,
      "flos": 19096397187840.0,
      "grad_norm": 1.9527062373056907,
      "language_loss": 0.68870354,
      "learning_rate": 3.6931819210102474e-06,
      "loss": 0.71109951,
      "num_input_tokens_seen": 35878890,
      "step": 1691,
      "time_per_iteration": 2.530916452407837
    },
    {
      "auxiliary_loss_clip": 0.01218143,
      "auxiliary_loss_mlp": 0.01031253,
      "balance_loss_clip": 1.06427014,
      "balance_loss_mlp": 1.02219915,
      "epoch": 0.2034509709613419,
      "flos": 18180962144640.0,
      "grad_norm": 1.7755316371720002,
      "language_loss": 0.84366202,
      "learning_rate": 3.6927671895568402e-06,
      "loss": 0.86615598,
      "num_input_tokens_seen": 35897950,
      "step": 1692,
      "time_per_iteration": 2.4939329624176025
    },
    {
      "auxiliary_loss_clip": 0.01223264,
      "auxiliary_loss_mlp": 0.01042803,
      "balance_loss_clip": 1.06919193,
      "balance_loss_mlp": 1.03336775,
      "epoch": 0.20357121385198101,
      "flos": 22923899648640.0,
      "grad_norm": 2.370514514673758,
      "language_loss": 0.87054586,
      "learning_rate": 3.692352201311996e-06,
      "loss": 0.89320648,
      "num_input_tokens_seen": 35916800,
      "step": 1693,
      "time_per_iteration": 2.477813959121704
    },
    {
      "auxiliary_loss_clip": 0.01165065,
      "auxiliary_loss_mlp": 0.01034243,
      "balance_loss_clip": 1.05643845,
      "balance_loss_mlp": 1.02477789,
      "epoch": 0.2036914567426201,
      "flos": 20922131629440.0,
      "grad_norm": 2.111977701821932,
      "language_loss": 0.76861882,
      "learning_rate": 3.6919369563386687e-06,
      "loss": 0.79061186,
      "num_input_tokens_seen": 35936600,
      "step": 1694,
      "time_per_iteration": 2.581807851791382
    },
    {
      "auxiliary_loss_clip": 0.01183342,
      "auxiliary_loss_mlp": 0.01041396,
      "balance_loss_clip": 1.06000626,
      "balance_loss_mlp": 1.03247857,
      "epoch": 0.20381169963325918,
      "flos": 15519155760000.0,
      "grad_norm": 3.1819032500554103,
      "language_loss": 0.78757375,
      "learning_rate": 3.69152145469985e-06,
      "loss": 0.80982107,
      "num_input_tokens_seen": 35953645,
      "step": 1695,
      "time_per_iteration": 2.5328786373138428
    },
    {
      "auxiliary_loss_clip": 0.01158971,
      "auxiliary_loss_mlp": 0.01036207,
      "balance_loss_clip": 1.05532157,
      "balance_loss_mlp": 1.02555537,
      "epoch": 0.20393194252389826,
      "flos": 28833143760000.0,
      "grad_norm": 1.9443057097346088,
      "language_loss": 0.81745565,
      "learning_rate": 3.691105696458572e-06,
      "loss": 0.83940738,
      "num_input_tokens_seen": 35970940,
      "step": 1696,
      "time_per_iteration": 2.6566383838653564
    },
    {
      "auxiliary_loss_clip": 0.01215741,
      "auxiliary_loss_mlp": 0.01032664,
      "balance_loss_clip": 1.06489623,
      "balance_loss_mlp": 1.02411664,
      "epoch": 0.20405218541453737,
      "flos": 22488554810880.0,
      "grad_norm": 2.5136253715035504,
      "language_loss": 0.68357927,
      "learning_rate": 3.690689681677904e-06,
      "loss": 0.70606333,
      "num_input_tokens_seen": 35989410,
      "step": 1697,
      "time_per_iteration": 2.5045814514160156
    },
    {
      "auxiliary_loss_clip": 0.01187778,
      "auxiliary_loss_mlp": 0.0103293,
      "balance_loss_clip": 1.05696249,
      "balance_loss_mlp": 1.02413237,
      "epoch": 0.20417242830517646,
      "flos": 25374408278400.0,
      "grad_norm": 1.7875445082910515,
      "language_loss": 0.88970387,
      "learning_rate": 3.690273410420956e-06,
      "loss": 0.91191089,
      "num_input_tokens_seen": 36009175,
      "step": 1698,
      "time_per_iteration": 2.5803349018096924
    },
    {
      "auxiliary_loss_clip": 0.0119846,
      "auxiliary_loss_mlp": 0.01043654,
      "balance_loss_clip": 1.06015813,
      "balance_loss_mlp": 1.03439784,
      "epoch": 0.20429267119581554,
      "flos": 14793078240000.0,
      "grad_norm": 2.260683465050588,
      "language_loss": 0.76718646,
      "learning_rate": 3.689856882750875e-06,
      "loss": 0.78960758,
      "num_input_tokens_seen": 36024375,
      "step": 1699,
      "time_per_iteration": 2.4736506938934326
    },
    {
      "auxiliary_loss_clip": 0.01199796,
      "auxiliary_loss_mlp": 0.01034893,
      "balance_loss_clip": 1.06316304,
      "balance_loss_mlp": 1.02629769,
      "epoch": 0.20441291408645465,
      "flos": 17781851151360.0,
      "grad_norm": 2.0535151041228574,
      "language_loss": 0.7834391,
      "learning_rate": 3.6894400987308486e-06,
      "loss": 0.80578595,
      "num_input_tokens_seen": 36041895,
      "step": 1700,
      "time_per_iteration": 2.4950199127197266
    },
    {
      "auxiliary_loss_clip": 0.01203361,
      "auxiliary_loss_mlp": 0.01036821,
      "balance_loss_clip": 1.06138933,
      "balance_loss_mlp": 1.02786279,
      "epoch": 0.20453315697709373,
      "flos": 16435668211200.0,
      "grad_norm": 1.9975758842294746,
      "language_loss": 0.84788382,
      "learning_rate": 3.6890230584241024e-06,
      "loss": 0.87028563,
      "num_input_tokens_seen": 36058825,
      "step": 1701,
      "time_per_iteration": 2.5253922939300537
    },
    {
      "auxiliary_loss_clip": 0.01106309,
      "auxiliary_loss_mlp": 0.0100488,
      "balance_loss_clip": 1.02346671,
      "balance_loss_mlp": 1.00254393,
      "epoch": 0.20465339986773282,
      "flos": 66713085653760.0,
      "grad_norm": 1.0356248958730427,
      "language_loss": 0.66446781,
      "learning_rate": 3.6886057618939016e-06,
      "loss": 0.68557978,
      "num_input_tokens_seen": 36121645,
      "step": 1702,
      "time_per_iteration": 3.1514835357666016
    },
    {
      "auxiliary_loss_clip": 0.01163414,
      "auxiliary_loss_mlp": 0.01038612,
      "balance_loss_clip": 1.05545211,
      "balance_loss_mlp": 1.02843714,
      "epoch": 0.2047736427583719,
      "flos": 41974114924800.0,
      "grad_norm": 2.0031599343588726,
      "language_loss": 0.69369829,
      "learning_rate": 3.6881882092035492e-06,
      "loss": 0.71571851,
      "num_input_tokens_seen": 36143030,
      "step": 1703,
      "time_per_iteration": 2.730560779571533
    },
    {
      "auxiliary_loss_clip": 0.01076705,
      "auxiliary_loss_mlp": 0.00753001,
      "balance_loss_clip": 1.02303708,
      "balance_loss_mlp": 1.00019574,
      "epoch": 0.204893885649011,
      "flos": 69940878641280.0,
      "grad_norm": 0.9263532792700441,
      "language_loss": 0.61256588,
      "learning_rate": 3.6877704004163873e-06,
      "loss": 0.63086295,
      "num_input_tokens_seen": 36203435,
      "step": 1704,
      "time_per_iteration": 3.3109655380249023
    },
    {
      "auxiliary_loss_clip": 0.01218918,
      "auxiliary_loss_mlp": 0.01040054,
      "balance_loss_clip": 1.06384122,
      "balance_loss_mlp": 1.03009963,
      "epoch": 0.2050141285396501,
      "flos": 22200012858240.0,
      "grad_norm": 2.9563300445726863,
      "language_loss": 0.77848095,
      "learning_rate": 3.6873523355957984e-06,
      "loss": 0.80107069,
      "num_input_tokens_seen": 36222435,
      "step": 1705,
      "time_per_iteration": 3.2323403358459473
    },
    {
      "auxiliary_loss_clip": 0.01102912,
      "auxiliary_loss_mlp": 0.01006345,
      "balance_loss_clip": 1.02033496,
      "balance_loss_mlp": 1.00409162,
      "epoch": 0.20513437143028918,
      "flos": 46283721730560.0,
      "grad_norm": 0.9818722437470498,
      "language_loss": 0.64074767,
      "learning_rate": 3.686934014805201e-06,
      "loss": 0.66184032,
      "num_input_tokens_seen": 36273065,
      "step": 1706,
      "time_per_iteration": 2.9311726093292236
    },
    {
      "auxiliary_loss_clip": 0.01199893,
      "auxiliary_loss_mlp": 0.01043211,
      "balance_loss_clip": 1.06301522,
      "balance_loss_mlp": 1.03378129,
      "epoch": 0.20525461432092829,
      "flos": 21904324099200.0,
      "grad_norm": 1.8631498404291655,
      "language_loss": 0.80870032,
      "learning_rate": 3.6865154381080552e-06,
      "loss": 0.83113134,
      "num_input_tokens_seen": 36293750,
      "step": 1707,
      "time_per_iteration": 2.5563607215881348
    },
    {
      "auxiliary_loss_clip": 0.01126515,
      "auxiliary_loss_mlp": 0.01040634,
      "balance_loss_clip": 1.05098343,
      "balance_loss_mlp": 1.03159153,
      "epoch": 0.20537485721156737,
      "flos": 21214264942080.0,
      "grad_norm": 3.1428348155918586,
      "language_loss": 0.82789433,
      "learning_rate": 3.6860966055678585e-06,
      "loss": 0.84956574,
      "num_input_tokens_seen": 36310105,
      "step": 1708,
      "time_per_iteration": 3.4999380111694336
    },
    {
      "auxiliary_loss_clip": 0.0120392,
      "auxiliary_loss_mlp": 0.01036333,
      "balance_loss_clip": 1.06402504,
      "balance_loss_mlp": 1.02706456,
      "epoch": 0.20549510010220645,
      "flos": 20191205773440.0,
      "grad_norm": 1.8090437184010248,
      "language_loss": 0.86228991,
      "learning_rate": 3.685677517248147e-06,
      "loss": 0.88469243,
      "num_input_tokens_seen": 36328995,
      "step": 1709,
      "time_per_iteration": 2.6079885959625244
    },
    {
      "auxiliary_loss_clip": 0.01187798,
      "auxiliary_loss_mlp": 0.00762274,
      "balance_loss_clip": 1.06430948,
      "balance_loss_mlp": 1.00022507,
      "epoch": 0.20561534299284553,
      "flos": 17016702612480.0,
      "grad_norm": 1.8883669939319985,
      "language_loss": 0.80434602,
      "learning_rate": 3.6852581732124967e-06,
      "loss": 0.82384682,
      "num_input_tokens_seen": 36346340,
      "step": 1710,
      "time_per_iteration": 2.522298574447632
    },
    {
      "auxiliary_loss_clip": 0.01203474,
      "auxiliary_loss_mlp": 0.01042497,
      "balance_loss_clip": 1.06396461,
      "balance_loss_mlp": 1.03261411,
      "epoch": 0.20573558588348465,
      "flos": 22890467064960.0,
      "grad_norm": 2.136583206972171,
      "language_loss": 0.76482683,
      "learning_rate": 3.6848385735245213e-06,
      "loss": 0.78728658,
      "num_input_tokens_seen": 36365430,
      "step": 1711,
      "time_per_iteration": 3.4694912433624268
    },
    {
      "auxiliary_loss_clip": 0.01183031,
      "auxiliary_loss_mlp": 0.01033961,
      "balance_loss_clip": 1.05427146,
      "balance_loss_mlp": 1.02519298,
      "epoch": 0.20585582877412373,
      "flos": 24643123286400.0,
      "grad_norm": 1.7498946892310732,
      "language_loss": 0.86333978,
      "learning_rate": 3.6844187182478734e-06,
      "loss": 0.88550973,
      "num_input_tokens_seen": 36386285,
      "step": 1712,
      "time_per_iteration": 3.3554019927978516
    },
    {
      "auxiliary_loss_clip": 0.01174061,
      "auxiliary_loss_mlp": 0.01039323,
      "balance_loss_clip": 1.0572331,
      "balance_loss_mlp": 1.0301497,
      "epoch": 0.2059760716647628,
      "flos": 24206952435840.0,
      "grad_norm": 2.0531090594482104,
      "language_loss": 0.75014567,
      "learning_rate": 3.683998607446246e-06,
      "loss": 0.7722795,
      "num_input_tokens_seen": 36404935,
      "step": 1713,
      "time_per_iteration": 2.5609142780303955
    },
    {
      "auxiliary_loss_clip": 0.01203878,
      "auxiliary_loss_mlp": 0.0104051,
      "balance_loss_clip": 1.06466377,
      "balance_loss_mlp": 1.03117621,
      "epoch": 0.20609631455540192,
      "flos": 20229522606720.0,
      "grad_norm": 1.9307027911850765,
      "language_loss": 0.74819231,
      "learning_rate": 3.6835782411833686e-06,
      "loss": 0.77063614,
      "num_input_tokens_seen": 36424455,
      "step": 1714,
      "time_per_iteration": 2.5156447887420654
    },
    {
      "auxiliary_loss_clip": 0.01158779,
      "auxiliary_loss_mlp": 0.01033332,
      "balance_loss_clip": 1.05409908,
      "balance_loss_mlp": 1.024755,
      "epoch": 0.206216557446041,
      "flos": 19864957518720.0,
      "grad_norm": 1.83381027900753,
      "language_loss": 0.74186343,
      "learning_rate": 3.68315761952301e-06,
      "loss": 0.76378453,
      "num_input_tokens_seen": 36441685,
      "step": 1715,
      "time_per_iteration": 2.5801353454589844
    },
    {
      "auxiliary_loss_clip": 0.01214273,
      "auxiliary_loss_mlp": 0.01035494,
      "balance_loss_clip": 1.06185961,
      "balance_loss_mlp": 1.02588546,
      "epoch": 0.2063368003366801,
      "flos": 24096311568000.0,
      "grad_norm": 1.9485976866747476,
      "language_loss": 0.82769096,
      "learning_rate": 3.6827367425289797e-06,
      "loss": 0.85018861,
      "num_input_tokens_seen": 36461460,
      "step": 1716,
      "time_per_iteration": 2.518188714981079
    },
    {
      "auxiliary_loss_clip": 0.01185074,
      "auxiliary_loss_mlp": 0.01042457,
      "balance_loss_clip": 1.06014419,
      "balance_loss_mlp": 1.03264034,
      "epoch": 0.2064570432273192,
      "flos": 20340163474560.0,
      "grad_norm": 2.175348423475904,
      "language_loss": 0.72804713,
      "learning_rate": 3.6823156102651225e-06,
      "loss": 0.75032246,
      "num_input_tokens_seen": 36479615,
      "step": 1717,
      "time_per_iteration": 2.5415356159210205
    },
    {
      "auxiliary_loss_clip": 0.01133626,
      "auxiliary_loss_mlp": 0.0103265,
      "balance_loss_clip": 1.05872178,
      "balance_loss_mlp": 1.02353668,
      "epoch": 0.20657728611795828,
      "flos": 20520363029760.0,
      "grad_norm": 1.7721069288658966,
      "language_loss": 0.71065778,
      "learning_rate": 3.6818942227953257e-06,
      "loss": 0.73232049,
      "num_input_tokens_seen": 36500160,
      "step": 1718,
      "time_per_iteration": 2.662010669708252
    },
    {
      "auxiliary_loss_clip": 0.01169801,
      "auxiliary_loss_mlp": 0.01036298,
      "balance_loss_clip": 1.05710077,
      "balance_loss_mlp": 1.02729201,
      "epoch": 0.20669752900859736,
      "flos": 21799285752960.0,
      "grad_norm": 2.824139918896068,
      "language_loss": 0.6921221,
      "learning_rate": 3.681472580183512e-06,
      "loss": 0.71418309,
      "num_input_tokens_seen": 36518810,
      "step": 1719,
      "time_per_iteration": 2.5904173851013184
    },
    {
      "auxiliary_loss_clip": 0.01196122,
      "auxiliary_loss_mlp": 0.01032823,
      "balance_loss_clip": 1.06124151,
      "balance_loss_mlp": 1.02428722,
      "epoch": 0.20681777189923645,
      "flos": 15122020014720.0,
      "grad_norm": 1.895962039055488,
      "language_loss": 0.86372888,
      "learning_rate": 3.6810506824936455e-06,
      "loss": 0.88601828,
      "num_input_tokens_seen": 36536890,
      "step": 1720,
      "time_per_iteration": 2.5047531127929688
    },
    {
      "auxiliary_loss_clip": 0.01078519,
      "auxiliary_loss_mlp": 0.01002905,
      "balance_loss_clip": 1.01857162,
      "balance_loss_mlp": 1.00084293,
      "epoch": 0.20693801478987556,
      "flos": 56481021509760.0,
      "grad_norm": 1.0545596526174006,
      "language_loss": 0.62560916,
      "learning_rate": 3.680628529789726e-06,
      "loss": 0.64642334,
      "num_input_tokens_seen": 36589300,
      "step": 1721,
      "time_per_iteration": 2.978342294692993
    },
    {
      "auxiliary_loss_clip": 0.01221991,
      "auxiliary_loss_mlp": 0.01047492,
      "balance_loss_clip": 1.06564212,
      "balance_loss_mlp": 1.03806221,
      "epoch": 0.20705825768051464,
      "flos": 21614201948160.0,
      "grad_norm": 2.377839410064548,
      "language_loss": 0.86219734,
      "learning_rate": 3.680206122135796e-06,
      "loss": 0.88489223,
      "num_input_tokens_seen": 36609905,
      "step": 1722,
      "time_per_iteration": 2.5234663486480713
    },
    {
      "auxiliary_loss_clip": 0.01161579,
      "auxiliary_loss_mlp": 0.01038923,
      "balance_loss_clip": 1.05980933,
      "balance_loss_mlp": 1.03054905,
      "epoch": 0.20717850057115372,
      "flos": 25848895962240.0,
      "grad_norm": 2.078232943449261,
      "language_loss": 0.78592545,
      "learning_rate": 3.6797834595959323e-06,
      "loss": 0.80793047,
      "num_input_tokens_seen": 36629805,
      "step": 1723,
      "time_per_iteration": 2.6566221714019775
    },
    {
      "auxiliary_loss_clip": 0.01147234,
      "auxiliary_loss_mlp": 0.01034499,
      "balance_loss_clip": 1.05285668,
      "balance_loss_mlp": 1.02524281,
      "epoch": 0.20729874346179283,
      "flos": 29130807767040.0,
      "grad_norm": 3.6089929127875875,
      "language_loss": 0.77825701,
      "learning_rate": 3.679360542234254e-06,
      "loss": 0.80007434,
      "num_input_tokens_seen": 36649150,
      "step": 1724,
      "time_per_iteration": 2.6744298934936523
    },
    {
      "auxiliary_loss_clip": 0.01177262,
      "auxiliary_loss_mlp": 0.00762241,
      "balance_loss_clip": 1.05568838,
      "balance_loss_mlp": 1.00030398,
      "epoch": 0.20741898635243192,
      "flos": 29023363209600.0,
      "grad_norm": 1.618609591984291,
      "language_loss": 0.72196198,
      "learning_rate": 3.678937370114916e-06,
      "loss": 0.74135697,
      "num_input_tokens_seen": 36668955,
      "step": 1725,
      "time_per_iteration": 2.621530771255493
    },
    {
      "auxiliary_loss_clip": 0.01179385,
      "auxiliary_loss_mlp": 0.01040664,
      "balance_loss_clip": 1.06071126,
      "balance_loss_mlp": 1.03249204,
      "epoch": 0.207539229243071,
      "flos": 15559447841280.0,
      "grad_norm": 1.9229893735302308,
      "language_loss": 0.78967357,
      "learning_rate": 3.678513943302114e-06,
      "loss": 0.81187403,
      "num_input_tokens_seen": 36685730,
      "step": 1726,
      "time_per_iteration": 2.523022174835205
    },
    {
      "auxiliary_loss_clip": 0.01212903,
      "auxiliary_loss_mlp": 0.0104172,
      "balance_loss_clip": 1.06368637,
      "balance_loss_mlp": 1.03290391,
      "epoch": 0.20765947213371008,
      "flos": 20521081301760.0,
      "grad_norm": 1.6969447698909321,
      "language_loss": 0.85079837,
      "learning_rate": 3.678090261860082e-06,
      "loss": 0.8733446,
      "num_input_tokens_seen": 36705460,
      "step": 1727,
      "time_per_iteration": 2.4954237937927246
    },
    {
      "auxiliary_loss_clip": 0.01171217,
      "auxiliary_loss_mlp": 0.01035832,
      "balance_loss_clip": 1.05476785,
      "balance_loss_mlp": 1.02708769,
      "epoch": 0.2077797150243492,
      "flos": 19354415558400.0,
      "grad_norm": 1.8891067464542552,
      "language_loss": 0.77366281,
      "learning_rate": 3.6776663258530906e-06,
      "loss": 0.79573333,
      "num_input_tokens_seen": 36724110,
      "step": 1728,
      "time_per_iteration": 2.6096787452697754
    },
    {
      "auxiliary_loss_clip": 0.01201233,
      "auxiliary_loss_mlp": 0.0103551,
      "balance_loss_clip": 1.06044686,
      "balance_loss_mlp": 1.02685523,
      "epoch": 0.20789995791498828,
      "flos": 21829952989440.0,
      "grad_norm": 2.0466398502185243,
      "language_loss": 0.71261871,
      "learning_rate": 3.6772421353454516e-06,
      "loss": 0.73498607,
      "num_input_tokens_seen": 36742705,
      "step": 1729,
      "time_per_iteration": 2.512300968170166
    },
    {
      "auxiliary_loss_clip": 0.01199164,
      "auxiliary_loss_mlp": 0.01037274,
      "balance_loss_clip": 1.06557655,
      "balance_loss_mlp": 1.02787423,
      "epoch": 0.20802020080562736,
      "flos": 23148844571520.0,
      "grad_norm": 1.8845983484778162,
      "language_loss": 0.88554311,
      "learning_rate": 3.6768176904015153e-06,
      "loss": 0.90790749,
      "num_input_tokens_seen": 36762510,
      "step": 1730,
      "time_per_iteration": 2.5216846466064453
    },
    {
      "auxiliary_loss_clip": 0.01200737,
      "auxiliary_loss_mlp": 0.0103806,
      "balance_loss_clip": 1.06084418,
      "balance_loss_mlp": 1.02941179,
      "epoch": 0.20814044369626647,
      "flos": 23072677781760.0,
      "grad_norm": 1.9541519116114452,
      "language_loss": 0.60291141,
      "learning_rate": 3.6763929910856674e-06,
      "loss": 0.62529945,
      "num_input_tokens_seen": 36780960,
      "step": 1731,
      "time_per_iteration": 3.3664777278900146
    },
    {
      "auxiliary_loss_clip": 0.01197779,
      "auxiliary_loss_mlp": 0.0103823,
      "balance_loss_clip": 1.06388271,
      "balance_loss_mlp": 1.02918792,
      "epoch": 0.20826068658690555,
      "flos": 19608016556160.0,
      "grad_norm": 2.341017024782291,
      "language_loss": 0.77822649,
      "learning_rate": 3.6759680374623365e-06,
      "loss": 0.80058658,
      "num_input_tokens_seen": 36798875,
      "step": 1732,
      "time_per_iteration": 2.495476484298706
    },
    {
      "auxiliary_loss_clip": 0.01211085,
      "auxiliary_loss_mlp": 0.01036754,
      "balance_loss_clip": 1.06361043,
      "balance_loss_mlp": 1.02725291,
      "epoch": 0.20838092947754464,
      "flos": 25374049142400.0,
      "grad_norm": 2.580660846124239,
      "language_loss": 0.7550745,
      "learning_rate": 3.675542829595986e-06,
      "loss": 0.77755284,
      "num_input_tokens_seen": 36818540,
      "step": 1733,
      "time_per_iteration": 2.5246336460113525
    },
    {
      "auxiliary_loss_clip": 0.01186363,
      "auxiliary_loss_mlp": 0.01031412,
      "balance_loss_clip": 1.0611403,
      "balance_loss_mlp": 1.02273965,
      "epoch": 0.20850117236818372,
      "flos": 24061729749120.0,
      "grad_norm": 1.5822633034063123,
      "language_loss": 0.79310966,
      "learning_rate": 3.6751173675511213e-06,
      "loss": 0.81528735,
      "num_input_tokens_seen": 36840585,
      "step": 1734,
      "time_per_iteration": 3.3607327938079834
    },
    {
      "auxiliary_loss_clip": 0.01180736,
      "auxiliary_loss_mlp": 0.01034402,
      "balance_loss_clip": 1.05496752,
      "balance_loss_mlp": 1.02546144,
      "epoch": 0.20862141525882283,
      "flos": 20077799558400.0,
      "grad_norm": 2.0361543056978677,
      "language_loss": 0.87288034,
      "learning_rate": 3.674691651392283e-06,
      "loss": 0.89503175,
      "num_input_tokens_seen": 36858255,
      "step": 1735,
      "time_per_iteration": 2.542043447494507
    },
    {
      "auxiliary_loss_clip": 0.01189124,
      "auxiliary_loss_mlp": 0.01032989,
      "balance_loss_clip": 1.06147313,
      "balance_loss_mlp": 1.02425122,
      "epoch": 0.2087416581494619,
      "flos": 39015183237120.0,
      "grad_norm": 2.122224133722387,
      "language_loss": 0.76095772,
      "learning_rate": 3.674265681184053e-06,
      "loss": 0.78317893,
      "num_input_tokens_seen": 36881515,
      "step": 1736,
      "time_per_iteration": 2.6891977787017822
    },
    {
      "auxiliary_loss_clip": 0.01184639,
      "auxiliary_loss_mlp": 0.01038714,
      "balance_loss_clip": 1.05818093,
      "balance_loss_mlp": 1.02988672,
      "epoch": 0.208861901040101,
      "flos": 26101994169600.0,
      "grad_norm": 1.933178039622886,
      "language_loss": 0.86454254,
      "learning_rate": 3.6738394569910504e-06,
      "loss": 0.88677609,
      "num_input_tokens_seen": 36902055,
      "step": 1737,
      "time_per_iteration": 3.3399436473846436
    },
    {
      "auxiliary_loss_clip": 0.01196359,
      "auxiliary_loss_mlp": 0.01046046,
      "balance_loss_clip": 1.06164217,
      "balance_loss_mlp": 1.03602087,
      "epoch": 0.2089821439307401,
      "flos": 28398732675840.0,
      "grad_norm": 2.0957394326222687,
      "language_loss": 0.82614797,
      "learning_rate": 3.6734129788779333e-06,
      "loss": 0.84857202,
      "num_input_tokens_seen": 36921230,
      "step": 1738,
      "time_per_iteration": 3.2663323879241943
    },
    {
      "auxiliary_loss_clip": 0.01169695,
      "auxiliary_loss_mlp": 0.01032764,
      "balance_loss_clip": 1.06135154,
      "balance_loss_mlp": 1.02441347,
      "epoch": 0.2091023868213792,
      "flos": 21069616872960.0,
      "grad_norm": 3.225955623691178,
      "language_loss": 0.90261054,
      "learning_rate": 3.6729862469093976e-06,
      "loss": 0.92463511,
      "num_input_tokens_seen": 36940325,
      "step": 1739,
      "time_per_iteration": 2.5885255336761475
    },
    {
      "auxiliary_loss_clip": 0.01171235,
      "auxiliary_loss_mlp": 0.0102762,
      "balance_loss_clip": 1.05731225,
      "balance_loss_mlp": 1.01937056,
      "epoch": 0.20922262971201827,
      "flos": 22455481363200.0,
      "grad_norm": 2.1544044735113683,
      "language_loss": 0.82850415,
      "learning_rate": 3.6725592611501782e-06,
      "loss": 0.85049272,
      "num_input_tokens_seen": 36959000,
      "step": 1740,
      "time_per_iteration": 2.5705885887145996
    },
    {
      "auxiliary_loss_clip": 0.01199459,
      "auxiliary_loss_mlp": 0.01032349,
      "balance_loss_clip": 1.06151414,
      "balance_loss_mlp": 1.02327132,
      "epoch": 0.20934287260265738,
      "flos": 27852244179840.0,
      "grad_norm": 2.104087499010352,
      "language_loss": 0.7617951,
      "learning_rate": 3.6721320216650496e-06,
      "loss": 0.78411317,
      "num_input_tokens_seen": 36979615,
      "step": 1741,
      "time_per_iteration": 2.5975582599639893
    },
    {
      "auxiliary_loss_clip": 0.01181877,
      "auxiliary_loss_mlp": 0.01033943,
      "balance_loss_clip": 1.0590353,
      "balance_loss_mlp": 1.0250138,
      "epoch": 0.20946311549329646,
      "flos": 16435309075200.0,
      "grad_norm": 1.9347283012239143,
      "language_loss": 0.83501828,
      "learning_rate": 3.6717045285188215e-06,
      "loss": 0.85717648,
      "num_input_tokens_seen": 36997310,
      "step": 1742,
      "time_per_iteration": 2.606879711151123
    },
    {
      "auxiliary_loss_clip": 0.01137058,
      "auxiliary_loss_mlp": 0.01038255,
      "balance_loss_clip": 1.0484798,
      "balance_loss_mlp": 1.02891529,
      "epoch": 0.20958335838393555,
      "flos": 22492720788480.0,
      "grad_norm": 2.113129818075374,
      "language_loss": 0.86752963,
      "learning_rate": 3.671276781776346e-06,
      "loss": 0.88928282,
      "num_input_tokens_seen": 37015965,
      "step": 1743,
      "time_per_iteration": 2.6138579845428467
    },
    {
      "auxiliary_loss_clip": 0.01175491,
      "auxiliary_loss_mlp": 0.01039342,
      "balance_loss_clip": 1.05578148,
      "balance_loss_mlp": 1.03059816,
      "epoch": 0.20970360127457463,
      "flos": 25224768218880.0,
      "grad_norm": 2.347298010821068,
      "language_loss": 0.66878963,
      "learning_rate": 3.6708487815025128e-06,
      "loss": 0.69093794,
      "num_input_tokens_seen": 37036545,
      "step": 1744,
      "time_per_iteration": 2.6337168216705322
    },
    {
      "auxiliary_loss_clip": 0.01168642,
      "auxiliary_loss_mlp": 0.01036007,
      "balance_loss_clip": 1.05750024,
      "balance_loss_mlp": 1.02690554,
      "epoch": 0.20982384416521374,
      "flos": 18479164855680.0,
      "grad_norm": 3.6620819286201405,
      "language_loss": 0.74346453,
      "learning_rate": 3.6704205277622463e-06,
      "loss": 0.76551104,
      "num_input_tokens_seen": 37054985,
      "step": 1745,
      "time_per_iteration": 2.5538344383239746
    },
    {
      "auxiliary_loss_clip": 0.01186533,
      "auxiliary_loss_mlp": 0.01033548,
      "balance_loss_clip": 1.05764127,
      "balance_loss_mlp": 1.02473259,
      "epoch": 0.20994408705585282,
      "flos": 25373546352000.0,
      "grad_norm": 1.7742188297390749,
      "language_loss": 0.80347633,
      "learning_rate": 3.6699920206205146e-06,
      "loss": 0.8256771,
      "num_input_tokens_seen": 37075725,
      "step": 1746,
      "time_per_iteration": 2.601717948913574
    },
    {
      "auxiliary_loss_clip": 0.01200015,
      "auxiliary_loss_mlp": 0.01035305,
      "balance_loss_clip": 1.06049395,
      "balance_loss_mlp": 1.02693677,
      "epoch": 0.2100643299464919,
      "flos": 21320955313920.0,
      "grad_norm": 1.669481246917338,
      "language_loss": 0.82206702,
      "learning_rate": 3.669563260142321e-06,
      "loss": 0.84442025,
      "num_input_tokens_seen": 37094615,
      "step": 1747,
      "time_per_iteration": 2.5198159217834473
    },
    {
      "auxiliary_loss_clip": 0.01181103,
      "auxiliary_loss_mlp": 0.01040101,
      "balance_loss_clip": 1.0599823,
      "balance_loss_mlp": 1.03022408,
      "epoch": 0.21018457283713102,
      "flos": 19354379644800.0,
      "grad_norm": 2.0743231561586066,
      "language_loss": 0.84197032,
      "learning_rate": 3.6691342463927083e-06,
      "loss": 0.86418235,
      "num_input_tokens_seen": 37113610,
      "step": 1748,
      "time_per_iteration": 2.5584867000579834
    },
    {
      "auxiliary_loss_clip": 0.01170852,
      "auxiliary_loss_mlp": 0.01039105,
      "balance_loss_clip": 1.05620074,
      "balance_loss_mlp": 1.03003311,
      "epoch": 0.2103048157277701,
      "flos": 28330035914880.0,
      "grad_norm": 1.768047482736279,
      "language_loss": 0.81989002,
      "learning_rate": 3.668704979436758e-06,
      "loss": 0.84198958,
      "num_input_tokens_seen": 37133705,
      "step": 1749,
      "time_per_iteration": 2.6231143474578857
    },
    {
      "auxiliary_loss_clip": 0.01174486,
      "auxiliary_loss_mlp": 0.01033843,
      "balance_loss_clip": 1.05513716,
      "balance_loss_mlp": 1.02428806,
      "epoch": 0.21042505861840918,
      "flos": 17457290835840.0,
      "grad_norm": 2.117796100332401,
      "language_loss": 0.79207546,
      "learning_rate": 3.668275459339588e-06,
      "loss": 0.81415874,
      "num_input_tokens_seen": 37152185,
      "step": 1750,
      "time_per_iteration": 2.53269624710083
    },
    {
      "auxiliary_loss_clip": 0.01211514,
      "auxiliary_loss_mlp": 0.01036415,
      "balance_loss_clip": 1.06376147,
      "balance_loss_mlp": 1.02796936,
      "epoch": 0.21054530150904827,
      "flos": 14209817195520.0,
      "grad_norm": 1.9025955401998134,
      "language_loss": 0.8034057,
      "learning_rate": 3.667845686166358e-06,
      "loss": 0.82588506,
      "num_input_tokens_seen": 37169110,
      "step": 1751,
      "time_per_iteration": 2.4512040615081787
    },
    {
      "auxiliary_loss_clip": 0.01150591,
      "auxiliary_loss_mlp": 0.01036908,
      "balance_loss_clip": 1.05257213,
      "balance_loss_mlp": 1.02765119,
      "epoch": 0.21066554439968738,
      "flos": 18618210403200.0,
      "grad_norm": 1.6970086128558828,
      "language_loss": 0.86149675,
      "learning_rate": 3.6674156599822634e-06,
      "loss": 0.88337171,
      "num_input_tokens_seen": 37184905,
      "step": 1752,
      "time_per_iteration": 2.5544517040252686
    },
    {
      "auxiliary_loss_clip": 0.01156551,
      "auxiliary_loss_mlp": 0.01040472,
      "balance_loss_clip": 1.05152369,
      "balance_loss_mlp": 1.03047061,
      "epoch": 0.21078578729032646,
      "flos": 23658883741440.0,
      "grad_norm": 1.9932189998018164,
      "language_loss": 0.81789386,
      "learning_rate": 3.666985380852539e-06,
      "loss": 0.83986413,
      "num_input_tokens_seen": 37203910,
      "step": 1753,
      "time_per_iteration": 2.6283626556396484
    },
    {
      "auxiliary_loss_clip": 0.01180695,
      "auxiliary_loss_mlp": 0.01034626,
      "balance_loss_clip": 1.0575819,
      "balance_loss_mlp": 1.02572107,
      "epoch": 0.21090603018096554,
      "flos": 29346379240320.0,
      "grad_norm": 2.2774132257437825,
      "language_loss": 0.73731703,
      "learning_rate": 3.6665548488424576e-06,
      "loss": 0.75947022,
      "num_input_tokens_seen": 37222670,
      "step": 1754,
      "time_per_iteration": 2.609576463699341
    },
    {
      "auxiliary_loss_clip": 0.01211866,
      "auxiliary_loss_mlp": 0.01036099,
      "balance_loss_clip": 1.0618242,
      "balance_loss_mlp": 1.0264554,
      "epoch": 0.21102627307160465,
      "flos": 23261245205760.0,
      "grad_norm": 1.6502455120307287,
      "language_loss": 0.87988019,
      "learning_rate": 3.6661240640173307e-06,
      "loss": 0.90235984,
      "num_input_tokens_seen": 37244140,
      "step": 1755,
      "time_per_iteration": 2.5311052799224854
    },
    {
      "auxiliary_loss_clip": 0.0106792,
      "auxiliary_loss_mlp": 0.01005612,
      "balance_loss_clip": 1.01985919,
      "balance_loss_mlp": 1.00339496,
      "epoch": 0.21114651596224374,
      "flos": 54633454577280.0,
      "grad_norm": 0.8594720847701138,
      "language_loss": 0.57869363,
      "learning_rate": 3.6656930264425085e-06,
      "loss": 0.59942895,
      "num_input_tokens_seen": 37308185,
      "step": 1756,
      "time_per_iteration": 3.2043297290802
    },
    {
      "auxiliary_loss_clip": 0.01214455,
      "auxiliary_loss_mlp": 0.01035317,
      "balance_loss_clip": 1.06466246,
      "balance_loss_mlp": 1.02716327,
      "epoch": 0.21126675885288282,
      "flos": 21543314457600.0,
      "grad_norm": 1.826455745303862,
      "language_loss": 0.75574994,
      "learning_rate": 3.665261736183378e-06,
      "loss": 0.77824765,
      "num_input_tokens_seen": 37328220,
      "step": 1757,
      "time_per_iteration": 3.245326519012451
    },
    {
      "auxiliary_loss_clip": 0.01165153,
      "auxiliary_loss_mlp": 0.01039398,
      "balance_loss_clip": 1.0557847,
      "balance_loss_mlp": 1.03010523,
      "epoch": 0.2113870017435219,
      "flos": 10961876678400.0,
      "grad_norm": 3.5767141692528144,
      "language_loss": 0.88900101,
      "learning_rate": 3.664830193305366e-06,
      "loss": 0.91104645,
      "num_input_tokens_seen": 37345995,
      "step": 1758,
      "time_per_iteration": 2.5554144382476807
    },
    {
      "auxiliary_loss_clip": 0.01163062,
      "auxiliary_loss_mlp": 0.01039473,
      "balance_loss_clip": 1.05455494,
      "balance_loss_mlp": 1.03062749,
      "epoch": 0.211507244634161,
      "flos": 16653825463680.0,
      "grad_norm": 3.2110263341744423,
      "language_loss": 0.77210337,
      "learning_rate": 3.6643983978739373e-06,
      "loss": 0.79412872,
      "num_input_tokens_seen": 37362610,
      "step": 1759,
      "time_per_iteration": 2.576718330383301
    },
    {
      "auxiliary_loss_clip": 0.01177733,
      "auxiliary_loss_mlp": 0.01035275,
      "balance_loss_clip": 1.0604732,
      "balance_loss_mlp": 1.02594078,
      "epoch": 0.2116274875248001,
      "flos": 20954091755520.0,
      "grad_norm": 1.8000282599774204,
      "language_loss": 0.81955791,
      "learning_rate": 3.663966349954596e-06,
      "loss": 0.84168792,
      "num_input_tokens_seen": 37382790,
      "step": 1760,
      "time_per_iteration": 3.3607428073883057
    },
    {
      "auxiliary_loss_clip": 0.01093449,
      "auxiliary_loss_mlp": 0.01002939,
      "balance_loss_clip": 1.02167797,
      "balance_loss_mlp": 1.00067377,
      "epoch": 0.21174773041543918,
      "flos": 68196949424640.0,
      "grad_norm": 0.7918078789410999,
      "language_loss": 0.59729487,
      "learning_rate": 3.6635340496128816e-06,
      "loss": 0.61825877,
      "num_input_tokens_seen": 37439720,
      "step": 1761,
      "time_per_iteration": 3.0550057888031006
    },
    {
      "auxiliary_loss_clip": 0.01149398,
      "auxiliary_loss_mlp": 0.01037339,
      "balance_loss_clip": 1.05496418,
      "balance_loss_mlp": 1.02948928,
      "epoch": 0.2118679733060783,
      "flos": 20668315150080.0,
      "grad_norm": 2.015024738659513,
      "language_loss": 0.92733908,
      "learning_rate": 3.6631014969143747e-06,
      "loss": 0.94920641,
      "num_input_tokens_seen": 37459410,
      "step": 1762,
      "time_per_iteration": 2.6127772331237793
    },
    {
      "auxiliary_loss_clip": 0.01199947,
      "auxiliary_loss_mlp": 0.01039268,
      "balance_loss_clip": 1.06366396,
      "balance_loss_mlp": 1.03060174,
      "epoch": 0.21198821619671737,
      "flos": 23223431162880.0,
      "grad_norm": 2.0612919908097416,
      "language_loss": 0.88741601,
      "learning_rate": 3.662668691924693e-06,
      "loss": 0.90980816,
      "num_input_tokens_seen": 37480460,
      "step": 1763,
      "time_per_iteration": 3.279811143875122
    },
    {
      "auxiliary_loss_clip": 0.0116441,
      "auxiliary_loss_mlp": 0.01032685,
      "balance_loss_clip": 1.05342889,
      "balance_loss_mlp": 1.02340424,
      "epoch": 0.21210845908735645,
      "flos": 24498547044480.0,
      "grad_norm": 1.8824011975507167,
      "language_loss": 0.71050769,
      "learning_rate": 3.6622356347094927e-06,
      "loss": 0.73247862,
      "num_input_tokens_seen": 37502025,
      "step": 1764,
      "time_per_iteration": 3.430847406387329
    },
    {
      "auxiliary_loss_clip": 0.01168173,
      "auxiliary_loss_mlp": 0.01033434,
      "balance_loss_clip": 1.05583835,
      "balance_loss_mlp": 1.02383709,
      "epoch": 0.21222870197799554,
      "flos": 27089789160960.0,
      "grad_norm": 2.202802816630925,
      "language_loss": 0.78354013,
      "learning_rate": 3.6618023253344684e-06,
      "loss": 0.80555618,
      "num_input_tokens_seen": 37520885,
      "step": 1765,
      "time_per_iteration": 2.6374897956848145
    },
    {
      "auxiliary_loss_clip": 0.01194786,
      "auxiliary_loss_mlp": 0.01028763,
      "balance_loss_clip": 1.0573498,
      "balance_loss_mlp": 1.01867747,
      "epoch": 0.21234894486863465,
      "flos": 16873850223360.0,
      "grad_norm": 2.5211904019408253,
      "language_loss": 0.83481383,
      "learning_rate": 3.6613687638653527e-06,
      "loss": 0.85704935,
      "num_input_tokens_seen": 37539055,
      "step": 1766,
      "time_per_iteration": 2.4871089458465576
    },
    {
      "auxiliary_loss_clip": 0.01180806,
      "auxiliary_loss_mlp": 0.01036768,
      "balance_loss_clip": 1.06087995,
      "balance_loss_mlp": 1.02729046,
      "epoch": 0.21246918775927373,
      "flos": 23474949171840.0,
      "grad_norm": 1.9057748826647214,
      "language_loss": 0.77999997,
      "learning_rate": 3.660934950367916e-06,
      "loss": 0.8021757,
      "num_input_tokens_seen": 37558300,
      "step": 1767,
      "time_per_iteration": 2.591860055923462
    },
    {
      "auxiliary_loss_clip": 0.01198786,
      "auxiliary_loss_mlp": 0.0103524,
      "balance_loss_clip": 1.06095779,
      "balance_loss_mlp": 1.02675867,
      "epoch": 0.21258943064991281,
      "flos": 22382295402240.0,
      "grad_norm": 1.7077428121956286,
      "language_loss": 0.83534491,
      "learning_rate": 3.660500884907968e-06,
      "loss": 0.85768521,
      "num_input_tokens_seen": 37579040,
      "step": 1768,
      "time_per_iteration": 2.5368566513061523
    },
    {
      "auxiliary_loss_clip": 0.01063548,
      "auxiliary_loss_mlp": 0.01003221,
      "balance_loss_clip": 1.02507973,
      "balance_loss_mlp": 1.00117064,
      "epoch": 0.21270967354055192,
      "flos": 59440168679040.0,
      "grad_norm": 0.8290222204009497,
      "language_loss": 0.60101128,
      "learning_rate": 3.660066567551356e-06,
      "loss": 0.62167895,
      "num_input_tokens_seen": 37639185,
      "step": 1769,
      "time_per_iteration": 3.111382484436035
    },
    {
      "auxiliary_loss_clip": 0.01199566,
      "auxiliary_loss_mlp": 0.00762044,
      "balance_loss_clip": 1.06232643,
      "balance_loss_mlp": 1.00023913,
      "epoch": 0.212829916431191,
      "flos": 21544032729600.0,
      "grad_norm": 4.05603683293989,
      "language_loss": 0.84390146,
      "learning_rate": 3.6596319983639657e-06,
      "loss": 0.86351752,
      "num_input_tokens_seen": 37657765,
      "step": 1770,
      "time_per_iteration": 2.5459229946136475
    },
    {
      "auxiliary_loss_clip": 0.01166532,
      "auxiliary_loss_mlp": 0.00762859,
      "balance_loss_clip": 1.05650389,
      "balance_loss_mlp": 1.00025606,
      "epoch": 0.2129501593218301,
      "flos": 28987739896320.0,
      "grad_norm": 1.5920855573543164,
      "language_loss": 0.86296868,
      "learning_rate": 3.6591971774117214e-06,
      "loss": 0.88226265,
      "num_input_tokens_seen": 37680740,
      "step": 1771,
      "time_per_iteration": 2.653820753097534
    },
    {
      "auxiliary_loss_clip": 0.01204382,
      "auxiliary_loss_mlp": 0.01041102,
      "balance_loss_clip": 1.06254399,
      "balance_loss_mlp": 1.03341889,
      "epoch": 0.2130704022124692,
      "flos": 18806993308800.0,
      "grad_norm": 1.991834400596459,
      "language_loss": 0.80474454,
      "learning_rate": 3.6587621047605833e-06,
      "loss": 0.82719934,
      "num_input_tokens_seen": 37697910,
      "step": 1772,
      "time_per_iteration": 2.5018656253814697
    },
    {
      "auxiliary_loss_clip": 0.01196055,
      "auxiliary_loss_mlp": 0.01033415,
      "balance_loss_clip": 1.06027615,
      "balance_loss_mlp": 1.02423608,
      "epoch": 0.21319064510310828,
      "flos": 13918150759680.0,
      "grad_norm": 2.01765581731561,
      "language_loss": 0.86696136,
      "learning_rate": 3.6583267804765542e-06,
      "loss": 0.88925606,
      "num_input_tokens_seen": 37712245,
      "step": 1773,
      "time_per_iteration": 2.4914960861206055
    },
    {
      "auxiliary_loss_clip": 0.0119632,
      "auxiliary_loss_mlp": 0.01030647,
      "balance_loss_clip": 1.06096101,
      "balance_loss_mlp": 1.02095556,
      "epoch": 0.21331088799374737,
      "flos": 20959694277120.0,
      "grad_norm": 2.3437876121451975,
      "language_loss": 0.85573518,
      "learning_rate": 3.6578912046256702e-06,
      "loss": 0.87800491,
      "num_input_tokens_seen": 37730765,
      "step": 1774,
      "time_per_iteration": 2.517674207687378
    },
    {
      "auxiliary_loss_clip": 0.01165609,
      "auxiliary_loss_mlp": 0.01040269,
      "balance_loss_clip": 1.05679083,
      "balance_loss_mlp": 1.03125644,
      "epoch": 0.21343113088438645,
      "flos": 18624638937600.0,
      "grad_norm": 2.221147308565393,
      "language_loss": 0.7601577,
      "learning_rate": 3.6574553772740083e-06,
      "loss": 0.78221655,
      "num_input_tokens_seen": 37748695,
      "step": 1775,
      "time_per_iteration": 2.5666632652282715
    },
    {
      "auxiliary_loss_clip": 0.01088109,
      "auxiliary_loss_mlp": 0.01005706,
      "balance_loss_clip": 1.02077961,
      "balance_loss_mlp": 1.00334537,
      "epoch": 0.21355137377502556,
      "flos": 67413128791680.0,
      "grad_norm": 1.1157005736514312,
      "language_loss": 0.61874032,
      "learning_rate": 3.657019298487684e-06,
      "loss": 0.63967842,
      "num_input_tokens_seen": 37813705,
      "step": 1776,
      "time_per_iteration": 3.1777141094207764
    },
    {
      "auxiliary_loss_clip": 0.01201158,
      "auxiliary_loss_mlp": 0.00762977,
      "balance_loss_clip": 1.05973494,
      "balance_loss_mlp": 1.00027299,
      "epoch": 0.21367161666566464,
      "flos": 34532095697280.0,
      "grad_norm": 2.060449031070092,
      "language_loss": 0.83694029,
      "learning_rate": 3.6565829683328495e-06,
      "loss": 0.85658169,
      "num_input_tokens_seen": 37836330,
      "step": 1777,
      "time_per_iteration": 2.6400482654571533
    },
    {
      "auxiliary_loss_clip": 0.01193776,
      "auxiliary_loss_mlp": 0.01032412,
      "balance_loss_clip": 1.0615021,
      "balance_loss_mlp": 1.02396619,
      "epoch": 0.21379185955630373,
      "flos": 18989347680000.0,
      "grad_norm": 1.8494887624061311,
      "language_loss": 0.86353528,
      "learning_rate": 3.6561463868756965e-06,
      "loss": 0.88579714,
      "num_input_tokens_seen": 37855030,
      "step": 1778,
      "time_per_iteration": 2.5014660358428955
    },
    {
      "auxiliary_loss_clip": 0.0119847,
      "auxiliary_loss_mlp": 0.01034901,
      "balance_loss_clip": 1.06270909,
      "balance_loss_mlp": 1.02605534,
      "epoch": 0.21391210244694284,
      "flos": 28218497207040.0,
      "grad_norm": 2.3285017347123387,
      "language_loss": 0.78146219,
      "learning_rate": 3.655709554182452e-06,
      "loss": 0.80379593,
      "num_input_tokens_seen": 37875370,
      "step": 1779,
      "time_per_iteration": 2.573662519454956
    },
    {
      "auxiliary_loss_clip": 0.01196722,
      "auxiliary_loss_mlp": 0.01029774,
      "balance_loss_clip": 1.0575552,
      "balance_loss_mlp": 1.02014208,
      "epoch": 0.21403234533758192,
      "flos": 17455064192640.0,
      "grad_norm": 2.16832471942644,
      "language_loss": 0.84347314,
      "learning_rate": 3.6552724703193855e-06,
      "loss": 0.86573809,
      "num_input_tokens_seen": 37892560,
      "step": 1780,
      "time_per_iteration": 2.495096445083618
    },
    {
      "auxiliary_loss_clip": 0.01054051,
      "auxiliary_loss_mlp": 0.01008153,
      "balance_loss_clip": 1.01822448,
      "balance_loss_mlp": 1.00574481,
      "epoch": 0.214152588228221,
      "flos": 51637606686720.0,
      "grad_norm": 0.7876101690633913,
      "language_loss": 0.55965704,
      "learning_rate": 3.654835135352801e-06,
      "loss": 0.58027905,
      "num_input_tokens_seen": 37947370,
      "step": 1781,
      "time_per_iteration": 3.070044994354248
    },
    {
      "auxiliary_loss_clip": 0.01151819,
      "auxiliary_loss_mlp": 0.01034252,
      "balance_loss_clip": 1.05004549,
      "balance_loss_mlp": 1.02528167,
      "epoch": 0.21427283111886009,
      "flos": 19496154625920.0,
      "grad_norm": 1.8644097719002517,
      "language_loss": 0.87401688,
      "learning_rate": 3.654397549349043e-06,
      "loss": 0.8958776,
      "num_input_tokens_seen": 37964745,
      "step": 1782,
      "time_per_iteration": 3.442904233932495
    },
    {
      "auxiliary_loss_clip": 0.0118353,
      "auxiliary_loss_mlp": 0.0104014,
      "balance_loss_clip": 1.06292689,
      "balance_loss_mlp": 1.03037643,
      "epoch": 0.2143930740094992,
      "flos": 20084802710400.0,
      "grad_norm": 2.1340067015272655,
      "language_loss": 0.75092316,
      "learning_rate": 3.653959712374491e-06,
      "loss": 0.77315986,
      "num_input_tokens_seen": 37982850,
      "step": 1783,
      "time_per_iteration": 2.5465593338012695
    },
    {
      "auxiliary_loss_clip": 0.01162182,
      "auxiliary_loss_mlp": 0.010364,
      "balance_loss_clip": 1.05793619,
      "balance_loss_mlp": 1.02683949,
      "epoch": 0.21451331690013828,
      "flos": 21798603394560.0,
      "grad_norm": 1.9170927765204293,
      "language_loss": 0.82445294,
      "learning_rate": 3.6535216244955663e-06,
      "loss": 0.84643877,
      "num_input_tokens_seen": 38002745,
      "step": 1784,
      "time_per_iteration": 2.5742788314819336
    },
    {
      "auxiliary_loss_clip": 0.0118132,
      "auxiliary_loss_mlp": 0.01037,
      "balance_loss_clip": 1.05824876,
      "balance_loss_mlp": 1.02783263,
      "epoch": 0.21463355979077736,
      "flos": 32853882412800.0,
      "grad_norm": 2.0606889487714652,
      "language_loss": 0.71236056,
      "learning_rate": 3.653083285778726e-06,
      "loss": 0.7345438,
      "num_input_tokens_seen": 38024115,
      "step": 1785,
      "time_per_iteration": 2.6426258087158203
    },
    {
      "auxiliary_loss_clip": 0.01204719,
      "auxiliary_loss_mlp": 0.01037674,
      "balance_loss_clip": 1.06204844,
      "balance_loss_mlp": 1.02829874,
      "epoch": 0.21475380268141647,
      "flos": 21543817248000.0,
      "grad_norm": 2.660373378007314,
      "language_loss": 0.81390363,
      "learning_rate": 3.6526446962904653e-06,
      "loss": 0.83632755,
      "num_input_tokens_seen": 38042830,
      "step": 1786,
      "time_per_iteration": 3.309589385986328
    },
    {
      "auxiliary_loss_clip": 0.01195731,
      "auxiliary_loss_mlp": 0.01038682,
      "balance_loss_clip": 1.06228411,
      "balance_loss_mlp": 1.02953815,
      "epoch": 0.21487404557205556,
      "flos": 32159082660480.0,
      "grad_norm": 1.6933737999206764,
      "language_loss": 0.74126923,
      "learning_rate": 3.652205856097318e-06,
      "loss": 0.7636134,
      "num_input_tokens_seen": 38066015,
      "step": 1787,
      "time_per_iteration": 2.6018002033233643
    },
    {
      "auxiliary_loss_clip": 0.01177919,
      "auxiliary_loss_mlp": 0.00763207,
      "balance_loss_clip": 1.05718303,
      "balance_loss_mlp": 1.00020647,
      "epoch": 0.21499428846269464,
      "flos": 12673091583360.0,
      "grad_norm": 2.007366788913792,
      "language_loss": 0.78839707,
      "learning_rate": 3.651766765265856e-06,
      "loss": 0.80780828,
      "num_input_tokens_seen": 38083025,
      "step": 1788,
      "time_per_iteration": 3.3061418533325195
    },
    {
      "auxiliary_loss_clip": 0.01178815,
      "auxiliary_loss_mlp": 0.0103754,
      "balance_loss_clip": 1.05647111,
      "balance_loss_mlp": 1.02748501,
      "epoch": 0.21511453135333372,
      "flos": 23471573293440.0,
      "grad_norm": 3.1315990973485044,
      "language_loss": 0.8108604,
      "learning_rate": 3.65132742386269e-06,
      "loss": 0.83302391,
      "num_input_tokens_seen": 38098245,
      "step": 1789,
      "time_per_iteration": 2.5554301738739014
    },
    {
      "auxiliary_loss_clip": 0.01213777,
      "auxiliary_loss_mlp": 0.01037394,
      "balance_loss_clip": 1.06201291,
      "balance_loss_mlp": 1.02859628,
      "epoch": 0.21523477424397283,
      "flos": 26943560893440.0,
      "grad_norm": 1.9836481635723249,
      "language_loss": 0.84704173,
      "learning_rate": 3.6508878319544656e-06,
      "loss": 0.86955345,
      "num_input_tokens_seen": 38118460,
      "step": 1790,
      "time_per_iteration": 3.2349603176116943
    },
    {
      "auxiliary_loss_clip": 0.01173601,
      "auxiliary_loss_mlp": 0.01040031,
      "balance_loss_clip": 1.05877256,
      "balance_loss_mlp": 1.03002381,
      "epoch": 0.21535501713461191,
      "flos": 18916161719040.0,
      "grad_norm": 3.398103684844648,
      "language_loss": 0.8169058,
      "learning_rate": 3.65044798960787e-06,
      "loss": 0.83904219,
      "num_input_tokens_seen": 38136800,
      "step": 1791,
      "time_per_iteration": 2.5242550373077393
    },
    {
      "auxiliary_loss_clip": 0.01159907,
      "auxiliary_loss_mlp": 0.01027143,
      "balance_loss_clip": 1.05470443,
      "balance_loss_mlp": 1.01847684,
      "epoch": 0.215475260025251,
      "flos": 17895113712000.0,
      "grad_norm": 1.930511700220896,
      "language_loss": 0.77896643,
      "learning_rate": 3.650007896889627e-06,
      "loss": 0.80083686,
      "num_input_tokens_seen": 38155380,
      "step": 1792,
      "time_per_iteration": 2.597355604171753
    },
    {
      "auxiliary_loss_clip": 0.01214436,
      "auxiliary_loss_mlp": 0.01040378,
      "balance_loss_clip": 1.06569278,
      "balance_loss_mlp": 1.03080535,
      "epoch": 0.2155955029158901,
      "flos": 16654292340480.0,
      "grad_norm": 1.7411558672954228,
      "language_loss": 0.80518937,
      "learning_rate": 3.6495675538664974e-06,
      "loss": 0.82773757,
      "num_input_tokens_seen": 38174395,
      "step": 1793,
      "time_per_iteration": 2.471722364425659
    },
    {
      "auxiliary_loss_clip": 0.01182711,
      "auxiliary_loss_mlp": 0.0103646,
      "balance_loss_clip": 1.05513537,
      "balance_loss_mlp": 1.02692318,
      "epoch": 0.2157157458065292,
      "flos": 23621213352960.0,
      "grad_norm": 1.7102353773213608,
      "language_loss": 0.82311577,
      "learning_rate": 3.649126960605282e-06,
      "loss": 0.84530753,
      "num_input_tokens_seen": 38195380,
      "step": 1794,
      "time_per_iteration": 2.583911657333374
    },
    {
      "auxiliary_loss_clip": 0.01179263,
      "auxiliary_loss_mlp": 0.01032466,
      "balance_loss_clip": 1.05741215,
      "balance_loss_mlp": 1.02339959,
      "epoch": 0.21583598869716827,
      "flos": 22127078292480.0,
      "grad_norm": 2.380151079925419,
      "language_loss": 0.83437598,
      "learning_rate": 3.6486861171728174e-06,
      "loss": 0.85649329,
      "num_input_tokens_seen": 38213775,
      "step": 1795,
      "time_per_iteration": 2.533357620239258
    },
    {
      "auxiliary_loss_clip": 0.01171051,
      "auxiliary_loss_mlp": 0.01033395,
      "balance_loss_clip": 1.0550909,
      "balance_loss_mlp": 1.02423429,
      "epoch": 0.21595623158780738,
      "flos": 23441229279360.0,
      "grad_norm": 1.6829178474386945,
      "language_loss": 0.78576922,
      "learning_rate": 3.6482450236359803e-06,
      "loss": 0.80781364,
      "num_input_tokens_seen": 38235630,
      "step": 1796,
      "time_per_iteration": 2.637712001800537
    },
    {
      "auxiliary_loss_clip": 0.0119605,
      "auxiliary_loss_mlp": 0.01035885,
      "balance_loss_clip": 1.06105447,
      "balance_loss_mlp": 1.02723026,
      "epoch": 0.21607647447844647,
      "flos": 26906501036160.0,
      "grad_norm": 2.287342732119108,
      "language_loss": 0.77358007,
      "learning_rate": 3.647803680061683e-06,
      "loss": 0.79589939,
      "num_input_tokens_seen": 38256045,
      "step": 1797,
      "time_per_iteration": 2.5509908199310303
    },
    {
      "auxiliary_loss_clip": 0.01184254,
      "auxiliary_loss_mlp": 0.01040004,
      "balance_loss_clip": 1.05920768,
      "balance_loss_mlp": 1.02983499,
      "epoch": 0.21619671736908555,
      "flos": 14495378319360.0,
      "grad_norm": 3.5241454397463445,
      "language_loss": 0.74816871,
      "learning_rate": 3.6473620865168776e-06,
      "loss": 0.77041125,
      "num_input_tokens_seen": 38272915,
      "step": 1798,
      "time_per_iteration": 2.5207464694976807
    },
    {
      "auxiliary_loss_clip": 0.01184296,
      "auxiliary_loss_mlp": 0.01043762,
      "balance_loss_clip": 1.06224215,
      "balance_loss_mlp": 1.03470802,
      "epoch": 0.21631696025972463,
      "flos": 17931096161280.0,
      "grad_norm": 1.8597321068265413,
      "language_loss": 0.81620193,
      "learning_rate": 3.646920243068554e-06,
      "loss": 0.8384825,
      "num_input_tokens_seen": 38290810,
      "step": 1799,
      "time_per_iteration": 2.5166287422180176
    },
    {
      "auxiliary_loss_clip": 0.01169509,
      "auxiliary_loss_mlp": 0.01040001,
      "balance_loss_clip": 1.05715346,
      "balance_loss_mlp": 1.02980852,
      "epoch": 0.21643720315036374,
      "flos": 24462385027200.0,
      "grad_norm": 1.6074014002718982,
      "language_loss": 0.74505556,
      "learning_rate": 3.6464781497837384e-06,
      "loss": 0.76715064,
      "num_input_tokens_seen": 38312785,
      "step": 1800,
      "time_per_iteration": 2.584984064102173
    },
    {
      "auxiliary_loss_clip": 0.0118571,
      "auxiliary_loss_mlp": 0.01039387,
      "balance_loss_clip": 1.05641818,
      "balance_loss_mlp": 1.0295341,
      "epoch": 0.21655744604100283,
      "flos": 28474432588800.0,
      "grad_norm": 2.0036948896099993,
      "language_loss": 0.72636598,
      "learning_rate": 3.6460358067294965e-06,
      "loss": 0.74861693,
      "num_input_tokens_seen": 38334015,
      "step": 1801,
      "time_per_iteration": 2.610901117324829
    },
    {
      "auxiliary_loss_clip": 0.01216437,
      "auxiliary_loss_mlp": 0.01034996,
      "balance_loss_clip": 1.06274533,
      "balance_loss_mlp": 1.02528644,
      "epoch": 0.2166776889316419,
      "flos": 20152960767360.0,
      "grad_norm": 2.152115693847454,
      "language_loss": 0.77449989,
      "learning_rate": 3.645593213972932e-06,
      "loss": 0.7970143,
      "num_input_tokens_seen": 38352920,
      "step": 1802,
      "time_per_iteration": 2.4989206790924072
    },
    {
      "auxiliary_loss_clip": 0.01192817,
      "auxiliary_loss_mlp": 0.01040117,
      "balance_loss_clip": 1.05891967,
      "balance_loss_mlp": 1.03019881,
      "epoch": 0.21679793182228102,
      "flos": 15193482122880.0,
      "grad_norm": 2.1722474029640666,
      "language_loss": 0.79912198,
      "learning_rate": 3.6451503715811852e-06,
      "loss": 0.82145131,
      "num_input_tokens_seen": 38371230,
      "step": 1803,
      "time_per_iteration": 2.4822099208831787
    },
    {
      "auxiliary_loss_clip": 0.01178706,
      "auxiliary_loss_mlp": 0.01029,
      "balance_loss_clip": 1.05854595,
      "balance_loss_mlp": 1.0208106,
      "epoch": 0.2169181747129201,
      "flos": 17384464010880.0,
      "grad_norm": 1.9734963162154404,
      "language_loss": 0.80126309,
      "learning_rate": 3.6447072796214345e-06,
      "loss": 0.82334012,
      "num_input_tokens_seen": 38389795,
      "step": 1804,
      "time_per_iteration": 2.5004353523254395
    },
    {
      "auxiliary_loss_clip": 0.01059468,
      "auxiliary_loss_mlp": 0.01002194,
      "balance_loss_clip": 1.02290678,
      "balance_loss_mlp": 0.99985713,
      "epoch": 0.21703841760355919,
      "flos": 58760955429120.0,
      "grad_norm": 0.925219396881176,
      "language_loss": 0.63134742,
      "learning_rate": 3.644263938160898e-06,
      "loss": 0.65196407,
      "num_input_tokens_seen": 38445760,
      "step": 1805,
      "time_per_iteration": 3.0824172496795654
    },
    {
      "auxiliary_loss_clip": 0.01166429,
      "auxiliary_loss_mlp": 0.01044619,
      "balance_loss_clip": 1.05720377,
      "balance_loss_mlp": 1.0348022,
      "epoch": 0.21715866049419827,
      "flos": 22418457419520.0,
      "grad_norm": 1.87844486554684,
      "language_loss": 0.71842462,
      "learning_rate": 3.6438203472668293e-06,
      "loss": 0.74053514,
      "num_input_tokens_seen": 38465405,
      "step": 1806,
      "time_per_iteration": 2.5857090950012207
    },
    {
      "auxiliary_loss_clip": 0.01187638,
      "auxiliary_loss_mlp": 0.0103725,
      "balance_loss_clip": 1.06016231,
      "balance_loss_mlp": 1.02861917,
      "epoch": 0.21727890338483738,
      "flos": 17237732952960.0,
      "grad_norm": 1.8965991438675915,
      "language_loss": 0.81905431,
      "learning_rate": 3.6433765070065206e-06,
      "loss": 0.84130323,
      "num_input_tokens_seen": 38483195,
      "step": 1807,
      "time_per_iteration": 2.53423810005188
    },
    {
      "auxiliary_loss_clip": 0.0121543,
      "auxiliary_loss_mlp": 0.01037318,
      "balance_loss_clip": 1.06327033,
      "balance_loss_mlp": 1.02750111,
      "epoch": 0.21739914627547646,
      "flos": 13434792416640.0,
      "grad_norm": 2.88233654437488,
      "language_loss": 0.87944436,
      "learning_rate": 3.6429324174473025e-06,
      "loss": 0.90197182,
      "num_input_tokens_seen": 38496735,
      "step": 1808,
      "time_per_iteration": 2.4329371452331543
    },
    {
      "auxiliary_loss_clip": 0.01202013,
      "auxiliary_loss_mlp": 0.01038364,
      "balance_loss_clip": 1.06155908,
      "balance_loss_mlp": 1.02889323,
      "epoch": 0.21751938916611555,
      "flos": 20959514709120.0,
      "grad_norm": 2.0222607603278147,
      "language_loss": 0.84500253,
      "learning_rate": 3.6424880786565425e-06,
      "loss": 0.86740625,
      "num_input_tokens_seen": 38512880,
      "step": 1809,
      "time_per_iteration": 3.270334243774414
    },
    {
      "auxiliary_loss_clip": 0.01149458,
      "auxiliary_loss_mlp": 0.01036837,
      "balance_loss_clip": 1.05595958,
      "balance_loss_mlp": 1.02665663,
      "epoch": 0.21763963205675466,
      "flos": 27599936071680.0,
      "grad_norm": 2.7817188172909275,
      "language_loss": 0.80076683,
      "learning_rate": 3.6420434907016482e-06,
      "loss": 0.82262975,
      "num_input_tokens_seen": 38532570,
      "step": 1810,
      "time_per_iteration": 2.6436312198638916
    },
    {
      "auxiliary_loss_clip": 0.01204455,
      "auxiliary_loss_mlp": 0.01040593,
      "balance_loss_clip": 1.06589079,
      "balance_loss_mlp": 1.0316937,
      "epoch": 0.21775987494739374,
      "flos": 21430411032960.0,
      "grad_norm": 1.8257748931333257,
      "language_loss": 0.81205678,
      "learning_rate": 3.6415986536500606e-06,
      "loss": 0.83450735,
      "num_input_tokens_seen": 38550900,
      "step": 1811,
      "time_per_iteration": 2.5162250995635986
    },
    {
      "auxiliary_loss_clip": 0.01149762,
      "auxiliary_loss_mlp": 0.01038284,
      "balance_loss_clip": 1.06046343,
      "balance_loss_mlp": 1.02965939,
      "epoch": 0.21788011783803282,
      "flos": 18332972501760.0,
      "grad_norm": 1.9240441529915235,
      "language_loss": 0.80348545,
      "learning_rate": 3.641153567569263e-06,
      "loss": 0.8253659,
      "num_input_tokens_seen": 38569215,
      "step": 1812,
      "time_per_iteration": 3.428393602371216
    },
    {
      "auxiliary_loss_clip": 0.01194554,
      "auxiliary_loss_mlp": 0.01034919,
      "balance_loss_clip": 1.06010473,
      "balance_loss_mlp": 1.02603793,
      "epoch": 0.2180003607286719,
      "flos": 30262748037120.0,
      "grad_norm": 2.1642542344481077,
      "language_loss": 0.95230532,
      "learning_rate": 3.640708232526774e-06,
      "loss": 0.97460008,
      "num_input_tokens_seen": 38587870,
      "step": 1813,
      "time_per_iteration": 2.5741560459136963
    },
    {
      "auxiliary_loss_clip": 0.01136357,
      "auxiliary_loss_mlp": 0.01036601,
      "balance_loss_clip": 1.04870605,
      "balance_loss_mlp": 1.02670097,
      "epoch": 0.21812060361931102,
      "flos": 25480272637440.0,
      "grad_norm": 1.6681753061438978,
      "language_loss": 0.78634769,
      "learning_rate": 3.6402626485901504e-06,
      "loss": 0.80807734,
      "num_input_tokens_seen": 38606965,
      "step": 1814,
      "time_per_iteration": 2.6998040676116943
    },
    {
      "auxiliary_loss_clip": 0.01198494,
      "auxiliary_loss_mlp": 0.01034868,
      "balance_loss_clip": 1.06461477,
      "balance_loss_mlp": 1.0256058,
      "epoch": 0.2182408465099501,
      "flos": 21908166854400.0,
      "grad_norm": 2.3349870223848224,
      "language_loss": 0.77970892,
      "learning_rate": 3.639816815826988e-06,
      "loss": 0.8020426,
      "num_input_tokens_seen": 38626290,
      "step": 1815,
      "time_per_iteration": 3.997248888015747
    },
    {
      "auxiliary_loss_clip": 0.01183767,
      "auxiliary_loss_mlp": 0.01029259,
      "balance_loss_clip": 1.05962157,
      "balance_loss_mlp": 1.02018118,
      "epoch": 0.21836108940058918,
      "flos": 23657339456640.0,
      "grad_norm": 1.9326084855370262,
      "language_loss": 0.77788258,
      "learning_rate": 3.6393707343049176e-06,
      "loss": 0.80001283,
      "num_input_tokens_seen": 38646620,
      "step": 1816,
      "time_per_iteration": 2.5739402770996094
    },
    {
      "auxiliary_loss_clip": 0.01201439,
      "auxiliary_loss_mlp": 0.01035053,
      "balance_loss_clip": 1.06011176,
      "balance_loss_mlp": 1.02534318,
      "epoch": 0.2184813322912283,
      "flos": 24681009156480.0,
      "grad_norm": 2.3638843607257494,
      "language_loss": 0.7390669,
      "learning_rate": 3.6389244040916104e-06,
      "loss": 0.76143181,
      "num_input_tokens_seen": 38665695,
      "step": 1817,
      "time_per_iteration": 2.5493388175964355
    },
    {
      "auxiliary_loss_clip": 0.01175062,
      "auxiliary_loss_mlp": 0.00764246,
      "balance_loss_clip": 1.05801857,
      "balance_loss_mlp": 1.00029755,
      "epoch": 0.21860157518186737,
      "flos": 26574650259840.0,
      "grad_norm": 2.071555003346373,
      "language_loss": 0.79437768,
      "learning_rate": 3.6384778252547747e-06,
      "loss": 0.81377077,
      "num_input_tokens_seen": 38681575,
      "step": 1818,
      "time_per_iteration": 2.569887161254883
    },
    {
      "auxiliary_loss_clip": 0.01183679,
      "auxiliary_loss_mlp": 0.0076314,
      "balance_loss_clip": 1.06510675,
      "balance_loss_mlp": 1.00028932,
      "epoch": 0.21872181807250646,
      "flos": 20886292834560.0,
      "grad_norm": 2.3638684582860456,
      "language_loss": 0.77852356,
      "learning_rate": 3.638030997862155e-06,
      "loss": 0.79799175,
      "num_input_tokens_seen": 38700510,
      "step": 1819,
      "time_per_iteration": 2.536895275115967
    },
    {
      "auxiliary_loss_clip": 0.01077158,
      "auxiliary_loss_mlp": 0.01006076,
      "balance_loss_clip": 1.02488899,
      "balance_loss_mlp": 1.00366771,
      "epoch": 0.21884206096314554,
      "flos": 61209452897280.0,
      "grad_norm": 0.7666662054381629,
      "language_loss": 0.59454888,
      "learning_rate": 3.6375839219815356e-06,
      "loss": 0.61538118,
      "num_input_tokens_seen": 38758310,
      "step": 1820,
      "time_per_iteration": 3.0439674854278564
    },
    {
      "auxiliary_loss_clip": 0.01213103,
      "auxiliary_loss_mlp": 0.01034858,
      "balance_loss_clip": 1.06146932,
      "balance_loss_mlp": 1.02485681,
      "epoch": 0.21896230385378465,
      "flos": 23473835850240.0,
      "grad_norm": 1.8835672874293465,
      "language_loss": 0.83094394,
      "learning_rate": 3.6371365976807375e-06,
      "loss": 0.85342354,
      "num_input_tokens_seen": 38778705,
      "step": 1821,
      "time_per_iteration": 2.5021250247955322
    },
    {
      "auxiliary_loss_clip": 0.01146234,
      "auxiliary_loss_mlp": 0.0104153,
      "balance_loss_clip": 1.05631769,
      "balance_loss_mlp": 1.0319519,
      "epoch": 0.21908254674442373,
      "flos": 25081915829760.0,
      "grad_norm": 1.6910193979020673,
      "language_loss": 0.83650446,
      "learning_rate": 3.6366890250276185e-06,
      "loss": 0.85838211,
      "num_input_tokens_seen": 38799660,
      "step": 1822,
      "time_per_iteration": 2.6365389823913574
    },
    {
      "auxiliary_loss_clip": 0.01212246,
      "auxiliary_loss_mlp": 0.01034846,
      "balance_loss_clip": 1.06205201,
      "balance_loss_mlp": 1.02567315,
      "epoch": 0.21920278963506282,
      "flos": 23513768795520.0,
      "grad_norm": 2.565131486087501,
      "language_loss": 0.89789653,
      "learning_rate": 3.6362412040900764e-06,
      "loss": 0.92036748,
      "num_input_tokens_seen": 38819450,
      "step": 1823,
      "time_per_iteration": 2.525851249694824
    },
    {
      "auxiliary_loss_clip": 0.01202778,
      "auxiliary_loss_mlp": 0.01034968,
      "balance_loss_clip": 1.0627079,
      "balance_loss_mlp": 1.02528775,
      "epoch": 0.21932303252570193,
      "flos": 29242238734080.0,
      "grad_norm": 1.8773757094131676,
      "language_loss": 0.80199826,
      "learning_rate": 3.635793134936044e-06,
      "loss": 0.82437575,
      "num_input_tokens_seen": 38840460,
      "step": 1824,
      "time_per_iteration": 2.6082136631011963
    },
    {
      "auxiliary_loss_clip": 0.0119634,
      "auxiliary_loss_mlp": 0.01038282,
      "balance_loss_clip": 1.06052113,
      "balance_loss_mlp": 1.02902579,
      "epoch": 0.219443275416341,
      "flos": 20806857907200.0,
      "grad_norm": 1.7518697399448588,
      "language_loss": 0.73258996,
      "learning_rate": 3.635344817633494e-06,
      "loss": 0.75493622,
      "num_input_tokens_seen": 38859775,
      "step": 1825,
      "time_per_iteration": 2.524298906326294
    },
    {
      "auxiliary_loss_clip": 0.01192391,
      "auxiliary_loss_mlp": 0.01036855,
      "balance_loss_clip": 1.05901122,
      "balance_loss_mlp": 1.02786064,
      "epoch": 0.2195635183069801,
      "flos": 14501555458560.0,
      "grad_norm": 4.1226091510224565,
      "language_loss": 0.75535458,
      "learning_rate": 3.634896252250436e-06,
      "loss": 0.77764696,
      "num_input_tokens_seen": 38876540,
      "step": 1826,
      "time_per_iteration": 2.483999490737915
    },
    {
      "auxiliary_loss_clip": 0.01219369,
      "auxiliary_loss_mlp": 0.01039355,
      "balance_loss_clip": 1.06670177,
      "balance_loss_mlp": 1.02895391,
      "epoch": 0.2196837611976192,
      "flos": 24243473589120.0,
      "grad_norm": 1.9221677693658599,
      "language_loss": 0.82259226,
      "learning_rate": 3.6344474388549157e-06,
      "loss": 0.84517944,
      "num_input_tokens_seen": 38896195,
      "step": 1827,
      "time_per_iteration": 2.502225399017334
    },
    {
      "auxiliary_loss_clip": 0.01199719,
      "auxiliary_loss_mlp": 0.01034937,
      "balance_loss_clip": 1.06288934,
      "balance_loss_mlp": 1.02457225,
      "epoch": 0.2198040040882583,
      "flos": 18074523168000.0,
      "grad_norm": 1.9736827601586437,
      "language_loss": 0.80324185,
      "learning_rate": 3.6339983775150183e-06,
      "loss": 0.82558846,
      "num_input_tokens_seen": 38912755,
      "step": 1828,
      "time_per_iteration": 2.4580917358398438
    },
    {
      "auxiliary_loss_clip": 0.01198992,
      "auxiliary_loss_mlp": 0.01040708,
      "balance_loss_clip": 1.06317544,
      "balance_loss_mlp": 1.0311234,
      "epoch": 0.21992424697889737,
      "flos": 17784185535360.0,
      "grad_norm": 3.2211218875311904,
      "language_loss": 0.84172028,
      "learning_rate": 3.6335490682988664e-06,
      "loss": 0.86411732,
      "num_input_tokens_seen": 38928365,
      "step": 1829,
      "time_per_iteration": 2.4820921421051025
    },
    {
      "auxiliary_loss_clip": 0.01131718,
      "auxiliary_loss_mlp": 0.01037307,
      "balance_loss_clip": 1.05155694,
      "balance_loss_mlp": 1.02793717,
      "epoch": 0.22004448986953645,
      "flos": 17638495971840.0,
      "grad_norm": 30.646845898126266,
      "language_loss": 0.82763278,
      "learning_rate": 3.63309951127462e-06,
      "loss": 0.84932303,
      "num_input_tokens_seen": 38945275,
      "step": 1830,
      "time_per_iteration": 2.6118664741516113
    },
    {
      "auxiliary_loss_clip": 0.01168673,
      "auxiliary_loss_mlp": 0.01042043,
      "balance_loss_clip": 1.05890727,
      "balance_loss_mlp": 1.03266096,
      "epoch": 0.22016473276017556,
      "flos": 22275533203200.0,
      "grad_norm": 2.6069377117147385,
      "language_loss": 0.7554816,
      "learning_rate": 3.6326497065104757e-06,
      "loss": 0.77758873,
      "num_input_tokens_seen": 38965740,
      "step": 1831,
      "time_per_iteration": 2.622213125228882
    },
    {
      "auxiliary_loss_clip": 0.01203808,
      "auxiliary_loss_mlp": 0.01042536,
      "balance_loss_clip": 1.06184101,
      "balance_loss_mlp": 1.0328567,
      "epoch": 0.22028497565081465,
      "flos": 25556259859200.0,
      "grad_norm": 2.0667166426303325,
      "language_loss": 0.78511298,
      "learning_rate": 3.6321996540746697e-06,
      "loss": 0.80757648,
      "num_input_tokens_seen": 38984815,
      "step": 1832,
      "time_per_iteration": 2.556617021560669
    },
    {
      "auxiliary_loss_clip": 0.01168508,
      "auxiliary_loss_mlp": 0.01040857,
      "balance_loss_clip": 1.05662847,
      "balance_loss_mlp": 1.03109348,
      "epoch": 0.22040521854145373,
      "flos": 36247332925440.0,
      "grad_norm": 1.7858488942817552,
      "language_loss": 0.80867457,
      "learning_rate": 3.6317493540354733e-06,
      "loss": 0.83076823,
      "num_input_tokens_seen": 39008230,
      "step": 1833,
      "time_per_iteration": 2.720679759979248
    },
    {
      "auxiliary_loss_clip": 0.01192827,
      "auxiliary_loss_mlp": 0.01033484,
      "balance_loss_clip": 1.05885386,
      "balance_loss_mlp": 1.02407241,
      "epoch": 0.22052546143209284,
      "flos": 11838420270720.0,
      "grad_norm": 2.010552468999786,
      "language_loss": 0.77242768,
      "learning_rate": 3.6312988064611976e-06,
      "loss": 0.79469073,
      "num_input_tokens_seen": 39026540,
      "step": 1834,
      "time_per_iteration": 3.3244943618774414
    },
    {
      "auxiliary_loss_clip": 0.01169492,
      "auxiliary_loss_mlp": 0.01036735,
      "balance_loss_clip": 1.05367279,
      "balance_loss_mlp": 1.02720439,
      "epoch": 0.22064570432273192,
      "flos": 24209250906240.0,
      "grad_norm": 1.8952734571863783,
      "language_loss": 0.81451273,
      "learning_rate": 3.6308480114201896e-06,
      "loss": 0.83657503,
      "num_input_tokens_seen": 39048460,
      "step": 1835,
      "time_per_iteration": 2.6163628101348877
    },
    {
      "auxiliary_loss_clip": 0.01216903,
      "auxiliary_loss_mlp": 0.01040161,
      "balance_loss_clip": 1.06598461,
      "balance_loss_mlp": 1.03067791,
      "epoch": 0.220765947213371,
      "flos": 17931347556480.0,
      "grad_norm": 1.849449296883682,
      "language_loss": 0.76427948,
      "learning_rate": 3.630396968980835e-06,
      "loss": 0.78685009,
      "num_input_tokens_seen": 39066335,
      "step": 1836,
      "time_per_iteration": 2.4813525676727295
    },
    {
      "auxiliary_loss_clip": 0.01183669,
      "auxiliary_loss_mlp": 0.01045498,
      "balance_loss_clip": 1.05789924,
      "balance_loss_mlp": 1.03519845,
      "epoch": 0.2208861901040101,
      "flos": 26757040544640.0,
      "grad_norm": 2.4955239342143027,
      "language_loss": 0.83745366,
      "learning_rate": 3.6299456792115575e-06,
      "loss": 0.85974532,
      "num_input_tokens_seen": 39087590,
      "step": 1837,
      "time_per_iteration": 3.374748706817627
    },
    {
      "auxiliary_loss_clip": 0.01105407,
      "auxiliary_loss_mlp": 0.01031632,
      "balance_loss_clip": 1.0460099,
      "balance_loss_mlp": 1.02253652,
      "epoch": 0.2210064329946492,
      "flos": 17817977255040.0,
      "grad_norm": 1.7741231884670803,
      "language_loss": 0.81190169,
      "learning_rate": 3.629494142180815e-06,
      "loss": 0.8332721,
      "num_input_tokens_seen": 39106335,
      "step": 1838,
      "time_per_iteration": 2.706540822982788
    },
    {
      "auxiliary_loss_clip": 0.01213527,
      "auxiliary_loss_mlp": 0.01034397,
      "balance_loss_clip": 1.0634129,
      "balance_loss_mlp": 1.02518821,
      "epoch": 0.22112667588528828,
      "flos": 17967401832960.0,
      "grad_norm": 2.0576280325168104,
      "language_loss": 0.85602421,
      "learning_rate": 3.6290423579571075e-06,
      "loss": 0.8785035,
      "num_input_tokens_seen": 39122875,
      "step": 1839,
      "time_per_iteration": 2.6072866916656494
    },
    {
      "auxiliary_loss_clip": 0.01196487,
      "auxiliary_loss_mlp": 0.01040481,
      "balance_loss_clip": 1.06289268,
      "balance_loss_mlp": 1.03109908,
      "epoch": 0.22124691877592736,
      "flos": 18369206346240.0,
      "grad_norm": 1.6387086328042546,
      "language_loss": 0.80326235,
      "learning_rate": 3.6285903266089694e-06,
      "loss": 0.82563198,
      "num_input_tokens_seen": 39142150,
      "step": 1840,
      "time_per_iteration": 2.4984676837921143
    },
    {
      "auxiliary_loss_clip": 0.01179879,
      "auxiliary_loss_mlp": 0.01028299,
      "balance_loss_clip": 1.05618691,
      "balance_loss_mlp": 1.01906586,
      "epoch": 0.22136716166656648,
      "flos": 20813286441600.0,
      "grad_norm": 1.694509582211568,
      "language_loss": 0.77017617,
      "learning_rate": 3.628138048204974e-06,
      "loss": 0.79225791,
      "num_input_tokens_seen": 39162835,
      "step": 1841,
      "time_per_iteration": 3.245981454849243
    },
    {
      "auxiliary_loss_clip": 0.01145619,
      "auxiliary_loss_mlp": 0.01040249,
      "balance_loss_clip": 1.05595756,
      "balance_loss_mlp": 1.02960408,
      "epoch": 0.22148740455720556,
      "flos": 17675699483520.0,
      "grad_norm": 2.093394560028967,
      "language_loss": 0.76006997,
      "learning_rate": 3.6276855228137304e-06,
      "loss": 0.7819286,
      "num_input_tokens_seen": 39181040,
      "step": 1842,
      "time_per_iteration": 3.3193488121032715
    },
    {
      "auxiliary_loss_clip": 0.01214137,
      "auxiliary_loss_mlp": 0.00763406,
      "balance_loss_clip": 1.06324983,
      "balance_loss_mlp": 1.00035822,
      "epoch": 0.22160764744784464,
      "flos": 21726710323200.0,
      "grad_norm": 2.0510147220362542,
      "language_loss": 0.81875038,
      "learning_rate": 3.6272327505038874e-06,
      "loss": 0.83852577,
      "num_input_tokens_seen": 39197505,
      "step": 1843,
      "time_per_iteration": 2.492896556854248
    },
    {
      "auxiliary_loss_clip": 0.01158678,
      "auxiliary_loss_mlp": 0.0103154,
      "balance_loss_clip": 1.05606616,
      "balance_loss_mlp": 1.02280235,
      "epoch": 0.22172789033848372,
      "flos": 23764712186880.0,
      "grad_norm": 2.7237161240798917,
      "language_loss": 0.78319204,
      "learning_rate": 3.626779731344131e-06,
      "loss": 0.80509424,
      "num_input_tokens_seen": 39217295,
      "step": 1844,
      "time_per_iteration": 2.615344762802124
    },
    {
      "auxiliary_loss_clip": 0.01209689,
      "auxiliary_loss_mlp": 0.01029382,
      "balance_loss_clip": 1.0602349,
      "balance_loss_mlp": 1.01994014,
      "epoch": 0.22184813322912283,
      "flos": 16982300361600.0,
      "grad_norm": 2.4405884354401666,
      "language_loss": 0.85374701,
      "learning_rate": 3.6263264654031814e-06,
      "loss": 0.87613773,
      "num_input_tokens_seen": 39234195,
      "step": 1845,
      "time_per_iteration": 2.4658052921295166
    },
    {
      "auxiliary_loss_clip": 0.01071997,
      "auxiliary_loss_mlp": 0.0100367,
      "balance_loss_clip": 1.02913725,
      "balance_loss_mlp": 1.00122583,
      "epoch": 0.22196837611976192,
      "flos": 61823740314240.0,
      "grad_norm": 0.6965947223285422,
      "language_loss": 0.59226727,
      "learning_rate": 3.6258729527498008e-06,
      "loss": 0.61302394,
      "num_input_tokens_seen": 39295040,
      "step": 1846,
      "time_per_iteration": 3.133504629135132
    },
    {
      "auxiliary_loss_clip": 0.01188678,
      "auxiliary_loss_mlp": 0.01035496,
      "balance_loss_clip": 1.06099463,
      "balance_loss_mlp": 1.02583385,
      "epoch": 0.222088619010401,
      "flos": 25558019625600.0,
      "grad_norm": 2.5171288989414684,
      "language_loss": 0.65436262,
      "learning_rate": 3.6254191934527854e-06,
      "loss": 0.67660439,
      "num_input_tokens_seen": 39314395,
      "step": 1847,
      "time_per_iteration": 2.594733476638794
    },
    {
      "auxiliary_loss_clip": 0.01168103,
      "auxiliary_loss_mlp": 0.01034382,
      "balance_loss_clip": 1.06054449,
      "balance_loss_mlp": 1.02395761,
      "epoch": 0.2222088619010401,
      "flos": 19318612677120.0,
      "grad_norm": 2.350690865316085,
      "language_loss": 0.65021193,
      "learning_rate": 3.6249651875809715e-06,
      "loss": 0.67223674,
      "num_input_tokens_seen": 39334275,
      "step": 1848,
      "time_per_iteration": 2.5676612854003906
    },
    {
      "auxiliary_loss_clip": 0.01182358,
      "auxiliary_loss_mlp": 0.01037923,
      "balance_loss_clip": 1.06233251,
      "balance_loss_mlp": 1.02767682,
      "epoch": 0.2223291047916792,
      "flos": 19099342103040.0,
      "grad_norm": 1.830306359786449,
      "language_loss": 0.89280736,
      "learning_rate": 3.62451093520323e-06,
      "loss": 0.91501009,
      "num_input_tokens_seen": 39352180,
      "step": 1849,
      "time_per_iteration": 2.5562593936920166
    },
    {
      "auxiliary_loss_clip": 0.01148038,
      "auxiliary_loss_mlp": 0.0103471,
      "balance_loss_clip": 1.05095196,
      "balance_loss_mlp": 1.02532244,
      "epoch": 0.22244934768231828,
      "flos": 20850418126080.0,
      "grad_norm": 1.9163221656370755,
      "language_loss": 0.90741706,
      "learning_rate": 3.6240564363884714e-06,
      "loss": 0.92924452,
      "num_input_tokens_seen": 39372125,
      "step": 1850,
      "time_per_iteration": 2.6089136600494385
    },
    {
      "auxiliary_loss_clip": 0.01199404,
      "auxiliary_loss_mlp": 0.01034082,
      "balance_loss_clip": 1.05825996,
      "balance_loss_mlp": 1.0242945,
      "epoch": 0.2225695905729574,
      "flos": 15632921111040.0,
      "grad_norm": 2.1256972724704863,
      "language_loss": 0.7026329,
      "learning_rate": 3.623601691205643e-06,
      "loss": 0.72496784,
      "num_input_tokens_seen": 39391200,
      "step": 1851,
      "time_per_iteration": 2.521548271179199
    },
    {
      "auxiliary_loss_clip": 0.01197021,
      "auxiliary_loss_mlp": 0.01032788,
      "balance_loss_clip": 1.06029177,
      "balance_loss_mlp": 1.023579,
      "epoch": 0.22268983346359647,
      "flos": 25373582265600.0,
      "grad_norm": 2.483894779579382,
      "language_loss": 0.82093096,
      "learning_rate": 3.623146699723729e-06,
      "loss": 0.84322906,
      "num_input_tokens_seen": 39410660,
      "step": 1852,
      "time_per_iteration": 2.550711154937744
    },
    {
      "auxiliary_loss_clip": 0.01186689,
      "auxiliary_loss_mlp": 0.01036369,
      "balance_loss_clip": 1.06509554,
      "balance_loss_mlp": 1.02679062,
      "epoch": 0.22281007635423555,
      "flos": 13261452359040.0,
      "grad_norm": 1.8539121793025382,
      "language_loss": 0.7756933,
      "learning_rate": 3.6226914620117507e-06,
      "loss": 0.79792386,
      "num_input_tokens_seen": 39429280,
      "step": 1853,
      "time_per_iteration": 2.5188686847686768
    },
    {
      "auxiliary_loss_clip": 0.01165419,
      "auxiliary_loss_mlp": 0.01033854,
      "balance_loss_clip": 1.05035377,
      "balance_loss_mlp": 1.02434659,
      "epoch": 0.22293031924487464,
      "flos": 15340536403200.0,
      "grad_norm": 1.9801601468344223,
      "language_loss": 0.80732369,
      "learning_rate": 3.622235978138768e-06,
      "loss": 0.82931644,
      "num_input_tokens_seen": 39446905,
      "step": 1854,
      "time_per_iteration": 2.5568552017211914
    },
    {
      "auxiliary_loss_clip": 0.01198843,
      "auxiliary_loss_mlp": 0.01033223,
      "balance_loss_clip": 1.0634228,
      "balance_loss_mlp": 1.02393627,
      "epoch": 0.22305056213551375,
      "flos": 22564649773440.0,
      "grad_norm": 2.4776494870249746,
      "language_loss": 0.81541133,
      "learning_rate": 3.621780248173877e-06,
      "loss": 0.83773196,
      "num_input_tokens_seen": 39465105,
      "step": 1855,
      "time_per_iteration": 2.530457019805908
    },
    {
      "auxiliary_loss_clip": 0.01095731,
      "auxiliary_loss_mlp": 0.01006594,
      "balance_loss_clip": 1.0273639,
      "balance_loss_mlp": 1.00416243,
      "epoch": 0.22317080502615283,
      "flos": 64880419887360.0,
      "grad_norm": 0.8291200037751248,
      "language_loss": 0.61084032,
      "learning_rate": 3.6213242721862125e-06,
      "loss": 0.63186353,
      "num_input_tokens_seen": 39523560,
      "step": 1856,
      "time_per_iteration": 3.130747079849243
    },
    {
      "auxiliary_loss_clip": 0.01175039,
      "auxiliary_loss_mlp": 0.01039374,
      "balance_loss_clip": 1.06037247,
      "balance_loss_mlp": 1.03040934,
      "epoch": 0.2232910479167919,
      "flos": 25775997310080.0,
      "grad_norm": 1.874202357832368,
      "language_loss": 0.75434506,
      "learning_rate": 3.620868050244945e-06,
      "loss": 0.77648914,
      "num_input_tokens_seen": 39544040,
      "step": 1857,
      "time_per_iteration": 2.582007646560669
    },
    {
      "auxiliary_loss_clip": 0.011778,
      "auxiliary_loss_mlp": 0.01045728,
      "balance_loss_clip": 1.05853522,
      "balance_loss_mlp": 1.03517795,
      "epoch": 0.22341129080743102,
      "flos": 23251799928960.0,
      "grad_norm": 1.7235808135200084,
      "language_loss": 0.7762146,
      "learning_rate": 3.6204115824192817e-06,
      "loss": 0.79844987,
      "num_input_tokens_seen": 39561515,
      "step": 1858,
      "time_per_iteration": 2.547947883605957
    },
    {
      "auxiliary_loss_clip": 0.01172614,
      "auxiliary_loss_mlp": 0.01033928,
      "balance_loss_clip": 1.05484819,
      "balance_loss_mlp": 1.02386642,
      "epoch": 0.2235315336980701,
      "flos": 21214552250880.0,
      "grad_norm": 2.523170148450079,
      "language_loss": 0.77555555,
      "learning_rate": 3.619954868778471e-06,
      "loss": 0.79762101,
      "num_input_tokens_seen": 39578210,
      "step": 1859,
      "time_per_iteration": 3.2807791233062744
    },
    {
      "auxiliary_loss_clip": 0.01182834,
      "auxiliary_loss_mlp": 0.01041229,
      "balance_loss_clip": 1.0572679,
      "balance_loss_mlp": 1.03146529,
      "epoch": 0.2236517765887092,
      "flos": 19901945548800.0,
      "grad_norm": 1.9687584915419636,
      "language_loss": 0.83095855,
      "learning_rate": 3.6194979093917944e-06,
      "loss": 0.85319918,
      "num_input_tokens_seen": 39597625,
      "step": 1860,
      "time_per_iteration": 2.5474274158477783
    },
    {
      "auxiliary_loss_clip": 0.01176393,
      "auxiliary_loss_mlp": 0.01047797,
      "balance_loss_clip": 1.05711746,
      "balance_loss_mlp": 1.03800404,
      "epoch": 0.22377201947934827,
      "flos": 23214847812480.0,
      "grad_norm": 1.8125389370462268,
      "language_loss": 0.87096012,
      "learning_rate": 3.6190407043285724e-06,
      "loss": 0.89320207,
      "num_input_tokens_seen": 39615360,
      "step": 1861,
      "time_per_iteration": 2.562239170074463
    },
    {
      "auxiliary_loss_clip": 0.01211864,
      "auxiliary_loss_mlp": 0.0104377,
      "balance_loss_clip": 1.06124127,
      "balance_loss_mlp": 1.03345847,
      "epoch": 0.22389226236998738,
      "flos": 26794244056320.0,
      "grad_norm": 1.8856752592588955,
      "language_loss": 0.75761425,
      "learning_rate": 3.618583253658163e-06,
      "loss": 0.78017056,
      "num_input_tokens_seen": 39635460,
      "step": 1862,
      "time_per_iteration": 2.5362656116485596
    },
    {
      "auxiliary_loss_clip": 0.01154768,
      "auxiliary_loss_mlp": 0.00763483,
      "balance_loss_clip": 1.05759501,
      "balance_loss_mlp": 1.00039923,
      "epoch": 0.22401250526062647,
      "flos": 24170359455360.0,
      "grad_norm": 1.796451844300776,
      "language_loss": 0.86452794,
      "learning_rate": 3.618125557449961e-06,
      "loss": 0.8837105,
      "num_input_tokens_seen": 39653515,
      "step": 1863,
      "time_per_iteration": 3.4278626441955566
    },
    {
      "auxiliary_loss_clip": 0.01197144,
      "auxiliary_loss_mlp": 0.01034193,
      "balance_loss_clip": 1.06355286,
      "balance_loss_mlp": 1.02499628,
      "epoch": 0.22413274815126555,
      "flos": 16759761649920.0,
      "grad_norm": 1.8411295110707282,
      "language_loss": 0.8285675,
      "learning_rate": 3.6176676157733983e-06,
      "loss": 0.85088086,
      "num_input_tokens_seen": 39668525,
      "step": 1864,
      "time_per_iteration": 2.4845263957977295
    },
    {
      "auxiliary_loss_clip": 0.01160131,
      "auxiliary_loss_mlp": 0.01044877,
      "balance_loss_clip": 1.05354643,
      "balance_loss_mlp": 1.03464842,
      "epoch": 0.22425299104190466,
      "flos": 21360205900800.0,
      "grad_norm": 2.1062668448182222,
      "language_loss": 0.75494969,
      "learning_rate": 3.6172094286979443e-06,
      "loss": 0.77699977,
      "num_input_tokens_seen": 39685895,
      "step": 1865,
      "time_per_iteration": 2.5725181102752686
    },
    {
      "auxiliary_loss_clip": 0.01180036,
      "auxiliary_loss_mlp": 0.01028055,
      "balance_loss_clip": 1.05538738,
      "balance_loss_mlp": 1.01862001,
      "epoch": 0.22437323393254374,
      "flos": 32165547108480.0,
      "grad_norm": 1.8129473885452336,
      "language_loss": 0.81321126,
      "learning_rate": 3.6167509962931064e-06,
      "loss": 0.83529216,
      "num_input_tokens_seen": 39711595,
      "step": 1866,
      "time_per_iteration": 2.6856613159179688
    },
    {
      "auxiliary_loss_clip": 0.01162677,
      "auxiliary_loss_mlp": 0.0104127,
      "balance_loss_clip": 1.05902338,
      "balance_loss_mlp": 1.03151894,
      "epoch": 0.22449347682318282,
      "flos": 18002809664640.0,
      "grad_norm": 2.9501471794501186,
      "language_loss": 0.77164179,
      "learning_rate": 3.6162923186284276e-06,
      "loss": 0.79368126,
      "num_input_tokens_seen": 39727555,
      "step": 1867,
      "time_per_iteration": 3.303394079208374
    },
    {
      "auxiliary_loss_clip": 0.01180947,
      "auxiliary_loss_mlp": 0.01038703,
      "balance_loss_clip": 1.05761194,
      "balance_loss_mlp": 1.02955937,
      "epoch": 0.2246137197138219,
      "flos": 18697286194560.0,
      "grad_norm": 2.1785657643537113,
      "language_loss": 0.85826862,
      "learning_rate": 3.6158333957734888e-06,
      "loss": 0.88046515,
      "num_input_tokens_seen": 39746145,
      "step": 1868,
      "time_per_iteration": 3.297990322113037
    },
    {
      "auxiliary_loss_clip": 0.01171434,
      "auxiliary_loss_mlp": 0.01034567,
      "balance_loss_clip": 1.05607045,
      "balance_loss_mlp": 1.02597797,
      "epoch": 0.22473396260446102,
      "flos": 15590653781760.0,
      "grad_norm": 2.109386333761803,
      "language_loss": 0.82767391,
      "learning_rate": 3.6153742277979088e-06,
      "loss": 0.84973395,
      "num_input_tokens_seen": 39763575,
      "step": 1869,
      "time_per_iteration": 2.5543501377105713
    },
    {
      "auxiliary_loss_clip": 0.01183716,
      "auxiliary_loss_mlp": 0.01041938,
      "balance_loss_clip": 1.05921221,
      "balance_loss_mlp": 1.03207898,
      "epoch": 0.2248542054951001,
      "flos": 14465501182080.0,
      "grad_norm": 2.4162409343010993,
      "language_loss": 0.78191817,
      "learning_rate": 3.6149148147713434e-06,
      "loss": 0.80417466,
      "num_input_tokens_seen": 39781810,
      "step": 1870,
      "time_per_iteration": 2.5225110054016113
    },
    {
      "auxiliary_loss_clip": 0.01205775,
      "auxiliary_loss_mlp": 0.01039485,
      "balance_loss_clip": 1.06619167,
      "balance_loss_mlp": 1.02989435,
      "epoch": 0.22497444838573918,
      "flos": 19243882431360.0,
      "grad_norm": 1.8313310016783932,
      "language_loss": 0.86329424,
      "learning_rate": 3.614455156763484e-06,
      "loss": 0.88574684,
      "num_input_tokens_seen": 39800115,
      "step": 1871,
      "time_per_iteration": 2.5161828994750977
    },
    {
      "auxiliary_loss_clip": 0.0114912,
      "auxiliary_loss_mlp": 0.01033337,
      "balance_loss_clip": 1.05112898,
      "balance_loss_mlp": 1.02375901,
      "epoch": 0.2250946912763783,
      "flos": 16910299549440.0,
      "grad_norm": 2.6437667958172617,
      "language_loss": 0.71468502,
      "learning_rate": 3.613995253844061e-06,
      "loss": 0.73650956,
      "num_input_tokens_seen": 39817795,
      "step": 1872,
      "time_per_iteration": 2.5896646976470947
    },
    {
      "auxiliary_loss_clip": 0.01192921,
      "auxiliary_loss_mlp": 0.01029331,
      "balance_loss_clip": 1.06008852,
      "balance_loss_mlp": 1.02030087,
      "epoch": 0.22521493416701738,
      "flos": 24681368292480.0,
      "grad_norm": 1.9038835185068268,
      "language_loss": 0.80674583,
      "learning_rate": 3.6135351060828414e-06,
      "loss": 0.82896835,
      "num_input_tokens_seen": 39838270,
      "step": 1873,
      "time_per_iteration": 2.551215887069702
    },
    {
      "auxiliary_loss_clip": 0.01214386,
      "auxiliary_loss_mlp": 0.01031308,
      "balance_loss_clip": 1.06327736,
      "balance_loss_mlp": 1.02124047,
      "epoch": 0.22533517705765646,
      "flos": 17821963664640.0,
      "grad_norm": 2.041245172053233,
      "language_loss": 0.69323742,
      "learning_rate": 3.6130747135496285e-06,
      "loss": 0.71569443,
      "num_input_tokens_seen": 39857270,
      "step": 1874,
      "time_per_iteration": 2.472790241241455
    },
    {
      "auxiliary_loss_clip": 0.01210305,
      "auxiliary_loss_mlp": 0.01037879,
      "balance_loss_clip": 1.06120753,
      "balance_loss_mlp": 1.02793658,
      "epoch": 0.22545541994829554,
      "flos": 33691390899840.0,
      "grad_norm": 1.90371822530015,
      "language_loss": 0.65816116,
      "learning_rate": 3.6126140763142646e-06,
      "loss": 0.68064302,
      "num_input_tokens_seen": 39882300,
      "step": 1875,
      "time_per_iteration": 2.589491605758667
    },
    {
      "auxiliary_loss_clip": 0.01211762,
      "auxiliary_loss_mlp": 0.01036997,
      "balance_loss_clip": 1.06267023,
      "balance_loss_mlp": 1.02692413,
      "epoch": 0.22557566283893465,
      "flos": 19171594310400.0,
      "grad_norm": 2.8709569794141694,
      "language_loss": 0.8641867,
      "learning_rate": 3.6121531944466275e-06,
      "loss": 0.88667434,
      "num_input_tokens_seen": 39899625,
      "step": 1876,
      "time_per_iteration": 2.472351312637329
    },
    {
      "auxiliary_loss_clip": 0.01196743,
      "auxiliary_loss_mlp": 0.01037134,
      "balance_loss_clip": 1.06118059,
      "balance_loss_mlp": 1.02831292,
      "epoch": 0.22569590572957374,
      "flos": 20773281669120.0,
      "grad_norm": 2.560105114809103,
      "language_loss": 0.78017592,
      "learning_rate": 3.611692068016633e-06,
      "loss": 0.80251467,
      "num_input_tokens_seen": 39915955,
      "step": 1877,
      "time_per_iteration": 2.503307342529297
    },
    {
      "auxiliary_loss_clip": 0.01162645,
      "auxiliary_loss_mlp": 0.01040152,
      "balance_loss_clip": 1.05454421,
      "balance_loss_mlp": 1.03038847,
      "epoch": 0.22581614862021282,
      "flos": 18442715529600.0,
      "grad_norm": 1.9942141732131287,
      "language_loss": 0.74857259,
      "learning_rate": 3.611230697094233e-06,
      "loss": 0.77060056,
      "num_input_tokens_seen": 39932655,
      "step": 1878,
      "time_per_iteration": 2.540508508682251
    },
    {
      "auxiliary_loss_clip": 0.01182474,
      "auxiliary_loss_mlp": 0.01037689,
      "balance_loss_clip": 1.05579758,
      "balance_loss_mlp": 1.02827168,
      "epoch": 0.22593639151085193,
      "flos": 20048389297920.0,
      "grad_norm": 1.887005556463023,
      "language_loss": 0.87388211,
      "learning_rate": 3.6107690817494173e-06,
      "loss": 0.89608371,
      "num_input_tokens_seen": 39952875,
      "step": 1879,
      "time_per_iteration": 2.54695463180542
    },
    {
      "auxiliary_loss_clip": 0.01145533,
      "auxiliary_loss_mlp": 0.01035844,
      "balance_loss_clip": 1.05218971,
      "balance_loss_mlp": 1.02654564,
      "epoch": 0.226056634401491,
      "flos": 13115116350720.0,
      "grad_norm": 2.171346060298478,
      "language_loss": 0.70407248,
      "learning_rate": 3.6103072220522117e-06,
      "loss": 0.72588629,
      "num_input_tokens_seen": 39968405,
      "step": 1880,
      "time_per_iteration": 2.5786023139953613
    },
    {
      "auxiliary_loss_clip": 0.01166809,
      "auxiliary_loss_mlp": 0.01041634,
      "balance_loss_clip": 1.05485153,
      "balance_loss_mlp": 1.03303885,
      "epoch": 0.2261768772921301,
      "flos": 18988378012800.0,
      "grad_norm": 1.7244562508886796,
      "language_loss": 0.9192915,
      "learning_rate": 3.609845118072682e-06,
      "loss": 0.94137591,
      "num_input_tokens_seen": 39987075,
      "step": 1881,
      "time_per_iteration": 2.572270393371582
    },
    {
      "auxiliary_loss_clip": 0.01201818,
      "auxiliary_loss_mlp": 0.00763669,
      "balance_loss_clip": 1.06059468,
      "balance_loss_mlp": 1.00038862,
      "epoch": 0.2262971201827692,
      "flos": 19974054101760.0,
      "grad_norm": 1.6692317203724176,
      "language_loss": 0.80261701,
      "learning_rate": 3.6093827698809276e-06,
      "loss": 0.82227188,
      "num_input_tokens_seen": 40006175,
      "step": 1882,
      "time_per_iteration": 2.5336966514587402
    },
    {
      "auxiliary_loss_clip": 0.0119022,
      "auxiliary_loss_mlp": 0.01026505,
      "balance_loss_clip": 1.05446768,
      "balance_loss_mlp": 1.01778483,
      "epoch": 0.2264173630734083,
      "flos": 16654543735680.0,
      "grad_norm": 2.4903815238412763,
      "language_loss": 0.84766835,
      "learning_rate": 3.6089201775470864e-06,
      "loss": 0.86983562,
      "num_input_tokens_seen": 40021630,
      "step": 1883,
      "time_per_iteration": 2.4835305213928223
    },
    {
      "auxiliary_loss_clip": 0.01154935,
      "auxiliary_loss_mlp": 0.01033683,
      "balance_loss_clip": 1.05408907,
      "balance_loss_mlp": 1.02467132,
      "epoch": 0.22653760596404737,
      "flos": 24389809597440.0,
      "grad_norm": 1.5347569096061404,
      "language_loss": 0.77418613,
      "learning_rate": 3.6084573411413334e-06,
      "loss": 0.79607236,
      "num_input_tokens_seen": 40041025,
      "step": 1884,
      "time_per_iteration": 2.632819414138794
    },
    {
      "auxiliary_loss_clip": 0.01166078,
      "auxiliary_loss_mlp": 0.01033809,
      "balance_loss_clip": 1.05730677,
      "balance_loss_mlp": 1.02471972,
      "epoch": 0.22665784885468646,
      "flos": 18332541538560.0,
      "grad_norm": 2.0933227833626766,
      "language_loss": 0.80962884,
      "learning_rate": 3.607994260733881e-06,
      "loss": 0.83162773,
      "num_input_tokens_seen": 40060265,
      "step": 1885,
      "time_per_iteration": 3.347014904022217
    },
    {
      "auxiliary_loss_clip": 0.01181819,
      "auxiliary_loss_mlp": 0.0103084,
      "balance_loss_clip": 1.05415034,
      "balance_loss_mlp": 1.02260828,
      "epoch": 0.22677809174532557,
      "flos": 24058102475520.0,
      "grad_norm": 1.6331163052628141,
      "language_loss": 0.74705458,
      "learning_rate": 3.6075309363949776e-06,
      "loss": 0.76918113,
      "num_input_tokens_seen": 40079435,
      "step": 1886,
      "time_per_iteration": 2.5312106609344482
    },
    {
      "auxiliary_loss_clip": 0.01208325,
      "auxiliary_loss_mlp": 0.01030971,
      "balance_loss_clip": 1.05939555,
      "balance_loss_mlp": 1.02145195,
      "epoch": 0.22689833463596465,
      "flos": 20374242503040.0,
      "grad_norm": 1.886772979988569,
      "language_loss": 0.81155455,
      "learning_rate": 3.6070673681949094e-06,
      "loss": 0.83394754,
      "num_input_tokens_seen": 40097800,
      "step": 1887,
      "time_per_iteration": 2.505464553833008
    },
    {
      "auxiliary_loss_clip": 0.01182526,
      "auxiliary_loss_mlp": 0.00762529,
      "balance_loss_clip": 1.0588901,
      "balance_loss_mlp": 1.00036085,
      "epoch": 0.22701857752660373,
      "flos": 30120398438400.0,
      "grad_norm": 1.6529379436572094,
      "language_loss": 0.81408787,
      "learning_rate": 3.606603556203999e-06,
      "loss": 0.83353841,
      "num_input_tokens_seen": 40122745,
      "step": 1888,
      "time_per_iteration": 2.6385016441345215
    },
    {
      "auxiliary_loss_clip": 0.01194917,
      "auxiliary_loss_mlp": 0.0104001,
      "balance_loss_clip": 1.05822968,
      "balance_loss_mlp": 1.03103924,
      "epoch": 0.22713882041724284,
      "flos": 22492182084480.0,
      "grad_norm": 2.1989570165227565,
      "language_loss": 0.83658361,
      "learning_rate": 3.6061395004926066e-06,
      "loss": 0.85893285,
      "num_input_tokens_seen": 40141680,
      "step": 1889,
      "time_per_iteration": 2.522618055343628
    },
    {
      "auxiliary_loss_clip": 0.01207661,
      "auxiliary_loss_mlp": 0.01035492,
      "balance_loss_clip": 1.05987263,
      "balance_loss_mlp": 1.02605116,
      "epoch": 0.22725906330788193,
      "flos": 20521548178560.0,
      "grad_norm": 2.848856048074611,
      "language_loss": 0.85260934,
      "learning_rate": 3.605675201131129e-06,
      "loss": 0.87504089,
      "num_input_tokens_seen": 40160140,
      "step": 1890,
      "time_per_iteration": 3.2511303424835205
    },
    {
      "auxiliary_loss_clip": 0.01197346,
      "auxiliary_loss_mlp": 0.01032462,
      "balance_loss_clip": 1.05915046,
      "balance_loss_mlp": 1.02300835,
      "epoch": 0.227379306198521,
      "flos": 18989922297600.0,
      "grad_norm": 2.197995193907923,
      "language_loss": 0.79822981,
      "learning_rate": 3.60521065819e-06,
      "loss": 0.82052791,
      "num_input_tokens_seen": 40177450,
      "step": 1891,
      "time_per_iteration": 2.4789669513702393
    },
    {
      "auxiliary_loss_clip": 0.01178226,
      "auxiliary_loss_mlp": 0.01030974,
      "balance_loss_clip": 1.05385184,
      "balance_loss_mlp": 1.02254057,
      "epoch": 0.2274995490891601,
      "flos": 21798351999360.0,
      "grad_norm": 1.9135752120304614,
      "language_loss": 0.87378722,
      "learning_rate": 3.60474587173969e-06,
      "loss": 0.89587921,
      "num_input_tokens_seen": 40195935,
      "step": 1892,
      "time_per_iteration": 2.5518712997436523
    },
    {
      "auxiliary_loss_clip": 0.01195431,
      "auxiliary_loss_mlp": 0.01031073,
      "balance_loss_clip": 1.06405771,
      "balance_loss_mlp": 1.02144659,
      "epoch": 0.2276197919797992,
      "flos": 19058654972160.0,
      "grad_norm": 1.880898699091292,
      "language_loss": 0.84186351,
      "learning_rate": 3.6042808418507084e-06,
      "loss": 0.86412859,
      "num_input_tokens_seen": 40213620,
      "step": 1893,
      "time_per_iteration": 3.2313616275787354
    },
    {
      "auxiliary_loss_clip": 0.01196099,
      "auxiliary_loss_mlp": 0.0103601,
      "balance_loss_clip": 1.06240106,
      "balance_loss_mlp": 1.02647877,
      "epoch": 0.22774003487043828,
      "flos": 18806777827200.0,
      "grad_norm": 2.4771437259053872,
      "language_loss": 0.76713437,
      "learning_rate": 3.6038155685935976e-06,
      "loss": 0.78945547,
      "num_input_tokens_seen": 40230190,
      "step": 1894,
      "time_per_iteration": 3.2739810943603516
    },
    {
      "auxiliary_loss_clip": 0.01191352,
      "auxiliary_loss_mlp": 0.01036591,
      "balance_loss_clip": 1.05800605,
      "balance_loss_mlp": 1.02672005,
      "epoch": 0.22786027776107737,
      "flos": 23002544476800.0,
      "grad_norm": 2.1053670696656934,
      "language_loss": 0.70590806,
      "learning_rate": 3.6033500520389404e-06,
      "loss": 0.72818756,
      "num_input_tokens_seen": 40246860,
      "step": 1895,
      "time_per_iteration": 2.5142710208892822
    },
    {
      "auxiliary_loss_clip": 0.01059769,
      "auxiliary_loss_mlp": 0.01002334,
      "balance_loss_clip": 1.0213356,
      "balance_loss_mlp": 1.00009286,
      "epoch": 0.22798052065171648,
      "flos": 66706872600960.0,
      "grad_norm": 0.7882888454372766,
      "language_loss": 0.6479944,
      "learning_rate": 3.6028842922573553e-06,
      "loss": 0.66861546,
      "num_input_tokens_seen": 40311005,
      "step": 1896,
      "time_per_iteration": 3.270555257797241
    },
    {
      "auxiliary_loss_clip": 0.01073782,
      "auxiliary_loss_mlp": 0.00753501,
      "balance_loss_clip": 1.02457738,
      "balance_loss_mlp": 1.00010252,
      "epoch": 0.22810076354235556,
      "flos": 62080896758400.0,
      "grad_norm": 0.8546828944597508,
      "language_loss": 0.62969607,
      "learning_rate": 3.602418289319497e-06,
      "loss": 0.64796889,
      "num_input_tokens_seen": 40369560,
      "step": 1897,
      "time_per_iteration": 3.1438465118408203
    },
    {
      "auxiliary_loss_clip": 0.01148279,
      "auxiliary_loss_mlp": 0.01034494,
      "balance_loss_clip": 1.0532012,
      "balance_loss_mlp": 1.02523744,
      "epoch": 0.22822100643299464,
      "flos": 23876358635520.0,
      "grad_norm": 2.5057844390170083,
      "language_loss": 0.72962904,
      "learning_rate": 3.601952043296059e-06,
      "loss": 0.75145674,
      "num_input_tokens_seen": 40389555,
      "step": 1898,
      "time_per_iteration": 2.612128734588623
    },
    {
      "auxiliary_loss_clip": 0.01184196,
      "auxiliary_loss_mlp": 0.01031288,
      "balance_loss_clip": 1.05551958,
      "balance_loss_mlp": 1.02153707,
      "epoch": 0.22834124932363373,
      "flos": 20991331180800.0,
      "grad_norm": 2.104706619519775,
      "language_loss": 0.80711555,
      "learning_rate": 3.6014855542577696e-06,
      "loss": 0.82927042,
      "num_input_tokens_seen": 40406765,
      "step": 1899,
      "time_per_iteration": 2.554781913757324
    },
    {
      "auxiliary_loss_clip": 0.01180515,
      "auxiliary_loss_mlp": 0.01041142,
      "balance_loss_clip": 1.05823493,
      "balance_loss_mlp": 1.03196335,
      "epoch": 0.22846149221427284,
      "flos": 24901572620160.0,
      "grad_norm": 1.7038358297561893,
      "language_loss": 0.84422112,
      "learning_rate": 3.6010188222753943e-06,
      "loss": 0.86643767,
      "num_input_tokens_seen": 40427535,
      "step": 1900,
      "time_per_iteration": 2.581376791000366
    },
    {
      "auxiliary_loss_clip": 0.01080726,
      "auxiliary_loss_mlp": 0.01002959,
      "balance_loss_clip": 1.02542686,
      "balance_loss_mlp": 1.00077713,
      "epoch": 0.22858173510491192,
      "flos": 56132294319360.0,
      "grad_norm": 0.9202511169432144,
      "language_loss": 0.64238048,
      "learning_rate": 3.6005518474197372e-06,
      "loss": 0.66321731,
      "num_input_tokens_seen": 40479580,
      "step": 1901,
      "time_per_iteration": 3.0441086292266846
    },
    {
      "auxiliary_loss_clip": 0.01194653,
      "auxiliary_loss_mlp": 0.01044283,
      "balance_loss_clip": 1.06113112,
      "balance_loss_mlp": 1.03403091,
      "epoch": 0.228701977995551,
      "flos": 24170826332160.0,
      "grad_norm": 2.248873604252607,
      "language_loss": 0.78329653,
      "learning_rate": 3.6000846297616373e-06,
      "loss": 0.80568588,
      "num_input_tokens_seen": 40497880,
      "step": 1902,
      "time_per_iteration": 2.5509328842163086
    },
    {
      "auxiliary_loss_clip": 0.01217675,
      "auxiliary_loss_mlp": 0.01036817,
      "balance_loss_clip": 1.0676347,
      "balance_loss_mlp": 1.02681541,
      "epoch": 0.22882222088619011,
      "flos": 21387892308480.0,
      "grad_norm": 2.0874817811671353,
      "language_loss": 0.72710383,
      "learning_rate": 3.5996171693719717e-06,
      "loss": 0.74964869,
      "num_input_tokens_seen": 40513975,
      "step": 1903,
      "time_per_iteration": 2.468595266342163
    },
    {
      "auxiliary_loss_clip": 0.01091794,
      "auxiliary_loss_mlp": 0.01007006,
      "balance_loss_clip": 1.02415943,
      "balance_loss_mlp": 1.00478864,
      "epoch": 0.2289424637768292,
      "flos": 64589615377920.0,
      "grad_norm": 0.9343989992060507,
      "language_loss": 0.64823055,
      "learning_rate": 3.5991494663216528e-06,
      "loss": 0.66921854,
      "num_input_tokens_seen": 40576960,
      "step": 1904,
      "time_per_iteration": 3.1764278411865234
    },
    {
      "auxiliary_loss_clip": 0.01211573,
      "auxiliary_loss_mlp": 0.01035965,
      "balance_loss_clip": 1.06358171,
      "balance_loss_mlp": 1.02651167,
      "epoch": 0.22906270666746828,
      "flos": 22163419877760.0,
      "grad_norm": 2.334156799380791,
      "language_loss": 0.87629145,
      "learning_rate": 3.5986815206816314e-06,
      "loss": 0.89876676,
      "num_input_tokens_seen": 40595780,
      "step": 1905,
      "time_per_iteration": 2.496854066848755
    },
    {
      "auxiliary_loss_clip": 0.01207554,
      "auxiliary_loss_mlp": 0.01037277,
      "balance_loss_clip": 1.05968499,
      "balance_loss_mlp": 1.02837217,
      "epoch": 0.2291829495581074,
      "flos": 25772334122880.0,
      "grad_norm": 1.7741217724551384,
      "language_loss": 0.74945891,
      "learning_rate": 3.598213332522895e-06,
      "loss": 0.77190721,
      "num_input_tokens_seen": 40615810,
      "step": 1906,
      "time_per_iteration": 2.544039249420166
    },
    {
      "auxiliary_loss_clip": 0.01192404,
      "auxiliary_loss_mlp": 0.01039873,
      "balance_loss_clip": 1.0583787,
      "balance_loss_mlp": 1.03071129,
      "epoch": 0.22930319244874647,
      "flos": 31172760126720.0,
      "grad_norm": 2.49109483612779,
      "language_loss": 0.77307087,
      "learning_rate": 3.597744901916466e-06,
      "loss": 0.79539359,
      "num_input_tokens_seen": 40637095,
      "step": 1907,
      "time_per_iteration": 2.5663156509399414
    },
    {
      "auxiliary_loss_clip": 0.01210178,
      "auxiliary_loss_mlp": 0.0103741,
      "balance_loss_clip": 1.05919099,
      "balance_loss_mlp": 1.02728868,
      "epoch": 0.22942343533938556,
      "flos": 23254098399360.0,
      "grad_norm": 2.091262418671253,
      "language_loss": 0.76855582,
      "learning_rate": 3.5972762289334058e-06,
      "loss": 0.79103172,
      "num_input_tokens_seen": 40656725,
      "step": 1908,
      "time_per_iteration": 2.4929656982421875
    },
    {
      "auxiliary_loss_clip": 0.0113097,
      "auxiliary_loss_mlp": 0.01041462,
      "balance_loss_clip": 1.05308437,
      "balance_loss_mlp": 1.03180635,
      "epoch": 0.22954367823002464,
      "flos": 14610903436800.0,
      "grad_norm": 1.9418205793081609,
      "language_loss": 0.84767354,
      "learning_rate": 3.5968073136448116e-06,
      "loss": 0.86939788,
      "num_input_tokens_seen": 40674745,
      "step": 1909,
      "time_per_iteration": 2.627415418624878
    },
    {
      "auxiliary_loss_clip": 0.01199274,
      "auxiliary_loss_mlp": 0.01032546,
      "balance_loss_clip": 1.06075549,
      "balance_loss_mlp": 1.02282417,
      "epoch": 0.22966392112066375,
      "flos": 16763604405120.0,
      "grad_norm": 2.1397466064419617,
      "language_loss": 0.9139207,
      "learning_rate": 3.596338156121818e-06,
      "loss": 0.93623888,
      "num_input_tokens_seen": 40693630,
      "step": 1910,
      "time_per_iteration": 2.4904751777648926
    },
    {
      "auxiliary_loss_clip": 0.01079282,
      "auxiliary_loss_mlp": 0.01002623,
      "balance_loss_clip": 1.02356744,
      "balance_loss_mlp": 1.00029814,
      "epoch": 0.22978416401130283,
      "flos": 67474247783040.0,
      "grad_norm": 0.762638861326994,
      "language_loss": 0.59357977,
      "learning_rate": 3.595868756435595e-06,
      "loss": 0.61439878,
      "num_input_tokens_seen": 40761310,
      "step": 1911,
      "time_per_iteration": 4.008214235305786
    },
    {
      "auxiliary_loss_clip": 0.01173355,
      "auxiliary_loss_mlp": 0.01038615,
      "balance_loss_clip": 1.06313705,
      "balance_loss_mlp": 1.02906632,
      "epoch": 0.22990440690194192,
      "flos": 19865137086720.0,
      "grad_norm": 2.19409478454432,
      "language_loss": 0.806615,
      "learning_rate": 3.5953991146573504e-06,
      "loss": 0.82873464,
      "num_input_tokens_seen": 40779955,
      "step": 1912,
      "time_per_iteration": 2.561598300933838
    },
    {
      "auxiliary_loss_clip": 0.01196481,
      "auxiliary_loss_mlp": 0.01040686,
      "balance_loss_clip": 1.05758238,
      "balance_loss_mlp": 1.03052902,
      "epoch": 0.23002464979258103,
      "flos": 13289246507520.0,
      "grad_norm": 2.2597784286697546,
      "language_loss": 0.83029336,
      "learning_rate": 3.5949292308583294e-06,
      "loss": 0.85266507,
      "num_input_tokens_seen": 40793200,
      "step": 1913,
      "time_per_iteration": 2.4531362056732178
    },
    {
      "auxiliary_loss_clip": 0.01211718,
      "auxiliary_loss_mlp": 0.01035532,
      "balance_loss_clip": 1.06396151,
      "balance_loss_mlp": 1.02577496,
      "epoch": 0.2301448926832201,
      "flos": 22163779013760.0,
      "grad_norm": 2.1135003960424195,
      "language_loss": 0.80764192,
      "learning_rate": 3.594459105109811e-06,
      "loss": 0.83011436,
      "num_input_tokens_seen": 40812380,
      "step": 1914,
      "time_per_iteration": 2.496546983718872
    },
    {
      "auxiliary_loss_clip": 0.01199574,
      "auxiliary_loss_mlp": 0.01036196,
      "balance_loss_clip": 1.06190777,
      "balance_loss_mlp": 1.02719593,
      "epoch": 0.2302651355738592,
      "flos": 20704477167360.0,
      "grad_norm": 1.7163182533266421,
      "language_loss": 0.81475717,
      "learning_rate": 3.593988737483115e-06,
      "loss": 0.83711481,
      "num_input_tokens_seen": 40832320,
      "step": 1915,
      "time_per_iteration": 2.499314308166504
    },
    {
      "auxiliary_loss_clip": 0.01184553,
      "auxiliary_loss_mlp": 0.01038583,
      "balance_loss_clip": 1.06150448,
      "balance_loss_mlp": 1.02909422,
      "epoch": 0.23038537846449827,
      "flos": 18588943797120.0,
      "grad_norm": 1.996662778407666,
      "language_loss": 0.78339398,
      "learning_rate": 3.5935181280495947e-06,
      "loss": 0.80562532,
      "num_input_tokens_seen": 40850900,
      "step": 1916,
      "time_per_iteration": 3.2908010482788086
    },
    {
      "auxiliary_loss_clip": 0.01078964,
      "auxiliary_loss_mlp": 0.01004563,
      "balance_loss_clip": 1.02784026,
      "balance_loss_mlp": 1.00224996,
      "epoch": 0.23050562135513739,
      "flos": 64224260190720.0,
      "grad_norm": 0.8106259318431462,
      "language_loss": 0.54262149,
      "learning_rate": 3.5930472768806412e-06,
      "loss": 0.56345677,
      "num_input_tokens_seen": 40909570,
      "step": 1917,
      "time_per_iteration": 3.1123931407928467
    },
    {
      "auxiliary_loss_clip": 0.0121057,
      "auxiliary_loss_mlp": 0.01037329,
      "balance_loss_clip": 1.0630039,
      "balance_loss_mlp": 1.02821565,
      "epoch": 0.23062586424577647,
      "flos": 17313396952320.0,
      "grad_norm": 1.8356560479041255,
      "language_loss": 0.77195567,
      "learning_rate": 3.5925761840476826e-06,
      "loss": 0.79443467,
      "num_input_tokens_seen": 40928180,
      "step": 1918,
      "time_per_iteration": 2.486764669418335
    },
    {
      "auxiliary_loss_clip": 0.01179679,
      "auxiliary_loss_mlp": 0.01034199,
      "balance_loss_clip": 1.06335425,
      "balance_loss_mlp": 1.02507997,
      "epoch": 0.23074610713641555,
      "flos": 27855979194240.0,
      "grad_norm": 1.8909966333300505,
      "language_loss": 0.80991161,
      "learning_rate": 3.592104849622183e-06,
      "loss": 0.83205032,
      "num_input_tokens_seen": 40950435,
      "step": 1919,
      "time_per_iteration": 3.27776837348938
    },
    {
      "auxiliary_loss_clip": 0.01142455,
      "auxiliary_loss_mlp": 0.01027256,
      "balance_loss_clip": 1.05472791,
      "balance_loss_mlp": 1.0178206,
      "epoch": 0.23086635002705466,
      "flos": 28841798937600.0,
      "grad_norm": 1.460238265629872,
      "language_loss": 0.73251581,
      "learning_rate": 3.591633273675644e-06,
      "loss": 0.75421298,
      "num_input_tokens_seen": 40972670,
      "step": 1920,
      "time_per_iteration": 3.42789888381958
    },
    {
      "auxiliary_loss_clip": 0.01061429,
      "auxiliary_loss_mlp": 0.01002131,
      "balance_loss_clip": 1.02862501,
      "balance_loss_mlp": 0.99994987,
      "epoch": 0.23098659291769374,
      "flos": 62923681566720.0,
      "grad_norm": 0.9280028185681791,
      "language_loss": 0.58187962,
      "learning_rate": 3.591161456279602e-06,
      "loss": 0.60251528,
      "num_input_tokens_seen": 41018215,
      "step": 1921,
      "time_per_iteration": 2.9824328422546387
    },
    {
      "auxiliary_loss_clip": 0.01187697,
      "auxiliary_loss_mlp": 0.01035829,
      "balance_loss_clip": 1.05890298,
      "balance_loss_mlp": 1.02608323,
      "epoch": 0.23110683580833283,
      "flos": 23476816679040.0,
      "grad_norm": 1.4965230820115563,
      "language_loss": 0.80543268,
      "learning_rate": 3.590689397505633e-06,
      "loss": 0.82766795,
      "num_input_tokens_seen": 41039125,
      "step": 1922,
      "time_per_iteration": 2.582834243774414
    },
    {
      "auxiliary_loss_clip": 0.01206983,
      "auxiliary_loss_mlp": 0.01038075,
      "balance_loss_clip": 1.06150413,
      "balance_loss_mlp": 1.02942085,
      "epoch": 0.2312270786989719,
      "flos": 27271066124160.0,
      "grad_norm": 1.9673967286647924,
      "language_loss": 0.86816901,
      "learning_rate": 3.590217097425347e-06,
      "loss": 0.89061958,
      "num_input_tokens_seen": 41059025,
      "step": 1923,
      "time_per_iteration": 2.5602033138275146
    },
    {
      "auxiliary_loss_clip": 0.01211423,
      "auxiliary_loss_mlp": 0.01035849,
      "balance_loss_clip": 1.0622586,
      "balance_loss_mlp": 1.02616942,
      "epoch": 0.23134732158961102,
      "flos": 13261344618240.0,
      "grad_norm": 2.0223802588773228,
      "language_loss": 0.7117269,
      "learning_rate": 3.589744556110391e-06,
      "loss": 0.73419964,
      "num_input_tokens_seen": 41077015,
      "step": 1924,
      "time_per_iteration": 2.4694690704345703
    },
    {
      "auxiliary_loss_clip": 0.01178777,
      "auxiliary_loss_mlp": 0.01041378,
      "balance_loss_clip": 1.05844867,
      "balance_loss_mlp": 1.03084612,
      "epoch": 0.2314675644802501,
      "flos": 36977648250240.0,
      "grad_norm": 1.7101525904267878,
      "language_loss": 0.84407687,
      "learning_rate": 3.58927177363245e-06,
      "loss": 0.86627847,
      "num_input_tokens_seen": 41099840,
      "step": 1925,
      "time_per_iteration": 2.691822052001953
    },
    {
      "auxiliary_loss_clip": 0.01161829,
      "auxiliary_loss_mlp": 0.01037412,
      "balance_loss_clip": 1.05437136,
      "balance_loss_mlp": 1.02656949,
      "epoch": 0.2315878073708892,
      "flos": 23842207779840.0,
      "grad_norm": 2.042121981488652,
      "language_loss": 0.72524631,
      "learning_rate": 3.5887987500632447e-06,
      "loss": 0.74723876,
      "num_input_tokens_seen": 41117845,
      "step": 1926,
      "time_per_iteration": 2.600806951522827
    },
    {
      "auxiliary_loss_clip": 0.0117458,
      "auxiliary_loss_mlp": 0.01035676,
      "balance_loss_clip": 1.0590018,
      "balance_loss_mlp": 1.02722359,
      "epoch": 0.2317080502615283,
      "flos": 23039424766080.0,
      "grad_norm": 1.7192930999374594,
      "language_loss": 0.84200466,
      "learning_rate": 3.5883254854745325e-06,
      "loss": 0.86410713,
      "num_input_tokens_seen": 41136235,
      "step": 1927,
      "time_per_iteration": 2.5778889656066895
    },
    {
      "auxiliary_loss_clip": 0.01200773,
      "auxiliary_loss_mlp": 0.01030972,
      "balance_loss_clip": 1.06021976,
      "balance_loss_mlp": 1.02112532,
      "epoch": 0.23182829315216738,
      "flos": 11254656435840.0,
      "grad_norm": 1.980152521474129,
      "language_loss": 0.74838573,
      "learning_rate": 3.587851979938107e-06,
      "loss": 0.7707032,
      "num_input_tokens_seen": 41153125,
      "step": 1928,
      "time_per_iteration": 2.4899919033050537
    },
    {
      "auxiliary_loss_clip": 0.01195319,
      "auxiliary_loss_mlp": 0.01043093,
      "balance_loss_clip": 1.06158829,
      "balance_loss_mlp": 1.03388977,
      "epoch": 0.23194853604280646,
      "flos": 19828939155840.0,
      "grad_norm": 2.15368063128977,
      "language_loss": 0.77221417,
      "learning_rate": 3.5873782335257985e-06,
      "loss": 0.79459834,
      "num_input_tokens_seen": 41171290,
      "step": 1929,
      "time_per_iteration": 2.4979662895202637
    },
    {
      "auxiliary_loss_clip": 0.01165972,
      "auxiliary_loss_mlp": 0.01042276,
      "balance_loss_clip": 1.05931294,
      "balance_loss_mlp": 1.03238177,
      "epoch": 0.23206877893344555,
      "flos": 15305020830720.0,
      "grad_norm": 2.2882219996758675,
      "language_loss": 0.78638667,
      "learning_rate": 3.5869042463094744e-06,
      "loss": 0.80846918,
      "num_input_tokens_seen": 41189005,
      "step": 1930,
      "time_per_iteration": 2.5520195960998535
    },
    {
      "auxiliary_loss_clip": 0.0113308,
      "auxiliary_loss_mlp": 0.010422,
      "balance_loss_clip": 1.0513823,
      "balance_loss_mlp": 1.03228736,
      "epoch": 0.23218902182408466,
      "flos": 22711488572160.0,
      "grad_norm": 1.934915508582788,
      "language_loss": 0.77237588,
      "learning_rate": 3.586430018361038e-06,
      "loss": 0.79412866,
      "num_input_tokens_seen": 41208775,
      "step": 1931,
      "time_per_iteration": 2.607043981552124
    },
    {
      "auxiliary_loss_clip": 0.01165191,
      "auxiliary_loss_mlp": 0.01036153,
      "balance_loss_clip": 1.05443096,
      "balance_loss_mlp": 1.02593708,
      "epoch": 0.23230926471472374,
      "flos": 22710734386560.0,
      "grad_norm": 2.3843396234048755,
      "language_loss": 0.76468837,
      "learning_rate": 3.5859555497524283e-06,
      "loss": 0.78670192,
      "num_input_tokens_seen": 41226010,
      "step": 1932,
      "time_per_iteration": 2.5569307804107666
    },
    {
      "auxiliary_loss_clip": 0.01200543,
      "auxiliary_loss_mlp": 0.01041051,
      "balance_loss_clip": 1.06632686,
      "balance_loss_mlp": 1.03157997,
      "epoch": 0.23242950760536282,
      "flos": 20375499479040.0,
      "grad_norm": 2.0212908925038517,
      "language_loss": 0.92140174,
      "learning_rate": 3.5854808405556237e-06,
      "loss": 0.94381762,
      "num_input_tokens_seen": 41245245,
      "step": 1933,
      "time_per_iteration": 2.5480642318725586
    },
    {
      "auxiliary_loss_clip": 0.01167925,
      "auxiliary_loss_mlp": 0.0103645,
      "balance_loss_clip": 1.05726433,
      "balance_loss_mlp": 1.02705622,
      "epoch": 0.23254975049600193,
      "flos": 16908324301440.0,
      "grad_norm": 2.274561139925823,
      "language_loss": 0.75617975,
      "learning_rate": 3.5850058908426355e-06,
      "loss": 0.77822351,
      "num_input_tokens_seen": 41263795,
      "step": 1934,
      "time_per_iteration": 2.541945695877075
    },
    {
      "auxiliary_loss_clip": 0.01183573,
      "auxiliary_loss_mlp": 0.01035274,
      "balance_loss_clip": 1.05652523,
      "balance_loss_mlp": 1.02622652,
      "epoch": 0.23266999338664102,
      "flos": 23294821443840.0,
      "grad_norm": 1.8009652748784892,
      "language_loss": 0.85681021,
      "learning_rate": 3.584530700685514e-06,
      "loss": 0.87899864,
      "num_input_tokens_seen": 41284055,
      "step": 1935,
      "time_per_iteration": 2.5566775798797607
    },
    {
      "auxiliary_loss_clip": 0.01178952,
      "auxiliary_loss_mlp": 0.0103645,
      "balance_loss_clip": 1.06225753,
      "balance_loss_mlp": 1.02758741,
      "epoch": 0.2327902362772801,
      "flos": 19569987031680.0,
      "grad_norm": 1.9080419084612432,
      "language_loss": 0.88682699,
      "learning_rate": 3.5840552701563448e-06,
      "loss": 0.90898108,
      "num_input_tokens_seen": 41300255,
      "step": 1936,
      "time_per_iteration": 2.5391664505004883
    },
    {
      "auxiliary_loss_clip": 0.01210769,
      "auxiliary_loss_mlp": 0.0104034,
      "balance_loss_clip": 1.06272459,
      "balance_loss_mlp": 1.03049958,
      "epoch": 0.2329104791679192,
      "flos": 16727514215040.0,
      "grad_norm": 2.4758434551127393,
      "language_loss": 0.81664717,
      "learning_rate": 3.5835795993272513e-06,
      "loss": 0.8391583,
      "num_input_tokens_seen": 41318540,
      "step": 1937,
      "time_per_iteration": 3.1778643131256104
    },
    {
      "auxiliary_loss_clip": 0.0109946,
      "auxiliary_loss_mlp": 0.01037017,
      "balance_loss_clip": 1.04839802,
      "balance_loss_mlp": 1.02793956,
      "epoch": 0.2330307220585583,
      "flos": 22163743100160.0,
      "grad_norm": 2.4643756857755426,
      "language_loss": 0.71002418,
      "learning_rate": 3.583103688270391e-06,
      "loss": 0.73138893,
      "num_input_tokens_seen": 41338320,
      "step": 1938,
      "time_per_iteration": 2.7549662590026855
    },
    {
      "auxiliary_loss_clip": 0.01166826,
      "auxiliary_loss_mlp": 0.01034104,
      "balance_loss_clip": 1.05527496,
      "balance_loss_mlp": 1.02428102,
      "epoch": 0.23315096494919738,
      "flos": 19317319787520.0,
      "grad_norm": 2.093965644250405,
      "language_loss": 0.89367807,
      "learning_rate": 3.58262753705796e-06,
      "loss": 0.91568738,
      "num_input_tokens_seen": 41353210,
      "step": 1939,
      "time_per_iteration": 2.7021915912628174
    },
    {
      "auxiliary_loss_clip": 0.01076783,
      "auxiliary_loss_mlp": 0.01007671,
      "balance_loss_clip": 1.02579188,
      "balance_loss_mlp": 1.00554872,
      "epoch": 0.23327120783983646,
      "flos": 53031048946560.0,
      "grad_norm": 0.7598811789932318,
      "language_loss": 0.55548251,
      "learning_rate": 3.5821511457621902e-06,
      "loss": 0.57632697,
      "num_input_tokens_seen": 41410510,
      "step": 1940,
      "time_per_iteration": 3.112018585205078
    },
    {
      "auxiliary_loss_clip": 0.01173182,
      "auxiliary_loss_mlp": 0.01033204,
      "balance_loss_clip": 1.05838549,
      "balance_loss_mlp": 1.02259433,
      "epoch": 0.23339145073047557,
      "flos": 17126984344320.0,
      "grad_norm": 13.854412402286249,
      "language_loss": 0.81403542,
      "learning_rate": 3.5816745144553497e-06,
      "loss": 0.83609927,
      "num_input_tokens_seen": 41425830,
      "step": 1941,
      "time_per_iteration": 2.5040903091430664
    },
    {
      "auxiliary_loss_clip": 0.01146931,
      "auxiliary_loss_mlp": 0.01038403,
      "balance_loss_clip": 1.05569947,
      "balance_loss_mlp": 1.02922428,
      "epoch": 0.23351169362111465,
      "flos": 13078918419840.0,
      "grad_norm": 2.4082427788422778,
      "language_loss": 0.75463349,
      "learning_rate": 3.5811976432097424e-06,
      "loss": 0.77648687,
      "num_input_tokens_seen": 41443500,
      "step": 1942,
      "time_per_iteration": 3.5705108642578125
    },
    {
      "auxiliary_loss_clip": 0.01195404,
      "auxiliary_loss_mlp": 0.00762539,
      "balance_loss_clip": 1.06213403,
      "balance_loss_mlp": 1.00027013,
      "epoch": 0.23363193651175373,
      "flos": 15851257931520.0,
      "grad_norm": 1.968659523671831,
      "language_loss": 0.844751,
      "learning_rate": 3.58072053209771e-06,
      "loss": 0.86433047,
      "num_input_tokens_seen": 41460055,
      "step": 1943,
      "time_per_iteration": 2.530427932739258
    },
    {
      "auxiliary_loss_clip": 0.01172055,
      "auxiliary_loss_mlp": 0.01044887,
      "balance_loss_clip": 1.05411434,
      "balance_loss_mlp": 1.03497469,
      "epoch": 0.23375217940239285,
      "flos": 21025769345280.0,
      "grad_norm": 1.902372856775736,
      "language_loss": 0.79027116,
      "learning_rate": 3.5802431811916296e-06,
      "loss": 0.81244051,
      "num_input_tokens_seen": 41476665,
      "step": 1944,
      "time_per_iteration": 3.267354726791382
    },
    {
      "auxiliary_loss_clip": 0.01177067,
      "auxiliary_loss_mlp": 0.01032203,
      "balance_loss_clip": 1.05877495,
      "balance_loss_mlp": 1.0234884,
      "epoch": 0.23387242229303193,
      "flos": 20594698225920.0,
      "grad_norm": 1.6536784843099788,
      "language_loss": 0.80555797,
      "learning_rate": 3.579765590563916e-06,
      "loss": 0.82765067,
      "num_input_tokens_seen": 41496065,
      "step": 1945,
      "time_per_iteration": 2.5258796215057373
    },
    {
      "auxiliary_loss_clip": 0.01183271,
      "auxiliary_loss_mlp": 0.01030168,
      "balance_loss_clip": 1.0581584,
      "balance_loss_mlp": 1.02128744,
      "epoch": 0.233992665183671,
      "flos": 24279491952000.0,
      "grad_norm": 1.788488310365844,
      "language_loss": 0.81923926,
      "learning_rate": 3.579287760287017e-06,
      "loss": 0.84137368,
      "num_input_tokens_seen": 41516815,
      "step": 1946,
      "time_per_iteration": 3.2344515323638916
    },
    {
      "auxiliary_loss_clip": 0.01191634,
      "auxiliary_loss_mlp": 0.01030933,
      "balance_loss_clip": 1.06104338,
      "balance_loss_mlp": 1.02223706,
      "epoch": 0.2341129080743101,
      "flos": 30154621121280.0,
      "grad_norm": 1.997742852629854,
      "language_loss": 0.72768056,
      "learning_rate": 3.578809690433421e-06,
      "loss": 0.74990624,
      "num_input_tokens_seen": 41538525,
      "step": 1947,
      "time_per_iteration": 2.5673327445983887
    },
    {
      "auxiliary_loss_clip": 0.01213955,
      "auxiliary_loss_mlp": 0.01039986,
      "balance_loss_clip": 1.06467271,
      "balance_loss_mlp": 1.02944827,
      "epoch": 0.2342331509649492,
      "flos": 22784135829120.0,
      "grad_norm": 2.453620508513181,
      "language_loss": 0.80996895,
      "learning_rate": 3.578331381075651e-06,
      "loss": 0.83250839,
      "num_input_tokens_seen": 41559025,
      "step": 1948,
      "time_per_iteration": 2.4825472831726074
    },
    {
      "auxiliary_loss_clip": 0.01193273,
      "auxiliary_loss_mlp": 0.01039617,
      "balance_loss_clip": 1.05786991,
      "balance_loss_mlp": 1.02955592,
      "epoch": 0.2343533938555883,
      "flos": 23623152687360.0,
      "grad_norm": 3.0717023312988476,
      "language_loss": 0.69698071,
      "learning_rate": 3.5778528322862646e-06,
      "loss": 0.71930957,
      "num_input_tokens_seen": 41577845,
      "step": 1949,
      "time_per_iteration": 2.512026071548462
    },
    {
      "auxiliary_loss_clip": 0.01195439,
      "auxiliary_loss_mlp": 0.01032487,
      "balance_loss_clip": 1.05953264,
      "balance_loss_mlp": 1.02398121,
      "epoch": 0.23447363674622737,
      "flos": 24570332375040.0,
      "grad_norm": 1.516010526340661,
      "language_loss": 0.86413616,
      "learning_rate": 3.5773740441378585e-06,
      "loss": 0.88641536,
      "num_input_tokens_seen": 41598600,
      "step": 1950,
      "time_per_iteration": 2.537116765975952
    },
    {
      "auxiliary_loss_clip": 0.01190885,
      "auxiliary_loss_mlp": 0.01031584,
      "balance_loss_clip": 1.05854762,
      "balance_loss_mlp": 1.02264929,
      "epoch": 0.23459387963686648,
      "flos": 53140322119680.0,
      "grad_norm": 2.240994932692058,
      "language_loss": 0.73923457,
      "learning_rate": 3.5768950167030633e-06,
      "loss": 0.76145923,
      "num_input_tokens_seen": 41623300,
      "step": 1951,
      "time_per_iteration": 2.8036646842956543
    },
    {
      "auxiliary_loss_clip": 0.01163753,
      "auxiliary_loss_mlp": 0.01032652,
      "balance_loss_clip": 1.05149293,
      "balance_loss_mlp": 1.02377105,
      "epoch": 0.23471412252750556,
      "flos": 23951412103680.0,
      "grad_norm": 1.7475957464706764,
      "language_loss": 0.78225642,
      "learning_rate": 3.576415750054548e-06,
      "loss": 0.80422044,
      "num_input_tokens_seen": 41643420,
      "step": 1952,
      "time_per_iteration": 2.5627200603485107
    },
    {
      "auxiliary_loss_clip": 0.01169234,
      "auxiliary_loss_mlp": 0.01037377,
      "balance_loss_clip": 1.05642009,
      "balance_loss_mlp": 1.0279479,
      "epoch": 0.23483436541814465,
      "flos": 15706573948800.0,
      "grad_norm": 2.283613778845595,
      "language_loss": 0.85904598,
      "learning_rate": 3.5759362442650172e-06,
      "loss": 0.8811121,
      "num_input_tokens_seen": 41660170,
      "step": 1953,
      "time_per_iteration": 2.5083799362182617
    },
    {
      "auxiliary_loss_clip": 0.01190633,
      "auxiliary_loss_mlp": 0.01031823,
      "balance_loss_clip": 1.05948448,
      "balance_loss_mlp": 1.02261388,
      "epoch": 0.23495460830878373,
      "flos": 24936262179840.0,
      "grad_norm": 2.1122007247560655,
      "language_loss": 0.85518754,
      "learning_rate": 3.5754564994072113e-06,
      "loss": 0.87741208,
      "num_input_tokens_seen": 41679010,
      "step": 1954,
      "time_per_iteration": 2.550895929336548
    },
    {
      "auxiliary_loss_clip": 0.01177539,
      "auxiliary_loss_mlp": 0.01032101,
      "balance_loss_clip": 1.05731463,
      "balance_loss_mlp": 1.02307677,
      "epoch": 0.23507485119942284,
      "flos": 30482665056000.0,
      "grad_norm": 1.8658319757544783,
      "language_loss": 0.60132426,
      "learning_rate": 3.5749765155539067e-06,
      "loss": 0.62342072,
      "num_input_tokens_seen": 41699495,
      "step": 1955,
      "time_per_iteration": 2.627254009246826
    },
    {
      "auxiliary_loss_clip": 0.01163213,
      "auxiliary_loss_mlp": 0.01033247,
      "balance_loss_clip": 1.05547774,
      "balance_loss_mlp": 1.02401376,
      "epoch": 0.23519509409006192,
      "flos": 18329129746560.0,
      "grad_norm": 2.371587653648976,
      "language_loss": 0.92172652,
      "learning_rate": 3.574496292777917e-06,
      "loss": 0.94369113,
      "num_input_tokens_seen": 41717705,
      "step": 1956,
      "time_per_iteration": 2.5463154315948486
    },
    {
      "auxiliary_loss_clip": 0.0118499,
      "auxiliary_loss_mlp": 0.01036919,
      "balance_loss_clip": 1.05857241,
      "balance_loss_mlp": 1.0264461,
      "epoch": 0.235315336980701,
      "flos": 29643217234560.0,
      "grad_norm": 2.016940641086042,
      "language_loss": 0.71546453,
      "learning_rate": 3.574015831152092e-06,
      "loss": 0.73768353,
      "num_input_tokens_seen": 41738120,
      "step": 1957,
      "time_per_iteration": 2.615809440612793
    },
    {
      "auxiliary_loss_clip": 0.01169963,
      "auxiliary_loss_mlp": 0.01032542,
      "balance_loss_clip": 1.05624402,
      "balance_loss_mlp": 1.02342296,
      "epoch": 0.23543557987134012,
      "flos": 18551704371840.0,
      "grad_norm": 2.7616166720564794,
      "language_loss": 0.83268374,
      "learning_rate": 3.573535130749316e-06,
      "loss": 0.85470885,
      "num_input_tokens_seen": 41756070,
      "step": 1958,
      "time_per_iteration": 2.5327565670013428
    },
    {
      "auxiliary_loss_clip": 0.01171524,
      "auxiliary_loss_mlp": 0.01034153,
      "balance_loss_clip": 1.05801272,
      "balance_loss_mlp": 1.02512288,
      "epoch": 0.2355558227619792,
      "flos": 24679033908480.0,
      "grad_norm": 1.9616547164735172,
      "language_loss": 0.73833132,
      "learning_rate": 3.5730541916425127e-06,
      "loss": 0.76038814,
      "num_input_tokens_seen": 41777550,
      "step": 1959,
      "time_per_iteration": 2.5775485038757324
    },
    {
      "auxiliary_loss_clip": 0.01164328,
      "auxiliary_loss_mlp": 0.01038693,
      "balance_loss_clip": 1.05340123,
      "balance_loss_mlp": 1.02992558,
      "epoch": 0.23567606565261828,
      "flos": 21944795748480.0,
      "grad_norm": 1.9065976588192277,
      "language_loss": 0.85866898,
      "learning_rate": 3.572573013904639e-06,
      "loss": 0.88069922,
      "num_input_tokens_seen": 41797460,
      "step": 1960,
      "time_per_iteration": 2.581597328186035
    },
    {
      "auxiliary_loss_clip": 0.01208691,
      "auxiliary_loss_mlp": 0.01035347,
      "balance_loss_clip": 1.06149447,
      "balance_loss_mlp": 1.02654314,
      "epoch": 0.2357963085432574,
      "flos": 13589352639360.0,
      "grad_norm": 1.9331691385627179,
      "language_loss": 0.91975331,
      "learning_rate": 3.572091597608689e-06,
      "loss": 0.94219375,
      "num_input_tokens_seen": 41815585,
      "step": 1961,
      "time_per_iteration": 2.4540953636169434
    },
    {
      "auxiliary_loss_clip": 0.01181341,
      "auxiliary_loss_mlp": 0.01034676,
      "balance_loss_clip": 1.05820346,
      "balance_loss_mlp": 1.02446032,
      "epoch": 0.23591655143389648,
      "flos": 22088689632000.0,
      "grad_norm": 3.076326390542941,
      "language_loss": 0.73454905,
      "learning_rate": 3.571609942827694e-06,
      "loss": 0.75670922,
      "num_input_tokens_seen": 41834700,
      "step": 1962,
      "time_per_iteration": 2.5505731105804443
    },
    {
      "auxiliary_loss_clip": 0.01174543,
      "auxiliary_loss_mlp": 0.01034605,
      "balance_loss_clip": 1.05405724,
      "balance_loss_mlp": 1.02540207,
      "epoch": 0.23603679432453556,
      "flos": 17017349057280.0,
      "grad_norm": 1.8026248125479925,
      "language_loss": 0.88578212,
      "learning_rate": 3.57112804963472e-06,
      "loss": 0.90787363,
      "num_input_tokens_seen": 41852915,
      "step": 1963,
      "time_per_iteration": 3.2919487953186035
    },
    {
      "auxiliary_loss_clip": 0.0115916,
      "auxiliary_loss_mlp": 0.01031258,
      "balance_loss_clip": 1.05656552,
      "balance_loss_mlp": 1.02246654,
      "epoch": 0.23615703721517464,
      "flos": 19171307001600.0,
      "grad_norm": 2.064159886850844,
      "language_loss": 0.76430243,
      "learning_rate": 3.57064591810287e-06,
      "loss": 0.78620666,
      "num_input_tokens_seen": 41870415,
      "step": 1964,
      "time_per_iteration": 2.523857593536377
    },
    {
      "auxiliary_loss_clip": 0.01207274,
      "auxiliary_loss_mlp": 0.00762472,
      "balance_loss_clip": 1.06205595,
      "balance_loss_mlp": 1.00028455,
      "epoch": 0.23627728010581375,
      "flos": 19098803399040.0,
      "grad_norm": 2.080113470344047,
      "language_loss": 0.80525947,
      "learning_rate": 3.570163548305284e-06,
      "loss": 0.82495689,
      "num_input_tokens_seen": 41889345,
      "step": 1965,
      "time_per_iteration": 2.500753164291382
    },
    {
      "auxiliary_loss_clip": 0.01176424,
      "auxiliary_loss_mlp": 0.01042409,
      "balance_loss_clip": 1.0567956,
      "balance_loss_mlp": 1.03224087,
      "epoch": 0.23639752299645284,
      "flos": 14282213057280.0,
      "grad_norm": 2.1019603884833056,
      "language_loss": 0.70253462,
      "learning_rate": 3.569680940315135e-06,
      "loss": 0.72472292,
      "num_input_tokens_seen": 41905745,
      "step": 1966,
      "time_per_iteration": 2.4957430362701416
    },
    {
      "auxiliary_loss_clip": 0.01169035,
      "auxiliary_loss_mlp": 0.01035293,
      "balance_loss_clip": 1.05685401,
      "balance_loss_mlp": 1.02512431,
      "epoch": 0.23651776588709192,
      "flos": 22893411980160.0,
      "grad_norm": 1.8038779279667028,
      "language_loss": 0.81704974,
      "learning_rate": 3.5691980942056356e-06,
      "loss": 0.83909303,
      "num_input_tokens_seen": 41925115,
      "step": 1967,
      "time_per_iteration": 2.568981170654297
    },
    {
      "auxiliary_loss_clip": 0.01194895,
      "auxiliary_loss_mlp": 0.01031849,
      "balance_loss_clip": 1.05738807,
      "balance_loss_mlp": 1.02254474,
      "epoch": 0.23663800877773103,
      "flos": 18624531196800.0,
      "grad_norm": 1.9710045497677662,
      "language_loss": 0.79586959,
      "learning_rate": 3.5687150100500332e-06,
      "loss": 0.81813693,
      "num_input_tokens_seen": 41944815,
      "step": 1968,
      "time_per_iteration": 3.2783961296081543
    },
    {
      "auxiliary_loss_clip": 0.01195053,
      "auxiliary_loss_mlp": 0.01034209,
      "balance_loss_clip": 1.05926847,
      "balance_loss_mlp": 1.02483881,
      "epoch": 0.2367582516683701,
      "flos": 25555828896000.0,
      "grad_norm": 1.5459793737879222,
      "language_loss": 0.74555248,
      "learning_rate": 3.568231687921611e-06,
      "loss": 0.76784509,
      "num_input_tokens_seen": 41964990,
      "step": 1969,
      "time_per_iteration": 2.5677361488342285
    },
    {
      "auxiliary_loss_clip": 0.01203026,
      "auxiliary_loss_mlp": 0.01038998,
      "balance_loss_clip": 1.05865788,
      "balance_loss_mlp": 1.03006363,
      "epoch": 0.2368784945590092,
      "flos": 23295072839040.0,
      "grad_norm": 1.5176859319574119,
      "language_loss": 0.80185992,
      "learning_rate": 3.5677481278936883e-06,
      "loss": 0.82428014,
      "num_input_tokens_seen": 41984570,
      "step": 1970,
      "time_per_iteration": 3.209676742553711
    },
    {
      "auxiliary_loss_clip": 0.01087552,
      "auxiliary_loss_mlp": 0.01003677,
      "balance_loss_clip": 1.03652489,
      "balance_loss_mlp": 1.00154293,
      "epoch": 0.23699873744964828,
      "flos": 69859291875840.0,
      "grad_norm": 0.826442959344438,
      "language_loss": 0.57792377,
      "learning_rate": 3.5672643300396214e-06,
      "loss": 0.59883606,
      "num_input_tokens_seen": 42053715,
      "step": 1971,
      "time_per_iteration": 3.1931276321411133
    },
    {
      "auxiliary_loss_clip": 0.01162301,
      "auxiliary_loss_mlp": 0.01035751,
      "balance_loss_clip": 1.05534077,
      "balance_loss_mlp": 1.02682185,
      "epoch": 0.2371189803402874,
      "flos": 21835052720640.0,
      "grad_norm": 2.2883617359961472,
      "language_loss": 0.67528361,
      "learning_rate": 3.566780294432802e-06,
      "loss": 0.69726413,
      "num_input_tokens_seen": 42070890,
      "step": 1972,
      "time_per_iteration": 3.3917527198791504
    },
    {
      "auxiliary_loss_clip": 0.01207256,
      "auxiliary_loss_mlp": 0.01030354,
      "balance_loss_clip": 1.06216753,
      "balance_loss_mlp": 1.02140689,
      "epoch": 0.23723922323092647,
      "flos": 21908490076800.0,
      "grad_norm": 2.3650443359334927,
      "language_loss": 0.75220799,
      "learning_rate": 3.566296021146657e-06,
      "loss": 0.77458411,
      "num_input_tokens_seen": 42090270,
      "step": 1973,
      "time_per_iteration": 2.4791407585144043
    },
    {
      "auxiliary_loss_clip": 0.01208808,
      "auxiliary_loss_mlp": 0.01036458,
      "balance_loss_clip": 1.06155145,
      "balance_loss_mlp": 1.02676046,
      "epoch": 0.23735946612156555,
      "flos": 32708803380480.0,
      "grad_norm": 1.7747653126593388,
      "language_loss": 0.73194784,
      "learning_rate": 3.565811510254652e-06,
      "loss": 0.75440049,
      "num_input_tokens_seen": 42111150,
      "step": 1974,
      "time_per_iteration": 2.5685746669769287
    },
    {
      "auxiliary_loss_clip": 0.01086809,
      "auxiliary_loss_mlp": 0.01002918,
      "balance_loss_clip": 1.02607262,
      "balance_loss_mlp": 1.00077176,
      "epoch": 0.23747970901220466,
      "flos": 70546944821760.0,
      "grad_norm": 0.8385790091261014,
      "language_loss": 0.5827831,
      "learning_rate": 3.5653267618302845e-06,
      "loss": 0.60368037,
      "num_input_tokens_seen": 42178730,
      "step": 1975,
      "time_per_iteration": 3.144622802734375
    },
    {
      "auxiliary_loss_clip": 0.01206732,
      "auxiliary_loss_mlp": 0.01032464,
      "balance_loss_clip": 1.06019032,
      "balance_loss_mlp": 1.02249217,
      "epoch": 0.23759995190284375,
      "flos": 20849807594880.0,
      "grad_norm": 1.6352899386936546,
      "language_loss": 0.85499787,
      "learning_rate": 3.564841775947093e-06,
      "loss": 0.87738985,
      "num_input_tokens_seen": 42199620,
      "step": 1976,
      "time_per_iteration": 2.506077289581299
    },
    {
      "auxiliary_loss_clip": 0.01162125,
      "auxiliary_loss_mlp": 0.0103304,
      "balance_loss_clip": 1.05427802,
      "balance_loss_mlp": 1.02406394,
      "epoch": 0.23772019479348283,
      "flos": 32921645420160.0,
      "grad_norm": 2.3654227863844324,
      "language_loss": 0.7636494,
      "learning_rate": 3.5643565526786475e-06,
      "loss": 0.78560114,
      "num_input_tokens_seen": 42219560,
      "step": 1977,
      "time_per_iteration": 2.671215295791626
    },
    {
      "auxiliary_loss_clip": 0.01210421,
      "auxiliary_loss_mlp": 0.0103438,
      "balance_loss_clip": 1.06404459,
      "balance_loss_mlp": 1.02515888,
      "epoch": 0.2378404376841219,
      "flos": 32342765834880.0,
      "grad_norm": 1.6482069416505059,
      "language_loss": 0.7702046,
      "learning_rate": 3.5638710920985574e-06,
      "loss": 0.79265261,
      "num_input_tokens_seen": 42241020,
      "step": 1978,
      "time_per_iteration": 2.5967066287994385
    },
    {
      "auxiliary_loss_clip": 0.01198165,
      "auxiliary_loss_mlp": 0.00762793,
      "balance_loss_clip": 1.05887723,
      "balance_loss_mlp": 1.00031805,
      "epoch": 0.23796068057476102,
      "flos": 22997624313600.0,
      "grad_norm": 2.640323676287713,
      "language_loss": 0.81991374,
      "learning_rate": 3.5633853942804655e-06,
      "loss": 0.83952338,
      "num_input_tokens_seen": 42259345,
      "step": 1979,
      "time_per_iteration": 2.5839288234710693
    },
    {
      "auxiliary_loss_clip": 0.01159776,
      "auxiliary_loss_mlp": 0.01038652,
      "balance_loss_clip": 1.05117679,
      "balance_loss_mlp": 1.02843022,
      "epoch": 0.2380809234654001,
      "flos": 13480938414720.0,
      "grad_norm": 2.015254671773996,
      "language_loss": 0.76298261,
      "learning_rate": 3.5628994592980527e-06,
      "loss": 0.78496689,
      "num_input_tokens_seen": 42277250,
      "step": 1980,
      "time_per_iteration": 2.5712084770202637
    },
    {
      "auxiliary_loss_clip": 0.01205871,
      "auxiliary_loss_mlp": 0.01032776,
      "balance_loss_clip": 1.059834,
      "balance_loss_mlp": 1.02372253,
      "epoch": 0.2382011663560392,
      "flos": 16871803148160.0,
      "grad_norm": 1.6412887049005456,
      "language_loss": 0.7041977,
      "learning_rate": 3.562413287225034e-06,
      "loss": 0.7265842,
      "num_input_tokens_seen": 42295360,
      "step": 1981,
      "time_per_iteration": 2.4527695178985596
    },
    {
      "auxiliary_loss_clip": 0.01189275,
      "auxiliary_loss_mlp": 0.01034026,
      "balance_loss_clip": 1.05976534,
      "balance_loss_mlp": 1.02482259,
      "epoch": 0.2383214092466783,
      "flos": 18441135331200.0,
      "grad_norm": 2.1190734102078292,
      "language_loss": 0.89265573,
      "learning_rate": 3.5619268781351623e-06,
      "loss": 0.91488874,
      "num_input_tokens_seen": 42313430,
      "step": 1982,
      "time_per_iteration": 2.487014055252075
    },
    {
      "auxiliary_loss_clip": 0.0117653,
      "auxiliary_loss_mlp": 0.0103638,
      "balance_loss_clip": 1.05964136,
      "balance_loss_mlp": 1.02799964,
      "epoch": 0.23844165213731738,
      "flos": 19755717281280.0,
      "grad_norm": 1.7595274969323675,
      "language_loss": 0.77049565,
      "learning_rate": 3.5614402321022256e-06,
      "loss": 0.79262477,
      "num_input_tokens_seen": 42331260,
      "step": 1983,
      "time_per_iteration": 2.530602216720581
    },
    {
      "auxiliary_loss_clip": 0.0114047,
      "auxiliary_loss_mlp": 0.01042229,
      "balance_loss_clip": 1.05213284,
      "balance_loss_mlp": 1.03290057,
      "epoch": 0.23856189502795647,
      "flos": 23367360960000.0,
      "grad_norm": 1.9057146581507358,
      "language_loss": 0.87410343,
      "learning_rate": 3.5609533492000463e-06,
      "loss": 0.89593041,
      "num_input_tokens_seen": 42350150,
      "step": 1984,
      "time_per_iteration": 2.6270885467529297
    },
    {
      "auxiliary_loss_clip": 0.01176811,
      "auxiliary_loss_mlp": 0.01034568,
      "balance_loss_clip": 1.06277537,
      "balance_loss_mlp": 1.02539444,
      "epoch": 0.23868213791859555,
      "flos": 23475056912640.0,
      "grad_norm": 2.046432777296419,
      "language_loss": 0.78512508,
      "learning_rate": 3.560466229502485e-06,
      "loss": 0.80723882,
      "num_input_tokens_seen": 42369495,
      "step": 1985,
      "time_per_iteration": 2.5581717491149902
    },
    {
      "auxiliary_loss_clip": 0.01174557,
      "auxiliary_loss_mlp": 0.00762841,
      "balance_loss_clip": 1.06091213,
      "balance_loss_mlp": 1.00029492,
      "epoch": 0.23880238080923466,
      "flos": 16617340224000.0,
      "grad_norm": 3.3041719414190522,
      "language_loss": 0.89842248,
      "learning_rate": 3.5599788730834384e-06,
      "loss": 0.91779643,
      "num_input_tokens_seen": 42387455,
      "step": 1986,
      "time_per_iteration": 2.500875473022461
    },
    {
      "auxiliary_loss_clip": 0.0119611,
      "auxiliary_loss_mlp": 0.01036622,
      "balance_loss_clip": 1.05950332,
      "balance_loss_mlp": 1.02670944,
      "epoch": 0.23892262369987374,
      "flos": 17348409734400.0,
      "grad_norm": 2.055682914630254,
      "language_loss": 0.78562576,
      "learning_rate": 3.559491280016836e-06,
      "loss": 0.80795312,
      "num_input_tokens_seen": 42405400,
      "step": 1987,
      "time_per_iteration": 2.488459587097168
    },
    {
      "auxiliary_loss_clip": 0.01177388,
      "auxiliary_loss_mlp": 0.01038194,
      "balance_loss_clip": 1.0567838,
      "balance_loss_mlp": 1.02890146,
      "epoch": 0.23904286659051283,
      "flos": 22309899540480.0,
      "grad_norm": 1.7889473250699526,
      "language_loss": 0.71063244,
      "learning_rate": 3.5590034503766465e-06,
      "loss": 0.73278826,
      "num_input_tokens_seen": 42425065,
      "step": 1988,
      "time_per_iteration": 2.5506553649902344
    },
    {
      "auxiliary_loss_clip": 0.01207892,
      "auxiliary_loss_mlp": 0.01041333,
      "balance_loss_clip": 1.06358266,
      "balance_loss_mlp": 1.03207088,
      "epoch": 0.23916310948115194,
      "flos": 21178246579200.0,
      "grad_norm": 2.284243880907437,
      "language_loss": 0.8094033,
      "learning_rate": 3.558515384236874e-06,
      "loss": 0.83189559,
      "num_input_tokens_seen": 42442495,
      "step": 1989,
      "time_per_iteration": 3.2694685459136963
    },
    {
      "auxiliary_loss_clip": 0.01151626,
      "auxiliary_loss_mlp": 0.00763276,
      "balance_loss_clip": 1.05509472,
      "balance_loss_mlp": 1.00031459,
      "epoch": 0.23928335237179102,
      "flos": 14137349506560.0,
      "grad_norm": 1.8378703517683261,
      "language_loss": 0.8371309,
      "learning_rate": 3.558027081671556e-06,
      "loss": 0.85627991,
      "num_input_tokens_seen": 42459480,
      "step": 1990,
      "time_per_iteration": 2.533486843109131
    },
    {
      "auxiliary_loss_clip": 0.01194025,
      "auxiliary_loss_mlp": 0.01031191,
      "balance_loss_clip": 1.05664587,
      "balance_loss_mlp": 1.02153504,
      "epoch": 0.2394035952624301,
      "flos": 23769596436480.0,
      "grad_norm": 1.8477700213275692,
      "language_loss": 0.68970048,
      "learning_rate": 3.557538542754769e-06,
      "loss": 0.71195263,
      "num_input_tokens_seen": 42479175,
      "step": 1991,
      "time_per_iteration": 2.5401432514190674
    },
    {
      "auxiliary_loss_clip": 0.0120799,
      "auxiliary_loss_mlp": 0.01034698,
      "balance_loss_clip": 1.06237936,
      "balance_loss_mlp": 1.02566147,
      "epoch": 0.2395238381530692,
      "flos": 24206198250240.0,
      "grad_norm": 1.7935020251532408,
      "language_loss": 0.66712642,
      "learning_rate": 3.557049767560623e-06,
      "loss": 0.68955326,
      "num_input_tokens_seen": 42498090,
      "step": 1992,
      "time_per_iteration": 2.4828922748565674
    },
    {
      "auxiliary_loss_clip": 0.0115148,
      "auxiliary_loss_mlp": 0.01030329,
      "balance_loss_clip": 1.05629671,
      "balance_loss_mlp": 1.02146566,
      "epoch": 0.2396440810437083,
      "flos": 25295763450240.0,
      "grad_norm": 2.450885374296754,
      "language_loss": 0.85910034,
      "learning_rate": 3.5565607561632655e-06,
      "loss": 0.88091838,
      "num_input_tokens_seen": 42516930,
      "step": 1993,
      "time_per_iteration": 2.6355857849121094
    },
    {
      "auxiliary_loss_clip": 0.01174972,
      "auxiliary_loss_mlp": 0.01030755,
      "balance_loss_clip": 1.05711079,
      "balance_loss_mlp": 1.02110481,
      "epoch": 0.23976432393434738,
      "flos": 28543093436160.0,
      "grad_norm": 2.7263925761049914,
      "language_loss": 0.79758215,
      "learning_rate": 3.5560715086368787e-06,
      "loss": 0.81963938,
      "num_input_tokens_seen": 42534800,
      "step": 1994,
      "time_per_iteration": 3.3718197345733643
    },
    {
      "auxiliary_loss_clip": 0.01173836,
      "auxiliary_loss_mlp": 0.01037816,
      "balance_loss_clip": 1.0590694,
      "balance_loss_mlp": 1.02864254,
      "epoch": 0.23988456682498646,
      "flos": 19494358945920.0,
      "grad_norm": 1.9421676909343608,
      "language_loss": 0.8211785,
      "learning_rate": 3.5555820250556816e-06,
      "loss": 0.84329504,
      "num_input_tokens_seen": 42552000,
      "step": 1995,
      "time_per_iteration": 2.5237960815429688
    },
    {
      "auxiliary_loss_clip": 0.01178616,
      "auxiliary_loss_mlp": 0.01039295,
      "balance_loss_clip": 1.05872345,
      "balance_loss_mlp": 1.02992523,
      "epoch": 0.24000480971562557,
      "flos": 20266331068800.0,
      "grad_norm": 2.1761493654806574,
      "language_loss": 0.69464111,
      "learning_rate": 3.5550923054939278e-06,
      "loss": 0.71682024,
      "num_input_tokens_seen": 42571455,
      "step": 1996,
      "time_per_iteration": 3.2587521076202393
    },
    {
      "auxiliary_loss_clip": 0.01136948,
      "auxiliary_loss_mlp": 0.01038292,
      "balance_loss_clip": 1.04900229,
      "balance_loss_mlp": 1.02932119,
      "epoch": 0.24012505260626466,
      "flos": 25443176866560.0,
      "grad_norm": 1.7486625121183688,
      "language_loss": 0.74281919,
      "learning_rate": 3.5546023500259083e-06,
      "loss": 0.76457155,
      "num_input_tokens_seen": 42592550,
      "step": 1997,
      "time_per_iteration": 2.6277942657470703
    },
    {
      "auxiliary_loss_clip": 0.01150816,
      "auxiliary_loss_mlp": 0.01037663,
      "balance_loss_clip": 1.05213046,
      "balance_loss_mlp": 1.02893686,
      "epoch": 0.24024529549690374,
      "flos": 15553342529280.0,
      "grad_norm": 1.8432603497315647,
      "language_loss": 0.80408478,
      "learning_rate": 3.5541121587259477e-06,
      "loss": 0.82596958,
      "num_input_tokens_seen": 42610385,
      "step": 1998,
      "time_per_iteration": 3.370062828063965
    },
    {
      "auxiliary_loss_clip": 0.01085386,
      "auxiliary_loss_mlp": 0.01003869,
      "balance_loss_clip": 1.02732277,
      "balance_loss_mlp": 1.00159252,
      "epoch": 0.24036553838754285,
      "flos": 57122351867520.0,
      "grad_norm": 0.9051604696045212,
      "language_loss": 0.57934153,
      "learning_rate": 3.553621731668408e-06,
      "loss": 0.60023409,
      "num_input_tokens_seen": 42673595,
      "step": 1999,
      "time_per_iteration": 3.0811831951141357
    },
    {
      "auxiliary_loss_clip": 0.01182294,
      "auxiliary_loss_mlp": 0.01033074,
      "balance_loss_clip": 1.05519915,
      "balance_loss_mlp": 1.02321529,
      "epoch": 0.24048578127818193,
      "flos": 24969946158720.0,
      "grad_norm": 2.115910824762654,
      "language_loss": 0.82974321,
      "learning_rate": 3.553131068927688e-06,
      "loss": 0.85189688,
      "num_input_tokens_seen": 42692000,
      "step": 2000,
      "time_per_iteration": 2.5622522830963135
    },
    {
      "auxiliary_loss_clip": 0.01160126,
      "auxiliary_loss_mlp": 0.01032434,
      "balance_loss_clip": 1.05595148,
      "balance_loss_mlp": 1.02344012,
      "epoch": 0.24060602416882101,
      "flos": 23330947547520.0,
      "grad_norm": 1.591205870386221,
      "language_loss": 0.80264258,
      "learning_rate": 3.552640170578219e-06,
      "loss": 0.82456821,
      "num_input_tokens_seen": 42712250,
      "step": 2001,
      "time_per_iteration": 2.5879533290863037
    },
    {
      "auxiliary_loss_clip": 0.01174166,
      "auxiliary_loss_mlp": 0.01033201,
      "balance_loss_clip": 1.05526066,
      "balance_loss_mlp": 1.02407598,
      "epoch": 0.2407262670594601,
      "flos": 14173260128640.0,
      "grad_norm": 2.3526902839227986,
      "language_loss": 0.78394145,
      "learning_rate": 3.5521490366944703e-06,
      "loss": 0.80601513,
      "num_input_tokens_seen": 42729900,
      "step": 2002,
      "time_per_iteration": 2.5069727897644043
    },
    {
      "auxiliary_loss_clip": 0.01161988,
      "auxiliary_loss_mlp": 0.01030928,
      "balance_loss_clip": 1.05648637,
      "balance_loss_mlp": 1.02221417,
      "epoch": 0.2408465099500992,
      "flos": 13663113217920.0,
      "grad_norm": 2.491432139765458,
      "language_loss": 0.80301785,
      "learning_rate": 3.5516576673509474e-06,
      "loss": 0.824947,
      "num_input_tokens_seen": 42747900,
      "step": 2003,
      "time_per_iteration": 2.532531976699829
    },
    {
      "auxiliary_loss_clip": 0.01207744,
      "auxiliary_loss_mlp": 0.0103969,
      "balance_loss_clip": 1.06081462,
      "balance_loss_mlp": 1.03015327,
      "epoch": 0.2409667528407383,
      "flos": 31248029076480.0,
      "grad_norm": 1.6572023190842826,
      "language_loss": 0.8611142,
      "learning_rate": 3.5511660626221896e-06,
      "loss": 0.88358855,
      "num_input_tokens_seen": 42768540,
      "step": 2004,
      "time_per_iteration": 2.546412229537964
    },
    {
      "auxiliary_loss_clip": 0.01177068,
      "auxiliary_loss_mlp": 0.00763545,
      "balance_loss_clip": 1.06065214,
      "balance_loss_mlp": 1.0003792,
      "epoch": 0.24108699573137737,
      "flos": 22199941031040.0,
      "grad_norm": 2.2442541005385785,
      "language_loss": 0.8914994,
      "learning_rate": 3.5506742225827744e-06,
      "loss": 0.91090548,
      "num_input_tokens_seen": 42785395,
      "step": 2005,
      "time_per_iteration": 2.5307891368865967
    },
    {
      "auxiliary_loss_clip": 0.01160792,
      "auxiliary_loss_mlp": 0.01036895,
      "balance_loss_clip": 1.05623555,
      "balance_loss_mlp": 1.02810907,
      "epoch": 0.24120723862201648,
      "flos": 26103035664000.0,
      "grad_norm": 2.284210840467783,
      "language_loss": 0.90229243,
      "learning_rate": 3.5501821473073116e-06,
      "loss": 0.92426932,
      "num_input_tokens_seen": 42801980,
      "step": 2006,
      "time_per_iteration": 2.5934860706329346
    },
    {
      "auxiliary_loss_clip": 0.01156125,
      "auxiliary_loss_mlp": 0.01043443,
      "balance_loss_clip": 1.05615366,
      "balance_loss_mlp": 1.03454995,
      "epoch": 0.24132748151265557,
      "flos": 18624926246400.0,
      "grad_norm": 1.9923388271111364,
      "language_loss": 0.86867446,
      "learning_rate": 3.54968983687045e-06,
      "loss": 0.89067018,
      "num_input_tokens_seen": 42818850,
      "step": 2007,
      "time_per_iteration": 2.5337555408477783
    },
    {
      "auxiliary_loss_clip": 0.0117583,
      "auxiliary_loss_mlp": 0.01040906,
      "balance_loss_clip": 1.0557282,
      "balance_loss_mlp": 1.03131557,
      "epoch": 0.24144772440329465,
      "flos": 15267673664640.0,
      "grad_norm": 2.6944936364528567,
      "language_loss": 0.89282227,
      "learning_rate": 3.549197291346872e-06,
      "loss": 0.91498965,
      "num_input_tokens_seen": 42835375,
      "step": 2008,
      "time_per_iteration": 2.50006365776062
    },
    {
      "auxiliary_loss_clip": 0.01191005,
      "auxiliary_loss_mlp": 0.01036046,
      "balance_loss_clip": 1.05666161,
      "balance_loss_mlp": 1.02691436,
      "epoch": 0.24156796729393373,
      "flos": 24024274842240.0,
      "grad_norm": 2.0266248379685807,
      "language_loss": 0.79381126,
      "learning_rate": 3.548704510811297e-06,
      "loss": 0.81608176,
      "num_input_tokens_seen": 42854570,
      "step": 2009,
      "time_per_iteration": 2.5244953632354736
    },
    {
      "auxiliary_loss_clip": 0.01152169,
      "auxiliary_loss_mlp": 0.0103599,
      "balance_loss_clip": 1.05248916,
      "balance_loss_mlp": 1.02698994,
      "epoch": 0.24168821018457284,
      "flos": 26286790665600.0,
      "grad_norm": 2.4467328807957167,
      "language_loss": 0.74703932,
      "learning_rate": 3.5482114953384787e-06,
      "loss": 0.7689209,
      "num_input_tokens_seen": 42873800,
      "step": 2010,
      "time_per_iteration": 2.631364583969116
    },
    {
      "auxiliary_loss_clip": 0.01194041,
      "auxiliary_loss_mlp": 0.01035799,
      "balance_loss_clip": 1.05972624,
      "balance_loss_mlp": 1.02607703,
      "epoch": 0.24180845307521193,
      "flos": 18223193560320.0,
      "grad_norm": 2.093571568753876,
      "language_loss": 0.84556746,
      "learning_rate": 3.5477182450032077e-06,
      "loss": 0.8678658,
      "num_input_tokens_seen": 42892400,
      "step": 2011,
      "time_per_iteration": 2.475832939147949
    },
    {
      "auxiliary_loss_clip": 0.01191702,
      "auxiliary_loss_mlp": 0.01036528,
      "balance_loss_clip": 1.05906332,
      "balance_loss_mlp": 1.02649117,
      "epoch": 0.241928695965851,
      "flos": 20449260057600.0,
      "grad_norm": 4.263210120591379,
      "language_loss": 0.83227575,
      "learning_rate": 3.5472247598803097e-06,
      "loss": 0.85455811,
      "num_input_tokens_seen": 42911745,
      "step": 2012,
      "time_per_iteration": 2.503694534301758
    },
    {
      "auxiliary_loss_clip": 0.01205802,
      "auxiliary_loss_mlp": 0.01033374,
      "balance_loss_clip": 1.05881751,
      "balance_loss_mlp": 1.02371764,
      "epoch": 0.24204893885649012,
      "flos": 25556475340800.0,
      "grad_norm": 2.3139497750952147,
      "language_loss": 0.85301667,
      "learning_rate": 3.546731040044645e-06,
      "loss": 0.87540841,
      "num_input_tokens_seen": 42926915,
      "step": 2013,
      "time_per_iteration": 2.5016987323760986
    },
    {
      "auxiliary_loss_clip": 0.01205498,
      "auxiliary_loss_mlp": 0.01035617,
      "balance_loss_clip": 1.06111646,
      "balance_loss_mlp": 1.02622962,
      "epoch": 0.2421691817471292,
      "flos": 30660207004800.0,
      "grad_norm": 1.7877170456863605,
      "language_loss": 0.74725783,
      "learning_rate": 3.546237085571112e-06,
      "loss": 0.76966894,
      "num_input_tokens_seen": 42945350,
      "step": 2014,
      "time_per_iteration": 3.248819589614868
    },
    {
      "auxiliary_loss_clip": 0.01194043,
      "auxiliary_loss_mlp": 0.01031299,
      "balance_loss_clip": 1.060992,
      "balance_loss_mlp": 1.02166653,
      "epoch": 0.24228942463776829,
      "flos": 21945011230080.0,
      "grad_norm": 2.0165849150602617,
      "language_loss": 0.7227695,
      "learning_rate": 3.5457428965346425e-06,
      "loss": 0.74502295,
      "num_input_tokens_seen": 42964290,
      "step": 2015,
      "time_per_iteration": 2.5165798664093018
    },
    {
      "auxiliary_loss_clip": 0.01134629,
      "auxiliary_loss_mlp": 0.01031491,
      "balance_loss_clip": 1.05250359,
      "balance_loss_mlp": 1.02195454,
      "epoch": 0.2424096675284074,
      "flos": 33984493879680.0,
      "grad_norm": 2.99718603739699,
      "language_loss": 0.75075448,
      "learning_rate": 3.545248473010205e-06,
      "loss": 0.77241564,
      "num_input_tokens_seen": 42987095,
      "step": 2016,
      "time_per_iteration": 2.7432687282562256
    },
    {
      "auxiliary_loss_clip": 0.01208451,
      "auxiliary_loss_mlp": 0.00763762,
      "balance_loss_clip": 1.06087685,
      "balance_loss_mlp": 1.00045896,
      "epoch": 0.24252991041904648,
      "flos": 21653416621440.0,
      "grad_norm": 2.2962792903012987,
      "language_loss": 0.8779695,
      "learning_rate": 3.544753815072802e-06,
      "loss": 0.89769161,
      "num_input_tokens_seen": 43005750,
      "step": 2017,
      "time_per_iteration": 2.511944532394409
    },
    {
      "auxiliary_loss_clip": 0.01111044,
      "auxiliary_loss_mlp": 0.01032311,
      "balance_loss_clip": 1.04531264,
      "balance_loss_mlp": 1.02258956,
      "epoch": 0.24265015330968556,
      "flos": 21870065502720.0,
      "grad_norm": 1.8175690353899094,
      "language_loss": 0.88453114,
      "learning_rate": 3.544258922797474e-06,
      "loss": 0.90596473,
      "num_input_tokens_seen": 43023870,
      "step": 2018,
      "time_per_iteration": 2.6642651557922363
    },
    {
      "auxiliary_loss_clip": 0.01205042,
      "auxiliary_loss_mlp": 0.01039417,
      "balance_loss_clip": 1.06026077,
      "balance_loss_mlp": 1.03032136,
      "epoch": 0.24277039620032465,
      "flos": 25628260671360.0,
      "grad_norm": 1.5714857818439214,
      "language_loss": 0.78236067,
      "learning_rate": 3.543763796259295e-06,
      "loss": 0.80480522,
      "num_input_tokens_seen": 43043825,
      "step": 2019,
      "time_per_iteration": 2.5179452896118164
    },
    {
      "auxiliary_loss_clip": 0.01190779,
      "auxiliary_loss_mlp": 0.01035655,
      "balance_loss_clip": 1.05706573,
      "balance_loss_mlp": 1.02559364,
      "epoch": 0.24289063909096376,
      "flos": 26286575184000.0,
      "grad_norm": 1.849859695734387,
      "language_loss": 0.91005909,
      "learning_rate": 3.5432684355333754e-06,
      "loss": 0.9323234,
      "num_input_tokens_seen": 43062480,
      "step": 2020,
      "time_per_iteration": 3.3127622604370117
    },
    {
      "auxiliary_loss_clip": 0.0119032,
      "auxiliary_loss_mlp": 0.01033683,
      "balance_loss_clip": 1.05648541,
      "balance_loss_mlp": 1.02508187,
      "epoch": 0.24301088198160284,
      "flos": 25075056332160.0,
      "grad_norm": 1.9701197398192898,
      "language_loss": 0.7641958,
      "learning_rate": 3.5427728406948613e-06,
      "loss": 0.78643584,
      "num_input_tokens_seen": 43081595,
      "step": 2021,
      "time_per_iteration": 2.5378661155700684
    },
    {
      "auxiliary_loss_clip": 0.01078473,
      "auxiliary_loss_mlp": 0.01003422,
      "balance_loss_clip": 1.02431774,
      "balance_loss_mlp": 1.00139511,
      "epoch": 0.24313112487224192,
      "flos": 69900948673920.0,
      "grad_norm": 0.7626906419244385,
      "language_loss": 0.57938707,
      "learning_rate": 3.542277011818934e-06,
      "loss": 0.60020602,
      "num_input_tokens_seen": 43145430,
      "step": 2022,
      "time_per_iteration": 3.9090769290924072
    },
    {
      "auxiliary_loss_clip": 0.0118045,
      "auxiliary_loss_mlp": 0.01034977,
      "balance_loss_clip": 1.06094098,
      "balance_loss_mlp": 1.02604818,
      "epoch": 0.24325136776288103,
      "flos": 40662334235520.0,
      "grad_norm": 2.0957282987055907,
      "language_loss": 0.74553931,
      "learning_rate": 3.5417809489808104e-06,
      "loss": 0.76769358,
      "num_input_tokens_seen": 43167040,
      "step": 2023,
      "time_per_iteration": 2.7028770446777344
    },
    {
      "auxiliary_loss_clip": 0.01195641,
      "auxiliary_loss_mlp": 0.01036561,
      "balance_loss_clip": 1.06112671,
      "balance_loss_mlp": 1.02827024,
      "epoch": 0.24337161065352012,
      "flos": 25046400257280.0,
      "grad_norm": 1.787530232634607,
      "language_loss": 0.72663289,
      "learning_rate": 3.5412846522557422e-06,
      "loss": 0.74895489,
      "num_input_tokens_seen": 43187930,
      "step": 2024,
      "time_per_iteration": 3.3037662506103516
    },
    {
      "auxiliary_loss_clip": 0.01205974,
      "auxiliary_loss_mlp": 0.01035699,
      "balance_loss_clip": 1.05987763,
      "balance_loss_mlp": 1.02572715,
      "epoch": 0.2434918535441592,
      "flos": 18661160090880.0,
      "grad_norm": 2.021311320657345,
      "language_loss": 0.7433567,
      "learning_rate": 3.540788121719018e-06,
      "loss": 0.76577342,
      "num_input_tokens_seen": 43206350,
      "step": 2025,
      "time_per_iteration": 2.450728178024292
    },
    {
      "auxiliary_loss_clip": 0.01152816,
      "auxiliary_loss_mlp": 0.01031565,
      "balance_loss_clip": 1.0572741,
      "balance_loss_mlp": 1.0219568,
      "epoch": 0.24361209643479828,
      "flos": 23915142345600.0,
      "grad_norm": 2.35238184545431,
      "language_loss": 0.81752419,
      "learning_rate": 3.5402913574459604e-06,
      "loss": 0.83936799,
      "num_input_tokens_seen": 43226255,
      "step": 2026,
      "time_per_iteration": 2.5941474437713623
    },
    {
      "auxiliary_loss_clip": 0.01127306,
      "auxiliary_loss_mlp": 0.01034765,
      "balance_loss_clip": 1.04753542,
      "balance_loss_mlp": 1.02615201,
      "epoch": 0.2437323393254374,
      "flos": 28657505232000.0,
      "grad_norm": 1.652752112566464,
      "language_loss": 0.86068058,
      "learning_rate": 3.5397943595119297e-06,
      "loss": 0.88230133,
      "num_input_tokens_seen": 43247675,
      "step": 2027,
      "time_per_iteration": 2.690549373626709
    },
    {
      "auxiliary_loss_clip": 0.01175971,
      "auxiliary_loss_mlp": 0.01035581,
      "balance_loss_clip": 1.05985713,
      "balance_loss_mlp": 1.02618742,
      "epoch": 0.24385258221607647,
      "flos": 23550325862400.0,
      "grad_norm": 2.25750268638344,
      "language_loss": 0.77470267,
      "learning_rate": 3.5392971279923177e-06,
      "loss": 0.79681814,
      "num_input_tokens_seen": 43265895,
      "step": 2028,
      "time_per_iteration": 2.554649591445923
    },
    {
      "auxiliary_loss_clip": 0.0115731,
      "auxiliary_loss_mlp": 0.01035961,
      "balance_loss_clip": 1.05277944,
      "balance_loss_mlp": 1.02675223,
      "epoch": 0.24397282510671556,
      "flos": 25336091445120.0,
      "grad_norm": 1.968471344433543,
      "language_loss": 0.82893121,
      "learning_rate": 3.5387996629625557e-06,
      "loss": 0.85086393,
      "num_input_tokens_seen": 43283485,
      "step": 2029,
      "time_per_iteration": 2.5910468101501465
    },
    {
      "auxiliary_loss_clip": 0.0109859,
      "auxiliary_loss_mlp": 0.01002684,
      "balance_loss_clip": 1.02401328,
      "balance_loss_mlp": 1.00078857,
      "epoch": 0.24409306799735467,
      "flos": 65187421430400.0,
      "grad_norm": 0.7970891012011679,
      "language_loss": 0.55021167,
      "learning_rate": 3.5383019644981083e-06,
      "loss": 0.57122439,
      "num_input_tokens_seen": 43347180,
      "step": 2030,
      "time_per_iteration": 3.0983214378356934
    },
    {
      "auxiliary_loss_clip": 0.01175172,
      "auxiliary_loss_mlp": 0.01044099,
      "balance_loss_clip": 1.05758572,
      "balance_loss_mlp": 1.03400171,
      "epoch": 0.24421331088799375,
      "flos": 19537093152000.0,
      "grad_norm": 2.219549242306186,
      "language_loss": 0.73066509,
      "learning_rate": 3.5378040326744763e-06,
      "loss": 0.7528578,
      "num_input_tokens_seen": 43366665,
      "step": 2031,
      "time_per_iteration": 2.5473358631134033
    },
    {
      "auxiliary_loss_clip": 0.01162869,
      "auxiliary_loss_mlp": 0.01046395,
      "balance_loss_clip": 1.05821848,
      "balance_loss_mlp": 1.03686976,
      "epoch": 0.24433355377863283,
      "flos": 21068575378560.0,
      "grad_norm": 2.021085035112213,
      "language_loss": 0.85706258,
      "learning_rate": 3.5373058675671946e-06,
      "loss": 0.87915528,
      "num_input_tokens_seen": 43384670,
      "step": 2032,
      "time_per_iteration": 2.5618579387664795
    },
    {
      "auxiliary_loss_clip": 0.01140193,
      "auxiliary_loss_mlp": 0.01035821,
      "balance_loss_clip": 1.05263758,
      "balance_loss_mlp": 1.02587271,
      "epoch": 0.24445379666927192,
      "flos": 22637189289600.0,
      "grad_norm": 2.288698870043243,
      "language_loss": 0.72526121,
      "learning_rate": 3.536807469251836e-06,
      "loss": 0.74702138,
      "num_input_tokens_seen": 43403825,
      "step": 2033,
      "time_per_iteration": 2.6115474700927734
    },
    {
      "auxiliary_loss_clip": 0.01165045,
      "auxiliary_loss_mlp": 0.01025909,
      "balance_loss_clip": 1.05133176,
      "balance_loss_mlp": 1.01663458,
      "epoch": 0.24457403955991103,
      "flos": 21251612108160.0,
      "grad_norm": 1.8276958591021424,
      "language_loss": 0.82867777,
      "learning_rate": 3.5363088378040055e-06,
      "loss": 0.85058737,
      "num_input_tokens_seen": 43422715,
      "step": 2034,
      "time_per_iteration": 2.5768396854400635
    },
    {
      "auxiliary_loss_clip": 0.01098434,
      "auxiliary_loss_mlp": 0.00752922,
      "balance_loss_clip": 1.02418447,
      "balance_loss_mlp": 1.00004292,
      "epoch": 0.2446942824505501,
      "flos": 66997820764800.0,
      "grad_norm": 0.7489855076037047,
      "language_loss": 0.64356792,
      "learning_rate": 3.5358099732993463e-06,
      "loss": 0.66208148,
      "num_input_tokens_seen": 43481825,
      "step": 2035,
      "time_per_iteration": 3.021388292312622
    },
    {
      "auxiliary_loss_clip": 0.01182785,
      "auxiliary_loss_mlp": 0.01035747,
      "balance_loss_clip": 1.05862236,
      "balance_loss_mlp": 1.02690768,
      "epoch": 0.2448145253411892,
      "flos": 20411122792320.0,
      "grad_norm": 1.7820900520839353,
      "language_loss": 0.89266747,
      "learning_rate": 3.535310875813535e-06,
      "loss": 0.9148528,
      "num_input_tokens_seen": 43500220,
      "step": 2036,
      "time_per_iteration": 2.5306057929992676
    },
    {
      "auxiliary_loss_clip": 0.01187181,
      "auxiliary_loss_mlp": 0.01033229,
      "balance_loss_clip": 1.05693364,
      "balance_loss_mlp": 1.0250572,
      "epoch": 0.2449347682318283,
      "flos": 28804739080320.0,
      "grad_norm": 1.678557160204201,
      "language_loss": 0.81141776,
      "learning_rate": 3.5348115454222843e-06,
      "loss": 0.83362186,
      "num_input_tokens_seen": 43522805,
      "step": 2037,
      "time_per_iteration": 2.5718765258789062
    },
    {
      "auxiliary_loss_clip": 0.01169307,
      "auxiliary_loss_mlp": 0.01036971,
      "balance_loss_clip": 1.05222261,
      "balance_loss_mlp": 1.02763057,
      "epoch": 0.2450550111224674,
      "flos": 22528990546560.0,
      "grad_norm": 2.0142434743497284,
      "language_loss": 0.86593926,
      "learning_rate": 3.5343119822013425e-06,
      "loss": 0.88800204,
      "num_input_tokens_seen": 43541915,
      "step": 2038,
      "time_per_iteration": 2.5527615547180176
    },
    {
      "auxiliary_loss_clip": 0.01196606,
      "auxiliary_loss_mlp": 0.01032423,
      "balance_loss_clip": 1.05854142,
      "balance_loss_mlp": 1.02242136,
      "epoch": 0.24517525401310647,
      "flos": 21759137326080.0,
      "grad_norm": 1.940488092489145,
      "language_loss": 0.77749395,
      "learning_rate": 3.533812186226493e-06,
      "loss": 0.79978424,
      "num_input_tokens_seen": 43562625,
      "step": 2039,
      "time_per_iteration": 2.518183469772339
    },
    {
      "auxiliary_loss_clip": 0.01199948,
      "auxiliary_loss_mlp": 0.01028216,
      "balance_loss_clip": 1.05753016,
      "balance_loss_mlp": 1.02033639,
      "epoch": 0.24529549690374555,
      "flos": 25043311687680.0,
      "grad_norm": 1.7055034204825394,
      "language_loss": 0.75839496,
      "learning_rate": 3.5333121575735545e-06,
      "loss": 0.78067654,
      "num_input_tokens_seen": 43582265,
      "step": 2040,
      "time_per_iteration": 3.232445478439331
    },
    {
      "auxiliary_loss_clip": 0.01179193,
      "auxiliary_loss_mlp": 0.0103759,
      "balance_loss_clip": 1.05984545,
      "balance_loss_mlp": 1.02894139,
      "epoch": 0.24541573979438466,
      "flos": 32123638915200.0,
      "grad_norm": 1.7834670793592902,
      "language_loss": 0.75444269,
      "learning_rate": 3.532811896318381e-06,
      "loss": 0.77661055,
      "num_input_tokens_seen": 43604335,
      "step": 2041,
      "time_per_iteration": 2.619645357131958
    },
    {
      "auxiliary_loss_clip": 0.01162368,
      "auxiliary_loss_mlp": 0.01041288,
      "balance_loss_clip": 1.05537295,
      "balance_loss_mlp": 1.031847,
      "epoch": 0.24553598268502375,
      "flos": 31357556622720.0,
      "grad_norm": 2.091033297684415,
      "language_loss": 0.8204518,
      "learning_rate": 3.5323114025368615e-06,
      "loss": 0.84248829,
      "num_input_tokens_seen": 43619400,
      "step": 2042,
      "time_per_iteration": 2.6189913749694824
    },
    {
      "auxiliary_loss_clip": 0.01184145,
      "auxiliary_loss_mlp": 0.0103161,
      "balance_loss_clip": 1.0530808,
      "balance_loss_mlp": 1.02248442,
      "epoch": 0.24565622557566283,
      "flos": 14027462824320.0,
      "grad_norm": 2.4633406898713006,
      "language_loss": 0.81721365,
      "learning_rate": 3.53181067630492e-06,
      "loss": 0.8393712,
      "num_input_tokens_seen": 43636870,
      "step": 2043,
      "time_per_iteration": 2.473313093185425
    },
    {
      "auxiliary_loss_clip": 0.0116864,
      "auxiliary_loss_mlp": 0.01035107,
      "balance_loss_clip": 1.05474043,
      "balance_loss_mlp": 1.02580309,
      "epoch": 0.24577646846630194,
      "flos": 16581465515520.0,
      "grad_norm": 2.0079178219909957,
      "language_loss": 0.7653147,
      "learning_rate": 3.5313097176985175e-06,
      "loss": 0.7873522,
      "num_input_tokens_seen": 43655180,
      "step": 2044,
      "time_per_iteration": 2.506767511367798
    },
    {
      "auxiliary_loss_clip": 0.01190721,
      "auxiliary_loss_mlp": 0.01034375,
      "balance_loss_clip": 1.05976248,
      "balance_loss_mlp": 1.02517796,
      "epoch": 0.24589671135694102,
      "flos": 18807424272000.0,
      "grad_norm": 1.7935581894155082,
      "language_loss": 0.81077057,
      "learning_rate": 3.5308085267936482e-06,
      "loss": 0.83302152,
      "num_input_tokens_seen": 43672895,
      "step": 2045,
      "time_per_iteration": 2.4861443042755127
    },
    {
      "auxiliary_loss_clip": 0.01137151,
      "auxiliary_loss_mlp": 0.00761856,
      "balance_loss_clip": 1.0561167,
      "balance_loss_mlp": 1.00044453,
      "epoch": 0.2460169542475801,
      "flos": 19938538529280.0,
      "grad_norm": 1.9186084300162136,
      "language_loss": 0.9009074,
      "learning_rate": 3.530307103666342e-06,
      "loss": 0.9198975,
      "num_input_tokens_seen": 43691975,
      "step": 2046,
      "time_per_iteration": 3.4017856121063232
    },
    {
      "auxiliary_loss_clip": 0.01165902,
      "auxiliary_loss_mlp": 0.01034752,
      "balance_loss_clip": 1.0556159,
      "balance_loss_mlp": 1.02494073,
      "epoch": 0.24613719713821922,
      "flos": 24171221381760.0,
      "grad_norm": 2.097803367019977,
      "language_loss": 0.80329895,
      "learning_rate": 3.5298054483926658e-06,
      "loss": 0.82530546,
      "num_input_tokens_seen": 43712670,
      "step": 2047,
      "time_per_iteration": 2.590782403945923
    },
    {
      "auxiliary_loss_clip": 0.01198909,
      "auxiliary_loss_mlp": 0.01039951,
      "balance_loss_clip": 1.06117129,
      "balance_loss_mlp": 1.03038454,
      "epoch": 0.2462574400288583,
      "flos": 30221055325440.0,
      "grad_norm": 1.8533408405161553,
      "language_loss": 0.83150482,
      "learning_rate": 3.5293035610487187e-06,
      "loss": 0.8538934,
      "num_input_tokens_seen": 43732035,
      "step": 2048,
      "time_per_iteration": 3.2985732555389404
    },
    {
      "auxiliary_loss_clip": 0.01070202,
      "auxiliary_loss_mlp": 0.0100323,
      "balance_loss_clip": 1.02213573,
      "balance_loss_mlp": 1.001001,
      "epoch": 0.24637768291949738,
      "flos": 68943030819840.0,
      "grad_norm": 0.7325648743332753,
      "language_loss": 0.62062263,
      "learning_rate": 3.5288014417106374e-06,
      "loss": 0.64135695,
      "num_input_tokens_seen": 43798055,
      "step": 2049,
      "time_per_iteration": 3.162182092666626
    },
    {
      "auxiliary_loss_clip": 0.01161717,
      "auxiliary_loss_mlp": 0.0103565,
      "balance_loss_clip": 1.05762959,
      "balance_loss_mlp": 1.02590477,
      "epoch": 0.24649792581013646,
      "flos": 34383999922560.0,
      "grad_norm": 1.865234087526529,
      "language_loss": 0.75888181,
      "learning_rate": 3.528299090454593e-06,
      "loss": 0.78085554,
      "num_input_tokens_seen": 43818590,
      "step": 2050,
      "time_per_iteration": 3.4445135593414307
    },
    {
      "auxiliary_loss_clip": 0.01191136,
      "auxiliary_loss_mlp": 0.01043227,
      "balance_loss_clip": 1.0559305,
      "balance_loss_mlp": 1.03380966,
      "epoch": 0.24661816870077558,
      "flos": 19680448331520.0,
      "grad_norm": 2.2678396637925458,
      "language_loss": 0.82837576,
      "learning_rate": 3.527796507356792e-06,
      "loss": 0.85071939,
      "num_input_tokens_seen": 43832480,
      "step": 2051,
      "time_per_iteration": 2.500074625015259
    },
    {
      "auxiliary_loss_clip": 0.01193363,
      "auxiliary_loss_mlp": 0.01031498,
      "balance_loss_clip": 1.0587976,
      "balance_loss_mlp": 1.02242625,
      "epoch": 0.24673841159141466,
      "flos": 20002279213440.0,
      "grad_norm": 3.3296230919624876,
      "language_loss": 0.8998152,
      "learning_rate": 3.527293692493475e-06,
      "loss": 0.92206389,
      "num_input_tokens_seen": 43848345,
      "step": 2052,
      "time_per_iteration": 2.4892115592956543
    },
    {
      "auxiliary_loss_clip": 0.01190571,
      "auxiliary_loss_mlp": 0.01031828,
      "balance_loss_clip": 1.05477738,
      "balance_loss_mlp": 1.02172542,
      "epoch": 0.24685865448205374,
      "flos": 21646593037440.0,
      "grad_norm": 2.5222571551502786,
      "language_loss": 0.73614401,
      "learning_rate": 3.52679064594092e-06,
      "loss": 0.75836802,
      "num_input_tokens_seen": 43865685,
      "step": 2053,
      "time_per_iteration": 2.6006484031677246
    },
    {
      "auxiliary_loss_clip": 0.01132797,
      "auxiliary_loss_mlp": 0.0103598,
      "balance_loss_clip": 1.04427433,
      "balance_loss_mlp": 1.02690172,
      "epoch": 0.24697889737269285,
      "flos": 17960470508160.0,
      "grad_norm": 2.2848898466985132,
      "language_loss": 0.74934554,
      "learning_rate": 3.5262873677754375e-06,
      "loss": 0.77103329,
      "num_input_tokens_seen": 43883690,
      "step": 2054,
      "time_per_iteration": 2.5893354415893555
    },
    {
      "auxiliary_loss_clip": 0.0120401,
      "auxiliary_loss_mlp": 0.01042171,
      "balance_loss_clip": 1.05970359,
      "balance_loss_mlp": 1.03309894,
      "epoch": 0.24709914026333193,
      "flos": 27344611221120.0,
      "grad_norm": 1.566083052593149,
      "language_loss": 0.80220222,
      "learning_rate": 3.5257838580733745e-06,
      "loss": 0.82466412,
      "num_input_tokens_seen": 43903295,
      "step": 2055,
      "time_per_iteration": 2.5330142974853516
    },
    {
      "auxiliary_loss_clip": 0.01195508,
      "auxiliary_loss_mlp": 0.0104386,
      "balance_loss_clip": 1.06034207,
      "balance_loss_mlp": 1.03423965,
      "epoch": 0.24721938315397102,
      "flos": 19275519335040.0,
      "grad_norm": 1.8923706671773803,
      "language_loss": 0.86993611,
      "learning_rate": 3.5252801169111138e-06,
      "loss": 0.89232969,
      "num_input_tokens_seen": 43920960,
      "step": 2056,
      "time_per_iteration": 2.4767906665802
    },
    {
      "auxiliary_loss_clip": 0.01174928,
      "auxiliary_loss_mlp": 0.01041866,
      "balance_loss_clip": 1.05937994,
      "balance_loss_mlp": 1.03252041,
      "epoch": 0.2473396260446101,
      "flos": 23185796688000.0,
      "grad_norm": 1.790332618827767,
      "language_loss": 0.8024618,
      "learning_rate": 3.524776144365072e-06,
      "loss": 0.82462978,
      "num_input_tokens_seen": 43939415,
      "step": 2057,
      "time_per_iteration": 2.5512232780456543
    },
    {
      "auxiliary_loss_clip": 0.01175422,
      "auxiliary_loss_mlp": 0.01034264,
      "balance_loss_clip": 1.06181335,
      "balance_loss_mlp": 1.02574062,
      "epoch": 0.2474598689352492,
      "flos": 21142443697920.0,
      "grad_norm": 1.549440954345555,
      "language_loss": 0.79298759,
      "learning_rate": 3.5242719405117016e-06,
      "loss": 0.81508446,
      "num_input_tokens_seen": 43959220,
      "step": 2058,
      "time_per_iteration": 2.5761241912841797
    },
    {
      "auxiliary_loss_clip": 0.0117647,
      "auxiliary_loss_mlp": 0.00762459,
      "balance_loss_clip": 1.05689943,
      "balance_loss_mlp": 1.00044179,
      "epoch": 0.2475801118258883,
      "flos": 21648352803840.0,
      "grad_norm": 5.204197120095358,
      "language_loss": 0.75026548,
      "learning_rate": 3.5237675054274893e-06,
      "loss": 0.76965475,
      "num_input_tokens_seen": 43978420,
      "step": 2059,
      "time_per_iteration": 2.5468244552612305
    },
    {
      "auxiliary_loss_clip": 0.01193114,
      "auxiliary_loss_mlp": 0.01039295,
      "balance_loss_clip": 1.05975783,
      "balance_loss_mlp": 1.02966273,
      "epoch": 0.24770035471652738,
      "flos": 22674500542080.0,
      "grad_norm": 2.0107763475090743,
      "language_loss": 0.80155277,
      "learning_rate": 3.5232628391889584e-06,
      "loss": 0.82387686,
      "num_input_tokens_seen": 43996710,
      "step": 2060,
      "time_per_iteration": 2.509671688079834
    },
    {
      "auxiliary_loss_clip": 0.01146515,
      "auxiliary_loss_mlp": 0.01035557,
      "balance_loss_clip": 1.05579603,
      "balance_loss_mlp": 1.0269202,
      "epoch": 0.2478205976071665,
      "flos": 22163814927360.0,
      "grad_norm": 2.449697804561298,
      "language_loss": 0.64268124,
      "learning_rate": 3.522757941872666e-06,
      "loss": 0.66450202,
      "num_input_tokens_seen": 44014865,
      "step": 2061,
      "time_per_iteration": 2.612285614013672
    },
    {
      "auxiliary_loss_clip": 0.01207944,
      "auxiliary_loss_mlp": 0.00762332,
      "balance_loss_clip": 1.06338608,
      "balance_loss_mlp": 1.00036645,
      "epoch": 0.24794084049780557,
      "flos": 24973106555520.0,
      "grad_norm": 1.5184630795666025,
      "language_loss": 0.82452261,
      "learning_rate": 3.5222528135552042e-06,
      "loss": 0.84422541,
      "num_input_tokens_seen": 44036325,
      "step": 2062,
      "time_per_iteration": 2.5098040103912354
    },
    {
      "auxiliary_loss_clip": 0.01191525,
      "auxiliary_loss_mlp": 0.01036135,
      "balance_loss_clip": 1.06176221,
      "balance_loss_mlp": 1.02699172,
      "epoch": 0.24806108338844465,
      "flos": 18296379521280.0,
      "grad_norm": 2.2323221488857756,
      "language_loss": 0.80416685,
      "learning_rate": 3.521747454313201e-06,
      "loss": 0.82644343,
      "num_input_tokens_seen": 44055005,
      "step": 2063,
      "time_per_iteration": 2.4983997344970703
    },
    {
      "auxiliary_loss_clip": 0.011522,
      "auxiliary_loss_mlp": 0.01032926,
      "balance_loss_clip": 1.0506078,
      "balance_loss_mlp": 1.02335382,
      "epoch": 0.24818132627908374,
      "flos": 19282163351040.0,
      "grad_norm": 1.918198881903483,
      "language_loss": 0.6701591,
      "learning_rate": 3.521241864223319e-06,
      "loss": 0.69201028,
      "num_input_tokens_seen": 44073965,
      "step": 2064,
      "time_per_iteration": 2.563106060028076
    },
    {
      "auxiliary_loss_clip": 0.01074165,
      "auxiliary_loss_mlp": 0.01008474,
      "balance_loss_clip": 1.01900458,
      "balance_loss_mlp": 1.00638759,
      "epoch": 0.24830156916972285,
      "flos": 70285837881600.0,
      "grad_norm": 0.8177070594454853,
      "language_loss": 0.62068152,
      "learning_rate": 3.5207360433622552e-06,
      "loss": 0.64150786,
      "num_input_tokens_seen": 44135965,
      "step": 2065,
      "time_per_iteration": 3.1466333866119385
    },
    {
      "auxiliary_loss_clip": 0.01171376,
      "auxiliary_loss_mlp": 0.01041482,
      "balance_loss_clip": 1.05841732,
      "balance_loss_mlp": 1.03248215,
      "epoch": 0.24842181206036193,
      "flos": 40409128287360.0,
      "grad_norm": 1.5564426910655398,
      "language_loss": 0.74747288,
      "learning_rate": 3.5202299918067437e-06,
      "loss": 0.76960146,
      "num_input_tokens_seen": 44159560,
      "step": 2066,
      "time_per_iteration": 3.5447330474853516
    },
    {
      "auxiliary_loss_clip": 0.01189841,
      "auxiliary_loss_mlp": 0.01034169,
      "balance_loss_clip": 1.06004012,
      "balance_loss_mlp": 1.02502561,
      "epoch": 0.248542054951001,
      "flos": 20082432412800.0,
      "grad_norm": 2.0580186322027294,
      "language_loss": 0.6887306,
      "learning_rate": 3.519723709633551e-06,
      "loss": 0.7109707,
      "num_input_tokens_seen": 44178320,
      "step": 2067,
      "time_per_iteration": 2.506460666656494
    },
    {
      "auxiliary_loss_clip": 0.01170836,
      "auxiliary_loss_mlp": 0.01038494,
      "balance_loss_clip": 1.05912971,
      "balance_loss_mlp": 1.02940392,
      "epoch": 0.24866229784164012,
      "flos": 23513948363520.0,
      "grad_norm": 1.7128055585671824,
      "language_loss": 0.83352184,
      "learning_rate": 3.519217196919479e-06,
      "loss": 0.85561514,
      "num_input_tokens_seen": 44197305,
      "step": 2068,
      "time_per_iteration": 2.567106008529663
    },
    {
      "auxiliary_loss_clip": 0.01181899,
      "auxiliary_loss_mlp": 0.010442,
      "balance_loss_clip": 1.0613538,
      "balance_loss_mlp": 1.03558707,
      "epoch": 0.2487825407322792,
      "flos": 19865101173120.0,
      "grad_norm": 1.7242000759708935,
      "language_loss": 0.73252726,
      "learning_rate": 3.518710453741367e-06,
      "loss": 0.75478828,
      "num_input_tokens_seen": 44216505,
      "step": 2069,
      "time_per_iteration": 2.5544862747192383
    },
    {
      "auxiliary_loss_clip": 0.01168025,
      "auxiliary_loss_mlp": 0.00763066,
      "balance_loss_clip": 1.05652595,
      "balance_loss_mlp": 1.00037408,
      "epoch": 0.2489027836229183,
      "flos": 22017622573440.0,
      "grad_norm": 2.178248446955164,
      "language_loss": 0.67952657,
      "learning_rate": 3.518203480176086e-06,
      "loss": 0.69883752,
      "num_input_tokens_seen": 44235435,
      "step": 2070,
      "time_per_iteration": 2.5554754734039307
    },
    {
      "auxiliary_loss_clip": 0.01117021,
      "auxiliary_loss_mlp": 0.01041812,
      "balance_loss_clip": 1.04755211,
      "balance_loss_mlp": 1.03318107,
      "epoch": 0.2490230265135574,
      "flos": 23294354567040.0,
      "grad_norm": 1.7149304999368582,
      "language_loss": 0.80937308,
      "learning_rate": 3.517696276300545e-06,
      "loss": 0.83096147,
      "num_input_tokens_seen": 44256975,
      "step": 2071,
      "time_per_iteration": 2.7007131576538086
    },
    {
      "auxiliary_loss_clip": 0.01189421,
      "auxiliary_loss_mlp": 0.01038009,
      "balance_loss_clip": 1.06144381,
      "balance_loss_mlp": 1.02859771,
      "epoch": 0.24914326940419648,
      "flos": 19826784339840.0,
      "grad_norm": 2.9071190556951394,
      "language_loss": 0.69728673,
      "learning_rate": 3.517188842191685e-06,
      "loss": 0.71956104,
      "num_input_tokens_seen": 44275125,
      "step": 2072,
      "time_per_iteration": 3.2773730754852295
    },
    {
      "auxiliary_loss_clip": 0.01189534,
      "auxiliary_loss_mlp": 0.0103721,
      "balance_loss_clip": 1.05817795,
      "balance_loss_mlp": 1.02797759,
      "epoch": 0.24926351229483557,
      "flos": 20229271211520.0,
      "grad_norm": 1.6248453422050848,
      "language_loss": 0.73864591,
      "learning_rate": 3.5166811779264837e-06,
      "loss": 0.76091337,
      "num_input_tokens_seen": 44295445,
      "step": 2073,
      "time_per_iteration": 2.5455198287963867
    },
    {
      "auxiliary_loss_clip": 0.01203516,
      "auxiliary_loss_mlp": 0.01037237,
      "balance_loss_clip": 1.05921721,
      "balance_loss_mlp": 1.02807021,
      "epoch": 0.24938375518547465,
      "flos": 23294570048640.0,
      "grad_norm": 1.9100819437822707,
      "language_loss": 0.78185451,
      "learning_rate": 3.5161732835819545e-06,
      "loss": 0.80426204,
      "num_input_tokens_seen": 44314755,
      "step": 2074,
      "time_per_iteration": 3.1958301067352295
    },
    {
      "auxiliary_loss_clip": 0.01204043,
      "auxiliary_loss_mlp": 0.01034546,
      "balance_loss_clip": 1.06138825,
      "balance_loss_mlp": 1.02485442,
      "epoch": 0.24950399807611376,
      "flos": 17311673099520.0,
      "grad_norm": 1.9277585970222786,
      "language_loss": 0.83428413,
      "learning_rate": 3.515665159235143e-06,
      "loss": 0.85667002,
      "num_input_tokens_seen": 44333640,
      "step": 2075,
      "time_per_iteration": 3.2876856327056885
    },
    {
      "auxiliary_loss_clip": 0.01175188,
      "auxiliary_loss_mlp": 0.01036312,
      "balance_loss_clip": 1.05554438,
      "balance_loss_mlp": 1.0281347,
      "epoch": 0.24962424096675284,
      "flos": 19024863252480.0,
      "grad_norm": 2.669701419654636,
      "language_loss": 0.74711627,
      "learning_rate": 3.5151568049631318e-06,
      "loss": 0.7692312,
      "num_input_tokens_seen": 44352355,
      "step": 2076,
      "time_per_iteration": 2.5215775966644287
    },
    {
      "auxiliary_loss_clip": 0.01206449,
      "auxiliary_loss_mlp": 0.01040118,
      "balance_loss_clip": 1.06116295,
      "balance_loss_mlp": 1.03134406,
      "epoch": 0.24974448385739192,
      "flos": 33398790710400.0,
      "grad_norm": 1.7002361584094747,
      "language_loss": 0.80480045,
      "learning_rate": 3.5146482208430385e-06,
      "loss": 0.8272661,
      "num_input_tokens_seen": 44374185,
      "step": 2077,
      "time_per_iteration": 2.5824756622314453
    },
    {
      "auxiliary_loss_clip": 0.01118486,
      "auxiliary_loss_mlp": 0.01035223,
      "balance_loss_clip": 1.04584372,
      "balance_loss_mlp": 1.02538824,
      "epoch": 0.24986472674803104,
      "flos": 30007279532160.0,
      "grad_norm": 2.499260368557853,
      "language_loss": 0.68201607,
      "learning_rate": 3.514139406952014e-06,
      "loss": 0.7035532,
      "num_input_tokens_seen": 44396210,
      "step": 2078,
      "time_per_iteration": 2.6846957206726074
    },
    {
      "auxiliary_loss_clip": 0.01190757,
      "auxiliary_loss_mlp": 0.01030373,
      "balance_loss_clip": 1.06097293,
      "balance_loss_mlp": 1.02126527,
      "epoch": 0.24998496963867012,
      "flos": 26613074833920.0,
      "grad_norm": 1.9040632706540188,
      "language_loss": 0.8373872,
      "learning_rate": 3.5136303633672454e-06,
      "loss": 0.85959852,
      "num_input_tokens_seen": 44416340,
      "step": 2079,
      "time_per_iteration": 2.5518417358398438
    },
    {
      "auxiliary_loss_clip": 0.0116788,
      "auxiliary_loss_mlp": 0.00762148,
      "balance_loss_clip": 1.05626631,
      "balance_loss_mlp": 1.00039148,
      "epoch": 0.25010521252930923,
      "flos": 23553989049600.0,
      "grad_norm": 1.865780153640428,
      "language_loss": 0.7474227,
      "learning_rate": 3.5131210901659544e-06,
      "loss": 0.76672298,
      "num_input_tokens_seen": 44438095,
      "step": 2080,
      "time_per_iteration": 2.6499993801116943
    },
    {
      "auxiliary_loss_clip": 0.01157323,
      "auxiliary_loss_mlp": 0.01035571,
      "balance_loss_clip": 1.05306995,
      "balance_loss_mlp": 1.02587962,
      "epoch": 0.2502254554199483,
      "flos": 23441193365760.0,
      "grad_norm": 2.323326325765046,
      "language_loss": 0.82425666,
      "learning_rate": 3.5126115874253967e-06,
      "loss": 0.84618562,
      "num_input_tokens_seen": 44457650,
      "step": 2081,
      "time_per_iteration": 2.5925650596618652
    },
    {
      "auxiliary_loss_clip": 0.01164205,
      "auxiliary_loss_mlp": 0.01036635,
      "balance_loss_clip": 1.05860615,
      "balance_loss_mlp": 1.02758682,
      "epoch": 0.2503456983105874,
      "flos": 28761681651840.0,
      "grad_norm": 1.7880296153258421,
      "language_loss": 0.81045371,
      "learning_rate": 3.5121018552228644e-06,
      "loss": 0.83246213,
      "num_input_tokens_seen": 44476155,
      "step": 2082,
      "time_per_iteration": 2.613863945007324
    },
    {
      "auxiliary_loss_clip": 0.01164347,
      "auxiliary_loss_mlp": 0.01033491,
      "balance_loss_clip": 1.05686331,
      "balance_loss_mlp": 1.02442479,
      "epoch": 0.2504659412012265,
      "flos": 18770256673920.0,
      "grad_norm": 1.9689271835918365,
      "language_loss": 0.76019555,
      "learning_rate": 3.5115918936356827e-06,
      "loss": 0.78217387,
      "num_input_tokens_seen": 44492910,
      "step": 2083,
      "time_per_iteration": 2.5620858669281006
    },
    {
      "auxiliary_loss_clip": 0.01141699,
      "auxiliary_loss_mlp": 0.01045445,
      "balance_loss_clip": 1.05294764,
      "balance_loss_mlp": 1.03678417,
      "epoch": 0.25058618409186556,
      "flos": 16873383346560.0,
      "grad_norm": 1.819986072040086,
      "language_loss": 0.78559077,
      "learning_rate": 3.5110817027412123e-06,
      "loss": 0.80746222,
      "num_input_tokens_seen": 44512000,
      "step": 2084,
      "time_per_iteration": 2.539010524749756
    },
    {
      "auxiliary_loss_clip": 0.01159257,
      "auxiliary_loss_mlp": 0.01029848,
      "balance_loss_clip": 1.05388248,
      "balance_loss_mlp": 1.02149701,
      "epoch": 0.25070642698250467,
      "flos": 24425540651520.0,
      "grad_norm": 1.789950671282387,
      "language_loss": 0.68647921,
      "learning_rate": 3.5105712826168493e-06,
      "loss": 0.70837021,
      "num_input_tokens_seen": 44531650,
      "step": 2085,
      "time_per_iteration": 2.6275196075439453
    },
    {
      "auxiliary_loss_clip": 0.01191029,
      "auxiliary_loss_mlp": 0.00762286,
      "balance_loss_clip": 1.05962276,
      "balance_loss_mlp": 1.00037646,
      "epoch": 0.2508266698731437,
      "flos": 20260944028800.0,
      "grad_norm": 1.8017950535184446,
      "language_loss": 0.71048778,
      "learning_rate": 3.5100606333400235e-06,
      "loss": 0.73002088,
      "num_input_tokens_seen": 44548785,
      "step": 2086,
      "time_per_iteration": 2.5097768306732178
    },
    {
      "auxiliary_loss_clip": 0.01186012,
      "auxiliary_loss_mlp": 0.01039894,
      "balance_loss_clip": 1.05873084,
      "balance_loss_mlp": 1.02977324,
      "epoch": 0.25094691276378284,
      "flos": 19245318975360.0,
      "grad_norm": 2.4573984633943464,
      "language_loss": 0.76862836,
      "learning_rate": 3.5095497549882006e-06,
      "loss": 0.79088742,
      "num_input_tokens_seen": 44567230,
      "step": 2087,
      "time_per_iteration": 2.5426225662231445
    },
    {
      "auxiliary_loss_clip": 0.01193081,
      "auxiliary_loss_mlp": 0.01035906,
      "balance_loss_clip": 1.06319857,
      "balance_loss_mlp": 1.02611935,
      "epoch": 0.25106715565442195,
      "flos": 26943237671040.0,
      "grad_norm": 1.8356329869092307,
      "language_loss": 0.72263098,
      "learning_rate": 3.50903864763888e-06,
      "loss": 0.74492085,
      "num_input_tokens_seen": 44588020,
      "step": 2088,
      "time_per_iteration": 2.623161554336548
    },
    {
      "auxiliary_loss_clip": 0.01196381,
      "auxiliary_loss_mlp": 0.010336,
      "balance_loss_clip": 1.06106842,
      "balance_loss_mlp": 1.02337766,
      "epoch": 0.251187398545061,
      "flos": 48359570572800.0,
      "grad_norm": 1.9717420473041052,
      "language_loss": 0.76628649,
      "learning_rate": 3.5085273113695965e-06,
      "loss": 0.78858632,
      "num_input_tokens_seen": 44612590,
      "step": 2089,
      "time_per_iteration": 2.7457163333892822
    },
    {
      "auxiliary_loss_clip": 0.01209828,
      "auxiliary_loss_mlp": 0.0104076,
      "balance_loss_clip": 1.06389046,
      "balance_loss_mlp": 1.03130674,
      "epoch": 0.2513076414357001,
      "flos": 27016100409600.0,
      "grad_norm": 1.734323545093817,
      "language_loss": 0.78405309,
      "learning_rate": 3.508015746257919e-06,
      "loss": 0.80655897,
      "num_input_tokens_seen": 44631630,
      "step": 2090,
      "time_per_iteration": 2.5342354774475098
    },
    {
      "auxiliary_loss_clip": 0.01166418,
      "auxiliary_loss_mlp": 0.01042085,
      "balance_loss_clip": 1.0587194,
      "balance_loss_mlp": 1.03219056,
      "epoch": 0.2514278843263392,
      "flos": 19463619882240.0,
      "grad_norm": 2.116030521161892,
      "language_loss": 0.83703411,
      "learning_rate": 3.5075039523814518e-06,
      "loss": 0.85911918,
      "num_input_tokens_seen": 44650820,
      "step": 2091,
      "time_per_iteration": 2.5807902812957764
    },
    {
      "auxiliary_loss_clip": 0.01195807,
      "auxiliary_loss_mlp": 0.01045744,
      "balance_loss_clip": 1.05988741,
      "balance_loss_mlp": 1.03544414,
      "epoch": 0.2515481272169783,
      "flos": 16866092885760.0,
      "grad_norm": 2.095580782107945,
      "language_loss": 0.81912535,
      "learning_rate": 3.506991929817834e-06,
      "loss": 0.84154093,
      "num_input_tokens_seen": 44667540,
      "step": 2092,
      "time_per_iteration": 3.259746551513672
    },
    {
      "auxiliary_loss_clip": 0.01204167,
      "auxiliary_loss_mlp": 0.01041876,
      "balance_loss_clip": 1.0632726,
      "balance_loss_mlp": 1.03263104,
      "epoch": 0.2516683701076174,
      "flos": 23732464752000.0,
      "grad_norm": 2.023346688494999,
      "language_loss": 0.82526082,
      "learning_rate": 3.506479678644738e-06,
      "loss": 0.84772122,
      "num_input_tokens_seen": 44687935,
      "step": 2093,
      "time_per_iteration": 2.5011889934539795
    },
    {
      "auxiliary_loss_clip": 0.01145885,
      "auxiliary_loss_mlp": 0.01043841,
      "balance_loss_clip": 1.0565896,
      "balance_loss_mlp": 1.03401804,
      "epoch": 0.2517886129982565,
      "flos": 27635954434560.0,
      "grad_norm": 3.581113409802518,
      "language_loss": 0.73980832,
      "learning_rate": 3.505967198939873e-06,
      "loss": 0.76170564,
      "num_input_tokens_seen": 44704975,
      "step": 2094,
      "time_per_iteration": 2.654013156890869
    },
    {
      "auxiliary_loss_clip": 0.01173998,
      "auxiliary_loss_mlp": 0.01036961,
      "balance_loss_clip": 1.05433595,
      "balance_loss_mlp": 1.02756119,
      "epoch": 0.25190885588889556,
      "flos": 38104596529920.0,
      "grad_norm": 2.115135815112557,
      "language_loss": 0.78394109,
      "learning_rate": 3.5054544907809813e-06,
      "loss": 0.80605066,
      "num_input_tokens_seen": 44725475,
      "step": 2095,
      "time_per_iteration": 2.6835176944732666
    },
    {
      "auxiliary_loss_clip": 0.011748,
      "auxiliary_loss_mlp": 0.00763251,
      "balance_loss_clip": 1.06069016,
      "balance_loss_mlp": 1.00034046,
      "epoch": 0.25202909877953467,
      "flos": 22269894768000.0,
      "grad_norm": 2.462911772878111,
      "language_loss": 0.80014294,
      "learning_rate": 3.50494155424584e-06,
      "loss": 0.81952345,
      "num_input_tokens_seen": 44744380,
      "step": 2096,
      "time_per_iteration": 2.5501749515533447
    },
    {
      "auxiliary_loss_clip": 0.01196865,
      "auxiliary_loss_mlp": 0.01035615,
      "balance_loss_clip": 1.06081152,
      "balance_loss_mlp": 1.02576864,
      "epoch": 0.2521493416701738,
      "flos": 21761759018880.0,
      "grad_norm": 1.972974160286277,
      "language_loss": 0.83279347,
      "learning_rate": 3.504428389412262e-06,
      "loss": 0.85511827,
      "num_input_tokens_seen": 44765190,
      "step": 2097,
      "time_per_iteration": 2.5111234188079834
    },
    {
      "auxiliary_loss_clip": 0.01192198,
      "auxiliary_loss_mlp": 0.01038947,
      "balance_loss_clip": 1.05998671,
      "balance_loss_mlp": 1.02970874,
      "epoch": 0.25226958456081283,
      "flos": 27746738956800.0,
      "grad_norm": 2.319830016867886,
      "language_loss": 0.72900105,
      "learning_rate": 3.5039149963580927e-06,
      "loss": 0.75131249,
      "num_input_tokens_seen": 44785210,
      "step": 2098,
      "time_per_iteration": 3.342381000518799
    },
    {
      "auxiliary_loss_clip": 0.01173367,
      "auxiliary_loss_mlp": 0.0103728,
      "balance_loss_clip": 1.05992961,
      "balance_loss_mlp": 1.02814806,
      "epoch": 0.25238982745145194,
      "flos": 30732171903360.0,
      "grad_norm": 2.900700741504068,
      "language_loss": 0.70031005,
      "learning_rate": 3.503401375161215e-06,
      "loss": 0.72241652,
      "num_input_tokens_seen": 44804955,
      "step": 2099,
      "time_per_iteration": 2.5933611392974854
    },
    {
      "auxiliary_loss_clip": 0.01205228,
      "auxiliary_loss_mlp": 0.01035232,
      "balance_loss_clip": 1.06217742,
      "balance_loss_mlp": 1.02559412,
      "epoch": 0.252510070342091,
      "flos": 20266331068800.0,
      "grad_norm": 1.4975971102277843,
      "language_loss": 0.84040827,
      "learning_rate": 3.502887525899544e-06,
      "loss": 0.86281288,
      "num_input_tokens_seen": 44823935,
      "step": 2100,
      "time_per_iteration": 3.214930534362793
    },
    {
      "auxiliary_loss_clip": 0.01180835,
      "auxiliary_loss_mlp": 0.01038398,
      "balance_loss_clip": 1.06130552,
      "balance_loss_mlp": 1.02837813,
      "epoch": 0.2526303132327301,
      "flos": 22747399194240.0,
      "grad_norm": 2.6544588653006915,
      "language_loss": 0.82965523,
      "learning_rate": 3.50237344865103e-06,
      "loss": 0.85184753,
      "num_input_tokens_seen": 44844935,
      "step": 2101,
      "time_per_iteration": 3.407270908355713
    },
    {
      "auxiliary_loss_clip": 0.01208147,
      "auxiliary_loss_mlp": 0.01039477,
      "balance_loss_clip": 1.06271935,
      "balance_loss_mlp": 1.03008342,
      "epoch": 0.2527505561233692,
      "flos": 30263466309120.0,
      "grad_norm": 2.5576713274023155,
      "language_loss": 0.76497495,
      "learning_rate": 3.501859143493658e-06,
      "loss": 0.78745115,
      "num_input_tokens_seen": 44865565,
      "step": 2102,
      "time_per_iteration": 2.5647566318511963
    },
    {
      "auxiliary_loss_clip": 0.0110545,
      "auxiliary_loss_mlp": 0.01009973,
      "balance_loss_clip": 1.03026092,
      "balance_loss_mlp": 1.00786293,
      "epoch": 0.2528707990140083,
      "flos": 58492917164160.0,
      "grad_norm": 0.9579107086231944,
      "language_loss": 0.60540617,
      "learning_rate": 3.5013446105054488e-06,
      "loss": 0.62656045,
      "num_input_tokens_seen": 44918485,
      "step": 2103,
      "time_per_iteration": 2.8456525802612305
    },
    {
      "auxiliary_loss_clip": 0.0114601,
      "auxiliary_loss_mlp": 0.01033429,
      "balance_loss_clip": 1.05182552,
      "balance_loss_mlp": 1.02411246,
      "epoch": 0.2529910419046474,
      "flos": 24645134448000.0,
      "grad_norm": 2.06065472405125,
      "language_loss": 0.74976468,
      "learning_rate": 3.5008298497644555e-06,
      "loss": 0.77155918,
      "num_input_tokens_seen": 44937530,
      "step": 2104,
      "time_per_iteration": 2.583155393600464
    },
    {
      "auxiliary_loss_clip": 0.01166673,
      "auxiliary_loss_mlp": 0.01036432,
      "balance_loss_clip": 1.05919147,
      "balance_loss_mlp": 1.02643633,
      "epoch": 0.2531112847952865,
      "flos": 23842135952640.0,
      "grad_norm": 1.6561198917650972,
      "language_loss": 0.87785357,
      "learning_rate": 3.500314861348767e-06,
      "loss": 0.89988458,
      "num_input_tokens_seen": 44958165,
      "step": 2105,
      "time_per_iteration": 2.618145704269409
    },
    {
      "auxiliary_loss_clip": 0.01154501,
      "auxiliary_loss_mlp": 0.01039661,
      "balance_loss_clip": 1.058846,
      "balance_loss_mlp": 1.02983856,
      "epoch": 0.25323152768592555,
      "flos": 16143822207360.0,
      "grad_norm": 1.7912937061033667,
      "language_loss": 0.77223045,
      "learning_rate": 3.499799645336507e-06,
      "loss": 0.79417205,
      "num_input_tokens_seen": 44975060,
      "step": 2106,
      "time_per_iteration": 2.5525567531585693
    },
    {
      "auxiliary_loss_clip": 0.01194877,
      "auxiliary_loss_mlp": 0.01031485,
      "balance_loss_clip": 1.06285405,
      "balance_loss_mlp": 1.02284861,
      "epoch": 0.25335177057656466,
      "flos": 28405161210240.0,
      "grad_norm": 2.0471050882725623,
      "language_loss": 0.86910975,
      "learning_rate": 3.4992842018058336e-06,
      "loss": 0.8913734,
      "num_input_tokens_seen": 44997960,
      "step": 2107,
      "time_per_iteration": 2.6110053062438965
    },
    {
      "auxiliary_loss_clip": 0.0116926,
      "auxiliary_loss_mlp": 0.01032052,
      "balance_loss_clip": 1.0570128,
      "balance_loss_mlp": 1.02326012,
      "epoch": 0.25347201346720377,
      "flos": 18799666934400.0,
      "grad_norm": 1.9113913928393098,
      "language_loss": 0.88225496,
      "learning_rate": 3.4987685308349384e-06,
      "loss": 0.90426815,
      "num_input_tokens_seen": 45015690,
      "step": 2108,
      "time_per_iteration": 2.5520761013031006
    },
    {
      "auxiliary_loss_clip": 0.01157722,
      "auxiliary_loss_mlp": 0.01038326,
      "balance_loss_clip": 1.05107427,
      "balance_loss_mlp": 1.02908134,
      "epoch": 0.2535922563578428,
      "flos": 15815490963840.0,
      "grad_norm": 2.0307845871271626,
      "language_loss": 0.61369467,
      "learning_rate": 3.4982526325020497e-06,
      "loss": 0.63565516,
      "num_input_tokens_seen": 45032660,
      "step": 2109,
      "time_per_iteration": 2.541287660598755
    },
    {
      "auxiliary_loss_clip": 0.01179013,
      "auxiliary_loss_mlp": 0.01038382,
      "balance_loss_clip": 1.0585525,
      "balance_loss_mlp": 1.02906537,
      "epoch": 0.25371249924848194,
      "flos": 16318922031360.0,
      "grad_norm": 2.049653496888196,
      "language_loss": 0.81709492,
      "learning_rate": 3.4977365068854273e-06,
      "loss": 0.8392688,
      "num_input_tokens_seen": 45048280,
      "step": 2110,
      "time_per_iteration": 2.5073964595794678
    },
    {
      "auxiliary_loss_clip": 0.01164415,
      "auxiliary_loss_mlp": 0.01033067,
      "balance_loss_clip": 1.05391324,
      "balance_loss_mlp": 1.02304125,
      "epoch": 0.25383274213912105,
      "flos": 21761615364480.0,
      "grad_norm": 1.7913571643324888,
      "language_loss": 0.73636186,
      "learning_rate": 3.4972201540633676e-06,
      "loss": 0.75833666,
      "num_input_tokens_seen": 45067635,
      "step": 2111,
      "time_per_iteration": 2.563694715499878
    },
    {
      "auxiliary_loss_clip": 0.01170028,
      "auxiliary_loss_mlp": 0.0103352,
      "balance_loss_clip": 1.05812955,
      "balance_loss_mlp": 1.02412033,
      "epoch": 0.2539529850297601,
      "flos": 21396870708480.0,
      "grad_norm": 1.8084924242117209,
      "language_loss": 0.85298598,
      "learning_rate": 3.4967035741142008e-06,
      "loss": 0.87502146,
      "num_input_tokens_seen": 45086455,
      "step": 2112,
      "time_per_iteration": 2.5316061973571777
    },
    {
      "auxiliary_loss_clip": 0.0117278,
      "auxiliary_loss_mlp": 0.01040226,
      "balance_loss_clip": 1.06622815,
      "balance_loss_mlp": 1.03164291,
      "epoch": 0.2540732279203992,
      "flos": 25228467319680.0,
      "grad_norm": 1.8243536631638415,
      "language_loss": 0.81574547,
      "learning_rate": 3.4961867671162917e-06,
      "loss": 0.83787549,
      "num_input_tokens_seen": 45106385,
      "step": 2113,
      "time_per_iteration": 2.587557554244995
    },
    {
      "auxiliary_loss_clip": 0.01205034,
      "auxiliary_loss_mlp": 0.01033504,
      "balance_loss_clip": 1.06028807,
      "balance_loss_mlp": 1.02365136,
      "epoch": 0.2541934708110383,
      "flos": 19427386037760.0,
      "grad_norm": 2.639936067526849,
      "language_loss": 0.77162588,
      "learning_rate": 3.4956697331480402e-06,
      "loss": 0.79401124,
      "num_input_tokens_seen": 45124955,
      "step": 2114,
      "time_per_iteration": 2.4816300868988037
    },
    {
      "auxiliary_loss_clip": 0.01167254,
      "auxiliary_loss_mlp": 0.01034429,
      "balance_loss_clip": 1.05588734,
      "balance_loss_mlp": 1.02552462,
      "epoch": 0.2543137137016774,
      "flos": 23949436855680.0,
      "grad_norm": 1.5278355720973897,
      "language_loss": 0.79945493,
      "learning_rate": 3.495152472287879e-06,
      "loss": 0.82147175,
      "num_input_tokens_seen": 45145665,
      "step": 2115,
      "time_per_iteration": 2.601029872894287
    },
    {
      "auxiliary_loss_clip": 0.01164769,
      "auxiliary_loss_mlp": 0.01038552,
      "balance_loss_clip": 1.05813777,
      "balance_loss_mlp": 1.02868724,
      "epoch": 0.2544339565923165,
      "flos": 25593283802880.0,
      "grad_norm": 1.8267900948925542,
      "language_loss": 0.73949993,
      "learning_rate": 3.4946349846142766e-06,
      "loss": 0.76153314,
      "num_input_tokens_seen": 45164805,
      "step": 2116,
      "time_per_iteration": 2.5854198932647705
    },
    {
      "auxiliary_loss_clip": 0.0120564,
      "auxiliary_loss_mlp": 0.01031083,
      "balance_loss_clip": 1.06076467,
      "balance_loss_mlp": 1.02137375,
      "epoch": 0.25455419948295555,
      "flos": 21689470897920.0,
      "grad_norm": 2.1520396583895334,
      "language_loss": 0.76209307,
      "learning_rate": 3.4941172702057353e-06,
      "loss": 0.78446031,
      "num_input_tokens_seen": 45184865,
      "step": 2117,
      "time_per_iteration": 2.477078914642334
    },
    {
      "auxiliary_loss_clip": 0.01179193,
      "auxiliary_loss_mlp": 0.01033098,
      "balance_loss_clip": 1.06077087,
      "balance_loss_mlp": 1.02341866,
      "epoch": 0.25467444237359466,
      "flos": 26250341339520.0,
      "grad_norm": 2.168707876738852,
      "language_loss": 0.80418956,
      "learning_rate": 3.4935993291407924e-06,
      "loss": 0.82631242,
      "num_input_tokens_seen": 45203690,
      "step": 2118,
      "time_per_iteration": 3.2881507873535156
    },
    {
      "auxiliary_loss_clip": 0.01172635,
      "auxiliary_loss_mlp": 0.01034101,
      "balance_loss_clip": 1.05657125,
      "balance_loss_mlp": 1.02478528,
      "epoch": 0.25479468526423377,
      "flos": 26979686997120.0,
      "grad_norm": 2.1901589668165258,
      "language_loss": 0.71054924,
      "learning_rate": 3.4930811614980183e-06,
      "loss": 0.7326166,
      "num_input_tokens_seen": 45225385,
      "step": 2119,
      "time_per_iteration": 2.574782371520996
    },
    {
      "auxiliary_loss_clip": 0.01185267,
      "auxiliary_loss_mlp": 0.01038367,
      "balance_loss_clip": 1.05944121,
      "balance_loss_mlp": 1.02800775,
      "epoch": 0.2549149281548728,
      "flos": 23475811098240.0,
      "grad_norm": 1.7464346767000507,
      "language_loss": 0.7960819,
      "learning_rate": 3.4925627673560198e-06,
      "loss": 0.81831819,
      "num_input_tokens_seen": 45246045,
      "step": 2120,
      "time_per_iteration": 2.5333878993988037
    },
    {
      "auxiliary_loss_clip": 0.01160651,
      "auxiliary_loss_mlp": 0.01033497,
      "balance_loss_clip": 1.05622363,
      "balance_loss_mlp": 1.02438354,
      "epoch": 0.25503517104551193,
      "flos": 25812302981760.0,
      "grad_norm": 2.0210532472431204,
      "language_loss": 0.88352448,
      "learning_rate": 3.4920441467934357e-06,
      "loss": 0.90546596,
      "num_input_tokens_seen": 45266560,
      "step": 2121,
      "time_per_iteration": 2.5967423915863037
    },
    {
      "auxiliary_loss_clip": 0.01155553,
      "auxiliary_loss_mlp": 0.01034812,
      "balance_loss_clip": 1.05687952,
      "balance_loss_mlp": 1.02560937,
      "epoch": 0.25515541393615104,
      "flos": 26645106787200.0,
      "grad_norm": 5.16058382535851,
      "language_loss": 0.82863748,
      "learning_rate": 3.491525299888941e-06,
      "loss": 0.85054111,
      "num_input_tokens_seen": 45285405,
      "step": 2122,
      "time_per_iteration": 2.6054956912994385
    },
    {
      "auxiliary_loss_clip": 0.01082217,
      "auxiliary_loss_mlp": 0.00753036,
      "balance_loss_clip": 1.04047155,
      "balance_loss_mlp": 1.00000405,
      "epoch": 0.2552756568267901,
      "flos": 65955945847680.0,
      "grad_norm": 0.9214712692686856,
      "language_loss": 0.62708187,
      "learning_rate": 3.491006226721244e-06,
      "loss": 0.64543438,
      "num_input_tokens_seen": 45349615,
      "step": 2123,
      "time_per_iteration": 3.9358623027801514
    },
    {
      "auxiliary_loss_clip": 0.01182365,
      "auxiliary_loss_mlp": 0.00762453,
      "balance_loss_clip": 1.06347585,
      "balance_loss_mlp": 1.00030303,
      "epoch": 0.2553958997174292,
      "flos": 17931096161280.0,
      "grad_norm": 2.104600949251835,
      "language_loss": 0.77562422,
      "learning_rate": 3.4904869273690882e-06,
      "loss": 0.79507244,
      "num_input_tokens_seen": 45367505,
      "step": 2124,
      "time_per_iteration": 2.5291388034820557
    },
    {
      "auxiliary_loss_clip": 0.01194659,
      "auxiliary_loss_mlp": 0.0103409,
      "balance_loss_clip": 1.05992293,
      "balance_loss_mlp": 1.02408862,
      "epoch": 0.2555161426080683,
      "flos": 23367791923200.0,
      "grad_norm": 1.926245537530003,
      "language_loss": 0.88723183,
      "learning_rate": 3.489967401911251e-06,
      "loss": 0.90951937,
      "num_input_tokens_seen": 45386805,
      "step": 2125,
      "time_per_iteration": 3.2687244415283203
    },
    {
      "auxiliary_loss_clip": 0.01208642,
      "auxiliary_loss_mlp": 0.01038527,
      "balance_loss_clip": 1.06253338,
      "balance_loss_mlp": 1.02859688,
      "epoch": 0.2556363854987074,
      "flos": 40625130723840.0,
      "grad_norm": 1.5987102161299322,
      "language_loss": 0.69592202,
      "learning_rate": 3.4894476504265428e-06,
      "loss": 0.71839368,
      "num_input_tokens_seen": 45411045,
      "step": 2126,
      "time_per_iteration": 2.6631603240966797
    },
    {
      "auxiliary_loss_clip": 0.01086103,
      "auxiliary_loss_mlp": 0.01001878,
      "balance_loss_clip": 1.02572727,
      "balance_loss_mlp": 0.99975616,
      "epoch": 0.2557566283893465,
      "flos": 68019443389440.0,
      "grad_norm": 0.7400381983311672,
      "language_loss": 0.54510289,
      "learning_rate": 3.4889276729938104e-06,
      "loss": 0.5659827,
      "num_input_tokens_seen": 45469575,
      "step": 2127,
      "time_per_iteration": 3.8222787380218506
    },
    {
      "auxiliary_loss_clip": 0.01172341,
      "auxiliary_loss_mlp": 0.01035155,
      "balance_loss_clip": 1.0580194,
      "balance_loss_mlp": 1.02580261,
      "epoch": 0.2558768712799856,
      "flos": 22635645004800.0,
      "grad_norm": 3.9151787878578794,
      "language_loss": 0.80676526,
      "learning_rate": 3.488407469691934e-06,
      "loss": 0.8288402,
      "num_input_tokens_seen": 45490270,
      "step": 2128,
      "time_per_iteration": 2.540405750274658
    },
    {
      "auxiliary_loss_clip": 0.01172698,
      "auxiliary_loss_mlp": 0.01037838,
      "balance_loss_clip": 1.05440748,
      "balance_loss_mlp": 1.02772927,
      "epoch": 0.25599711417062465,
      "flos": 26396354125440.0,
      "grad_norm": 2.036809592874757,
      "language_loss": 0.8094548,
      "learning_rate": 3.487887040599828e-06,
      "loss": 0.83156013,
      "num_input_tokens_seen": 45510070,
      "step": 2129,
      "time_per_iteration": 2.603438377380371
    },
    {
      "auxiliary_loss_clip": 0.01209265,
      "auxiliary_loss_mlp": 0.01036479,
      "balance_loss_clip": 1.06360817,
      "balance_loss_mlp": 1.02717423,
      "epoch": 0.25611735706126376,
      "flos": 22852042490880.0,
      "grad_norm": 2.6839999517770234,
      "language_loss": 0.76344538,
      "learning_rate": 3.4873663857964407e-06,
      "loss": 0.7859028,
      "num_input_tokens_seen": 45527285,
      "step": 2130,
      "time_per_iteration": 2.4817733764648438
    },
    {
      "auxiliary_loss_clip": 0.01147692,
      "auxiliary_loss_mlp": 0.01041692,
      "balance_loss_clip": 1.05758286,
      "balance_loss_mlp": 1.03275764,
      "epoch": 0.2562375999519028,
      "flos": 23367863750400.0,
      "grad_norm": 1.7994728708736876,
      "language_loss": 0.66438794,
      "learning_rate": 3.4868455053607556e-06,
      "loss": 0.6862818,
      "num_input_tokens_seen": 45546900,
      "step": 2131,
      "time_per_iteration": 2.613725185394287
    },
    {
      "auxiliary_loss_clip": 0.01195561,
      "auxiliary_loss_mlp": 0.01034167,
      "balance_loss_clip": 1.05935526,
      "balance_loss_mlp": 1.02393878,
      "epoch": 0.2563578428425419,
      "flos": 22856962654080.0,
      "grad_norm": 2.2093335964952305,
      "language_loss": 0.71906871,
      "learning_rate": 3.486324399371789e-06,
      "loss": 0.74136603,
      "num_input_tokens_seen": 45566200,
      "step": 2132,
      "time_per_iteration": 2.5071308612823486
    },
    {
      "auxiliary_loss_clip": 0.01159795,
      "auxiliary_loss_mlp": 0.01030304,
      "balance_loss_clip": 1.0591054,
      "balance_loss_mlp": 1.02213836,
      "epoch": 0.25647808573318104,
      "flos": 21653883498240.0,
      "grad_norm": 1.985262301160189,
      "language_loss": 0.78388047,
      "learning_rate": 3.485803067908593e-06,
      "loss": 0.80578148,
      "num_input_tokens_seen": 45585710,
      "step": 2133,
      "time_per_iteration": 2.5696043968200684
    },
    {
      "auxiliary_loss_clip": 0.01106925,
      "auxiliary_loss_mlp": 0.01031301,
      "balance_loss_clip": 1.04517937,
      "balance_loss_mlp": 1.02140033,
      "epoch": 0.2565983286238201,
      "flos": 33730569659520.0,
      "grad_norm": 2.177025844082378,
      "language_loss": 0.7928834,
      "learning_rate": 3.485281511050253e-06,
      "loss": 0.81426561,
      "num_input_tokens_seen": 45607845,
      "step": 2134,
      "time_per_iteration": 2.725576639175415
    },
    {
      "auxiliary_loss_clip": 0.0119225,
      "auxiliary_loss_mlp": 0.01034995,
      "balance_loss_clip": 1.05870152,
      "balance_loss_mlp": 1.02563739,
      "epoch": 0.2567185715144592,
      "flos": 16216002587520.0,
      "grad_norm": 2.500840084269615,
      "language_loss": 0.90160578,
      "learning_rate": 3.484759728875889e-06,
      "loss": 0.92387819,
      "num_input_tokens_seen": 45623210,
      "step": 2135,
      "time_per_iteration": 2.470620632171631
    },
    {
      "auxiliary_loss_clip": 0.011373,
      "auxiliary_loss_mlp": 0.01039782,
      "balance_loss_clip": 1.05258298,
      "balance_loss_mlp": 1.03037083,
      "epoch": 0.2568388144050983,
      "flos": 17458475984640.0,
      "grad_norm": 1.9506446087134908,
      "language_loss": 0.80931163,
      "learning_rate": 3.4842377214646543e-06,
      "loss": 0.83108246,
      "num_input_tokens_seen": 45641505,
      "step": 2136,
      "time_per_iteration": 2.573622703552246
    },
    {
      "auxiliary_loss_clip": 0.01205844,
      "auxiliary_loss_mlp": 0.01035308,
      "balance_loss_clip": 1.06353736,
      "balance_loss_mlp": 1.02668917,
      "epoch": 0.25695905729573737,
      "flos": 20887442069760.0,
      "grad_norm": 1.8315349626897552,
      "language_loss": 0.66627455,
      "learning_rate": 3.483715488895737e-06,
      "loss": 0.68868601,
      "num_input_tokens_seen": 45661835,
      "step": 2137,
      "time_per_iteration": 2.4838802814483643
    },
    {
      "auxiliary_loss_clip": 0.0114672,
      "auxiliary_loss_mlp": 0.01032545,
      "balance_loss_clip": 1.05340874,
      "balance_loss_mlp": 1.02355707,
      "epoch": 0.2570793001863765,
      "flos": 24717278914560.0,
      "grad_norm": 2.0158119872474765,
      "language_loss": 0.78894293,
      "learning_rate": 3.48319303124836e-06,
      "loss": 0.81073558,
      "num_input_tokens_seen": 45682215,
      "step": 2138,
      "time_per_iteration": 2.63031005859375
    },
    {
      "auxiliary_loss_clip": 0.01172604,
      "auxiliary_loss_mlp": 0.01033173,
      "balance_loss_clip": 1.06052518,
      "balance_loss_mlp": 1.02430952,
      "epoch": 0.2571995430770156,
      "flos": 26906896085760.0,
      "grad_norm": 2.793456369143242,
      "language_loss": 0.66679883,
      "learning_rate": 3.4826703486017798e-06,
      "loss": 0.6888566,
      "num_input_tokens_seen": 45701840,
      "step": 2139,
      "time_per_iteration": 2.582402467727661
    },
    {
      "auxiliary_loss_clip": 0.01190277,
      "auxiliary_loss_mlp": 0.0103021,
      "balance_loss_clip": 1.06154346,
      "balance_loss_mlp": 1.02091169,
      "epoch": 0.25731978596765465,
      "flos": 19792561656960.0,
      "grad_norm": 1.6031812941327097,
      "language_loss": 0.76937962,
      "learning_rate": 3.4821474410352867e-06,
      "loss": 0.79158449,
      "num_input_tokens_seen": 45720500,
      "step": 2140,
      "time_per_iteration": 2.5038199424743652
    },
    {
      "auxiliary_loss_clip": 0.01060706,
      "auxiliary_loss_mlp": 0.01007477,
      "balance_loss_clip": 1.02169466,
      "balance_loss_mlp": 1.00536728,
      "epoch": 0.25744002885829376,
      "flos": 70564970471040.0,
      "grad_norm": 0.9061539242779558,
      "language_loss": 0.62688702,
      "learning_rate": 3.481624308628205e-06,
      "loss": 0.64756888,
      "num_input_tokens_seen": 45781870,
      "step": 2141,
      "time_per_iteration": 3.2573835849761963
    },
    {
      "auxiliary_loss_clip": 0.01175275,
      "auxiliary_loss_mlp": 0.01031658,
      "balance_loss_clip": 1.05849123,
      "balance_loss_mlp": 1.02153158,
      "epoch": 0.25756027174893287,
      "flos": 18038181582720.0,
      "grad_norm": 2.556395930723889,
      "language_loss": 1.00268567,
      "learning_rate": 3.481100951459893e-06,
      "loss": 1.024755,
      "num_input_tokens_seen": 45794890,
      "step": 2142,
      "time_per_iteration": 2.503222703933716
    },
    {
      "auxiliary_loss_clip": 0.01187605,
      "auxiliary_loss_mlp": 0.01036973,
      "balance_loss_clip": 1.05879259,
      "balance_loss_mlp": 1.02869415,
      "epoch": 0.2576805146395719,
      "flos": 22674069578880.0,
      "grad_norm": 2.274219109170958,
      "language_loss": 0.78566885,
      "learning_rate": 3.4805773696097453e-06,
      "loss": 0.80791461,
      "num_input_tokens_seen": 45815780,
      "step": 2143,
      "time_per_iteration": 2.5410521030426025
    },
    {
      "auxiliary_loss_clip": 0.01166605,
      "auxiliary_loss_mlp": 0.01028364,
      "balance_loss_clip": 1.05726957,
      "balance_loss_mlp": 1.01969147,
      "epoch": 0.25780075753021103,
      "flos": 16472225278080.0,
      "grad_norm": 2.004631116964035,
      "language_loss": 0.8739922,
      "learning_rate": 3.4800535631571874e-06,
      "loss": 0.89594185,
      "num_input_tokens_seen": 45831310,
      "step": 2144,
      "time_per_iteration": 3.221395969390869
    },
    {
      "auxiliary_loss_clip": 0.0117922,
      "auxiliary_loss_mlp": 0.01033953,
      "balance_loss_clip": 1.05859828,
      "balance_loss_mlp": 1.02435052,
      "epoch": 0.25792100042085014,
      "flos": 22820297846400.0,
      "grad_norm": 1.936042699235944,
      "language_loss": 0.75970483,
      "learning_rate": 3.4795295321816804e-06,
      "loss": 0.78183663,
      "num_input_tokens_seen": 45850135,
      "step": 2145,
      "time_per_iteration": 2.5444693565368652
    },
    {
      "auxiliary_loss_clip": 0.01161884,
      "auxiliary_loss_mlp": 0.01032447,
      "balance_loss_clip": 1.05561686,
      "balance_loss_mlp": 1.02289844,
      "epoch": 0.2580412433114892,
      "flos": 18697286194560.0,
      "grad_norm": 2.0181284868841667,
      "language_loss": 0.91005611,
      "learning_rate": 3.47900527676272e-06,
      "loss": 0.93199944,
      "num_input_tokens_seen": 45868470,
      "step": 2146,
      "time_per_iteration": 2.507291078567505
    },
    {
      "auxiliary_loss_clip": 0.01207153,
      "auxiliary_loss_mlp": 0.01033045,
      "balance_loss_clip": 1.06362677,
      "balance_loss_mlp": 1.02353263,
      "epoch": 0.2581614862021283,
      "flos": 14283146810880.0,
      "grad_norm": 1.9621115680944143,
      "language_loss": 0.88024795,
      "learning_rate": 3.478480796979835e-06,
      "loss": 0.90264988,
      "num_input_tokens_seen": 45886355,
      "step": 2147,
      "time_per_iteration": 2.460406541824341
    },
    {
      "auxiliary_loss_clip": 0.0117092,
      "auxiliary_loss_mlp": 0.01035417,
      "balance_loss_clip": 1.05667269,
      "balance_loss_mlp": 1.02690005,
      "epoch": 0.25828172909276736,
      "flos": 29498281856640.0,
      "grad_norm": 1.6179682922230085,
      "language_loss": 0.7777583,
      "learning_rate": 3.4779560929125894e-06,
      "loss": 0.79982167,
      "num_input_tokens_seen": 45907900,
      "step": 2148,
      "time_per_iteration": 2.6158714294433594
    },
    {
      "auxiliary_loss_clip": 0.01059305,
      "auxiliary_loss_mlp": 0.01010434,
      "balance_loss_clip": 1.02169824,
      "balance_loss_mlp": 1.00833571,
      "epoch": 0.2584019719834065,
      "flos": 67114387376640.0,
      "grad_norm": 0.7028056811002855,
      "language_loss": 0.56916738,
      "learning_rate": 3.4774311646405783e-06,
      "loss": 0.58986473,
      "num_input_tokens_seen": 45977805,
      "step": 2149,
      "time_per_iteration": 3.2564096450805664
    },
    {
      "auxiliary_loss_clip": 0.01147708,
      "auxiliary_loss_mlp": 0.01032817,
      "balance_loss_clip": 1.05097175,
      "balance_loss_mlp": 1.02382851,
      "epoch": 0.2585222148740456,
      "flos": 22893555634560.0,
      "grad_norm": 1.9874780152607852,
      "language_loss": 0.83605719,
      "learning_rate": 3.476906012243435e-06,
      "loss": 0.85786235,
      "num_input_tokens_seen": 45996715,
      "step": 2150,
      "time_per_iteration": 3.3424301147460938
    },
    {
      "auxiliary_loss_clip": 0.01178557,
      "auxiliary_loss_mlp": 0.01037597,
      "balance_loss_clip": 1.05830884,
      "balance_loss_mlp": 1.02852476,
      "epoch": 0.25864245776468464,
      "flos": 28909202808960.0,
      "grad_norm": 1.5785337566136524,
      "language_loss": 0.8104434,
      "learning_rate": 3.476380635800824e-06,
      "loss": 0.832605,
      "num_input_tokens_seen": 46017915,
      "step": 2151,
      "time_per_iteration": 2.560530424118042
    },
    {
      "auxiliary_loss_clip": 0.01174491,
      "auxiliary_loss_mlp": 0.0103272,
      "balance_loss_clip": 1.06084287,
      "balance_loss_mlp": 1.02392197,
      "epoch": 0.25876270065532375,
      "flos": 14793185980800.0,
      "grad_norm": 2.188430144407419,
      "language_loss": 0.85692233,
      "learning_rate": 3.475855035392444e-06,
      "loss": 0.87899446,
      "num_input_tokens_seen": 46033235,
      "step": 2152,
      "time_per_iteration": 3.222622871398926
    },
    {
      "auxiliary_loss_clip": 0.01128042,
      "auxiliary_loss_mlp": 0.01040217,
      "balance_loss_clip": 1.052665,
      "balance_loss_mlp": 1.03158081,
      "epoch": 0.25888294354596286,
      "flos": 60467821810560.0,
      "grad_norm": 1.865161822656067,
      "language_loss": 0.71090758,
      "learning_rate": 3.475329211098029e-06,
      "loss": 0.73259008,
      "num_input_tokens_seen": 46056390,
      "step": 2153,
      "time_per_iteration": 3.640693426132202
    },
    {
      "auxiliary_loss_clip": 0.01150272,
      "auxiliary_loss_mlp": 0.01035775,
      "balance_loss_clip": 1.05565381,
      "balance_loss_mlp": 1.02696538,
      "epoch": 0.2590031864366019,
      "flos": 27851166771840.0,
      "grad_norm": 2.0458219043637524,
      "language_loss": 0.82278407,
      "learning_rate": 3.4748031629973453e-06,
      "loss": 0.84464455,
      "num_input_tokens_seen": 46077120,
      "step": 2154,
      "time_per_iteration": 2.6299588680267334
    },
    {
      "auxiliary_loss_clip": 0.01038693,
      "auxiliary_loss_mlp": 0.01001958,
      "balance_loss_clip": 1.01459205,
      "balance_loss_mlp": 0.99977684,
      "epoch": 0.25912342932724103,
      "flos": 62422444206720.0,
      "grad_norm": 0.9211520584524355,
      "language_loss": 0.56637025,
      "learning_rate": 3.4742768911701944e-06,
      "loss": 0.58677673,
      "num_input_tokens_seen": 46139815,
      "step": 2155,
      "time_per_iteration": 3.261922597885132
    },
    {
      "auxiliary_loss_clip": 0.01193268,
      "auxiliary_loss_mlp": 0.01046639,
      "balance_loss_clip": 1.06346941,
      "balance_loss_mlp": 1.03632176,
      "epoch": 0.25924367221788014,
      "flos": 12378839368320.0,
      "grad_norm": 3.697315383357547,
      "language_loss": 0.70035362,
      "learning_rate": 3.4737503956964113e-06,
      "loss": 0.72275269,
      "num_input_tokens_seen": 46152120,
      "step": 2156,
      "time_per_iteration": 2.4388608932495117
    },
    {
      "auxiliary_loss_clip": 0.01166998,
      "auxiliary_loss_mlp": 0.01030629,
      "balance_loss_clip": 1.05508125,
      "balance_loss_mlp": 1.02065122,
      "epoch": 0.2593639151085192,
      "flos": 14575208296320.0,
      "grad_norm": 2.2188666164800503,
      "language_loss": 0.66898608,
      "learning_rate": 3.473223676655865e-06,
      "loss": 0.69096237,
      "num_input_tokens_seen": 46170120,
      "step": 2157,
      "time_per_iteration": 2.4890077114105225
    },
    {
      "auxiliary_loss_clip": 0.01166242,
      "auxiliary_loss_mlp": 0.01034256,
      "balance_loss_clip": 1.05243301,
      "balance_loss_mlp": 1.02449894,
      "epoch": 0.2594841579991583,
      "flos": 15230937029760.0,
      "grad_norm": 1.8775092978324723,
      "language_loss": 0.79973745,
      "learning_rate": 3.472696734128459e-06,
      "loss": 0.82174242,
      "num_input_tokens_seen": 46187985,
      "step": 2158,
      "time_per_iteration": 2.5076100826263428
    },
    {
      "auxiliary_loss_clip": 0.01191141,
      "auxiliary_loss_mlp": 0.01035301,
      "balance_loss_clip": 1.06105661,
      "balance_loss_mlp": 1.02634287,
      "epoch": 0.2596044008897974,
      "flos": 23623583650560.0,
      "grad_norm": 1.7017729438129465,
      "language_loss": 0.75454652,
      "learning_rate": 3.4721695681941286e-06,
      "loss": 0.77681094,
      "num_input_tokens_seen": 46207025,
      "step": 2159,
      "time_per_iteration": 2.519624710083008
    },
    {
      "auxiliary_loss_clip": 0.01174461,
      "auxiliary_loss_mlp": 0.00762126,
      "balance_loss_clip": 1.05773032,
      "balance_loss_mlp": 1.00029922,
      "epoch": 0.25972464378043647,
      "flos": 13772281628160.0,
      "grad_norm": 1.980150612696572,
      "language_loss": 0.82714975,
      "learning_rate": 3.471642178932845e-06,
      "loss": 0.8465156,
      "num_input_tokens_seen": 46225670,
      "step": 2160,
      "time_per_iteration": 2.5327861309051514
    },
    {
      "auxiliary_loss_clip": 0.01174649,
      "auxiliary_loss_mlp": 0.01034334,
      "balance_loss_clip": 1.05456042,
      "balance_loss_mlp": 1.02507758,
      "epoch": 0.2598448866710756,
      "flos": 19573578391680.0,
      "grad_norm": 2.2367871869853295,
      "language_loss": 0.89613599,
      "learning_rate": 3.471114566424613e-06,
      "loss": 0.91822582,
      "num_input_tokens_seen": 46244130,
      "step": 2161,
      "time_per_iteration": 2.520111560821533
    },
    {
      "auxiliary_loss_clip": 0.01173615,
      "auxiliary_loss_mlp": 0.01037601,
      "balance_loss_clip": 1.05972505,
      "balance_loss_mlp": 1.02843332,
      "epoch": 0.25996512956171464,
      "flos": 21653237053440.0,
      "grad_norm": 3.4019081036117758,
      "language_loss": 0.75799209,
      "learning_rate": 3.4705867307494715e-06,
      "loss": 0.78010428,
      "num_input_tokens_seen": 46263200,
      "step": 2162,
      "time_per_iteration": 2.5772922039031982
    },
    {
      "auxiliary_loss_clip": 0.01190555,
      "auxiliary_loss_mlp": 0.01033844,
      "balance_loss_clip": 1.05747652,
      "balance_loss_mlp": 1.02409315,
      "epoch": 0.26008537245235375,
      "flos": 18223480869120.0,
      "grad_norm": 2.5485588748149866,
      "language_loss": 0.84810233,
      "learning_rate": 3.470058671987492e-06,
      "loss": 0.87034631,
      "num_input_tokens_seen": 46281465,
      "step": 2163,
      "time_per_iteration": 2.489537477493286
    },
    {
      "auxiliary_loss_clip": 0.01192774,
      "auxiliary_loss_mlp": 0.01036866,
      "balance_loss_clip": 1.05773246,
      "balance_loss_mlp": 1.02716267,
      "epoch": 0.26020561534299286,
      "flos": 24645385843200.0,
      "grad_norm": 2.9041781916593954,
      "language_loss": 0.84432018,
      "learning_rate": 3.4695303902187805e-06,
      "loss": 0.86661661,
      "num_input_tokens_seen": 46301020,
      "step": 2164,
      "time_per_iteration": 2.6259748935699463
    },
    {
      "auxiliary_loss_clip": 0.01153807,
      "auxiliary_loss_mlp": 0.0103503,
      "balance_loss_clip": 1.05318832,
      "balance_loss_mlp": 1.02606547,
      "epoch": 0.2603258582336319,
      "flos": 25773662926080.0,
      "grad_norm": 1.9600515373652982,
      "language_loss": 0.78715575,
      "learning_rate": 3.469001885523478e-06,
      "loss": 0.80904412,
      "num_input_tokens_seen": 46321740,
      "step": 2165,
      "time_per_iteration": 2.6274774074554443
    },
    {
      "auxiliary_loss_clip": 0.01204329,
      "auxiliary_loss_mlp": 0.01034884,
      "balance_loss_clip": 1.06091666,
      "balance_loss_mlp": 1.02569318,
      "epoch": 0.260446101124271,
      "flos": 28766314506240.0,
      "grad_norm": 1.5581847010125418,
      "language_loss": 0.81077945,
      "learning_rate": 3.4684731579817568e-06,
      "loss": 0.83317155,
      "num_input_tokens_seen": 46342730,
      "step": 2166,
      "time_per_iteration": 2.544976234436035
    },
    {
      "auxiliary_loss_clip": 0.0112493,
      "auxiliary_loss_mlp": 0.0103822,
      "balance_loss_clip": 1.05285692,
      "balance_loss_mlp": 1.02946448,
      "epoch": 0.26056634401491013,
      "flos": 25666757072640.0,
      "grad_norm": 1.9586428654441055,
      "language_loss": 0.7680524,
      "learning_rate": 3.4679442076738247e-06,
      "loss": 0.78968394,
      "num_input_tokens_seen": 46362445,
      "step": 2167,
      "time_per_iteration": 2.6525888442993164
    },
    {
      "auxiliary_loss_clip": 0.01206663,
      "auxiliary_loss_mlp": 0.01035429,
      "balance_loss_clip": 1.06308985,
      "balance_loss_mlp": 1.02526641,
      "epoch": 0.2606865869055492,
      "flos": 27052765217280.0,
      "grad_norm": 1.7606547668052164,
      "language_loss": 0.83506858,
      "learning_rate": 3.4674150346799245e-06,
      "loss": 0.85748953,
      "num_input_tokens_seen": 46382145,
      "step": 2168,
      "time_per_iteration": 2.51381254196167
    },
    {
      "auxiliary_loss_clip": 0.01172954,
      "auxiliary_loss_mlp": 0.0103204,
      "balance_loss_clip": 1.05756521,
      "balance_loss_mlp": 1.02329564,
      "epoch": 0.2608068297961883,
      "flos": 17712615686400.0,
      "grad_norm": 2.5088389714281245,
      "language_loss": 0.80450433,
      "learning_rate": 3.4668856390803295e-06,
      "loss": 0.82655424,
      "num_input_tokens_seen": 46400025,
      "step": 2169,
      "time_per_iteration": 2.494565963745117
    },
    {
      "auxiliary_loss_clip": 0.01175939,
      "auxiliary_loss_mlp": 0.01037079,
      "balance_loss_clip": 1.05661988,
      "balance_loss_mlp": 1.02728021,
      "epoch": 0.2609270726868274,
      "flos": 18551632544640.0,
      "grad_norm": 1.8443108956758443,
      "language_loss": 0.89834225,
      "learning_rate": 3.4663560209553495e-06,
      "loss": 0.92047238,
      "num_input_tokens_seen": 46418090,
      "step": 2170,
      "time_per_iteration": 3.2861616611480713
    },
    {
      "auxiliary_loss_clip": 0.01164631,
      "auxiliary_loss_mlp": 0.01035064,
      "balance_loss_clip": 1.05582082,
      "balance_loss_mlp": 1.026088,
      "epoch": 0.26104731557746647,
      "flos": 21835699165440.0,
      "grad_norm": 1.623082782506732,
      "language_loss": 0.79230714,
      "learning_rate": 3.4658261803853267e-06,
      "loss": 0.81430411,
      "num_input_tokens_seen": 46436015,
      "step": 2171,
      "time_per_iteration": 2.548447847366333
    },
    {
      "auxiliary_loss_clip": 0.01169157,
      "auxiliary_loss_mlp": 0.01030095,
      "balance_loss_clip": 1.05906343,
      "balance_loss_mlp": 1.02044463,
      "epoch": 0.2611675584681056,
      "flos": 21689650465920.0,
      "grad_norm": 3.7303626463122104,
      "language_loss": 0.80973983,
      "learning_rate": 3.4652961174506383e-06,
      "loss": 0.83173233,
      "num_input_tokens_seen": 46455885,
      "step": 2172,
      "time_per_iteration": 2.554018259048462
    },
    {
      "auxiliary_loss_clip": 0.01073455,
      "auxiliary_loss_mlp": 0.01002989,
      "balance_loss_clip": 1.01766753,
      "balance_loss_mlp": 1.00090301,
      "epoch": 0.2612878013587447,
      "flos": 71862101389440.0,
      "grad_norm": 0.9765485020574719,
      "language_loss": 0.58150864,
      "learning_rate": 3.464765832231694e-06,
      "loss": 0.60227311,
      "num_input_tokens_seen": 46510050,
      "step": 2173,
      "time_per_iteration": 3.121633529663086
    },
    {
      "auxiliary_loss_clip": 0.01191774,
      "auxiliary_loss_mlp": 0.01036924,
      "balance_loss_clip": 1.06181264,
      "balance_loss_mlp": 1.02800119,
      "epoch": 0.26140804424938374,
      "flos": 20227511445120.0,
      "grad_norm": 3.7741997148089825,
      "language_loss": 0.70536321,
      "learning_rate": 3.4642353248089373e-06,
      "loss": 0.72765017,
      "num_input_tokens_seen": 46528810,
      "step": 2174,
      "time_per_iteration": 2.508643388748169
    },
    {
      "auxiliary_loss_clip": 0.01166922,
      "auxiliary_loss_mlp": 0.01033969,
      "balance_loss_clip": 1.05439675,
      "balance_loss_mlp": 1.02449787,
      "epoch": 0.26152828714002285,
      "flos": 25557085872000.0,
      "grad_norm": 2.240772286644814,
      "language_loss": 0.80300713,
      "learning_rate": 3.463704595262846e-06,
      "loss": 0.82501596,
      "num_input_tokens_seen": 46549690,
      "step": 2175,
      "time_per_iteration": 2.6237759590148926
    },
    {
      "auxiliary_loss_clip": 0.01156064,
      "auxiliary_loss_mlp": 0.01032682,
      "balance_loss_clip": 1.05527639,
      "balance_loss_mlp": 1.02354431,
      "epoch": 0.26164853003066196,
      "flos": 25446516831360.0,
      "grad_norm": 2.201421375841158,
      "language_loss": 0.70782,
      "learning_rate": 3.463173643673931e-06,
      "loss": 0.72970742,
      "num_input_tokens_seen": 46572215,
      "step": 2176,
      "time_per_iteration": 3.418454170227051
    },
    {
      "auxiliary_loss_clip": 0.010813,
      "auxiliary_loss_mlp": 0.01005249,
      "balance_loss_clip": 1.01859689,
      "balance_loss_mlp": 1.00312757,
      "epoch": 0.261768772921301,
      "flos": 53944580568960.0,
      "grad_norm": 0.8965113123203803,
      "language_loss": 0.63509458,
      "learning_rate": 3.4626424701227387e-06,
      "loss": 0.65596008,
      "num_input_tokens_seen": 46627275,
      "step": 2177,
      "time_per_iteration": 3.707408905029297
    },
    {
      "auxiliary_loss_clip": 0.01090076,
      "auxiliary_loss_mlp": 0.01004961,
      "balance_loss_clip": 1.01870632,
      "balance_loss_mlp": 1.00292289,
      "epoch": 0.26188901581194013,
      "flos": 70687606481280.0,
      "grad_norm": 0.8167687227736299,
      "language_loss": 0.55761981,
      "learning_rate": 3.4621110746898452e-06,
      "loss": 0.57857019,
      "num_input_tokens_seen": 46695135,
      "step": 2178,
      "time_per_iteration": 3.215350389480591
    },
    {
      "auxiliary_loss_clip": 0.0119003,
      "auxiliary_loss_mlp": 0.01034766,
      "balance_loss_clip": 1.05866778,
      "balance_loss_mlp": 1.02628374,
      "epoch": 0.2620092587025792,
      "flos": 21069580959360.0,
      "grad_norm": 1.80968952913865,
      "language_loss": 0.74861968,
      "learning_rate": 3.4615794574558654e-06,
      "loss": 0.77086765,
      "num_input_tokens_seen": 46714145,
      "step": 2179,
      "time_per_iteration": 3.482170820236206
    },
    {
      "auxiliary_loss_clip": 0.01174945,
      "auxiliary_loss_mlp": 0.01033176,
      "balance_loss_clip": 1.05689931,
      "balance_loss_mlp": 1.02428293,
      "epoch": 0.2621295015932183,
      "flos": 18369601395840.0,
      "grad_norm": 2.3522861750394815,
      "language_loss": 0.84010243,
      "learning_rate": 3.4610476185014436e-06,
      "loss": 0.86218363,
      "num_input_tokens_seen": 46731405,
      "step": 2180,
      "time_per_iteration": 2.518636703491211
    },
    {
      "auxiliary_loss_clip": 0.01204751,
      "auxiliary_loss_mlp": 0.01042658,
      "balance_loss_clip": 1.06116176,
      "balance_loss_mlp": 1.03303826,
      "epoch": 0.2622497444838574,
      "flos": 23659997063040.0,
      "grad_norm": 1.8094181141048817,
      "language_loss": 0.79075956,
      "learning_rate": 3.4605155579072597e-06,
      "loss": 0.81323361,
      "num_input_tokens_seen": 46751260,
      "step": 2181,
      "time_per_iteration": 2.5096688270568848
    },
    {
      "auxiliary_loss_clip": 0.01134629,
      "auxiliary_loss_mlp": 0.01033761,
      "balance_loss_clip": 1.04964185,
      "balance_loss_mlp": 1.02476668,
      "epoch": 0.26236998737449646,
      "flos": 22123810154880.0,
      "grad_norm": 1.8425704283932451,
      "language_loss": 0.70879638,
      "learning_rate": 3.459983275754027e-06,
      "loss": 0.73048031,
      "num_input_tokens_seen": 46770155,
      "step": 2182,
      "time_per_iteration": 2.6323044300079346
    },
    {
      "auxiliary_loss_clip": 0.0120241,
      "auxiliary_loss_mlp": 0.0103667,
      "balance_loss_clip": 1.05937922,
      "balance_loss_mlp": 1.02758002,
      "epoch": 0.26249023026513557,
      "flos": 17895185539200.0,
      "grad_norm": 2.1592073188038134,
      "language_loss": 0.79753935,
      "learning_rate": 3.4594507721224918e-06,
      "loss": 0.8199302,
      "num_input_tokens_seen": 46788805,
      "step": 2183,
      "time_per_iteration": 2.485175609588623
    },
    {
      "auxiliary_loss_clip": 0.01175287,
      "auxiliary_loss_mlp": 0.01036619,
      "balance_loss_clip": 1.05538905,
      "balance_loss_mlp": 1.02816164,
      "epoch": 0.2626104731557747,
      "flos": 18332936588160.0,
      "grad_norm": 1.6267557995845998,
      "language_loss": 0.81657732,
      "learning_rate": 3.4589180470934353e-06,
      "loss": 0.83869636,
      "num_input_tokens_seen": 46808670,
      "step": 2184,
      "time_per_iteration": 2.5769431591033936
    },
    {
      "auxiliary_loss_clip": 0.01192703,
      "auxiliary_loss_mlp": 0.01032175,
      "balance_loss_clip": 1.05643451,
      "balance_loss_mlp": 1.02199447,
      "epoch": 0.26273071604641374,
      "flos": 19317714837120.0,
      "grad_norm": 1.778481225118753,
      "language_loss": 0.76767683,
      "learning_rate": 3.4583851007476713e-06,
      "loss": 0.78992563,
      "num_input_tokens_seen": 46827140,
      "step": 2185,
      "time_per_iteration": 2.503451108932495
    },
    {
      "auxiliary_loss_clip": 0.01161679,
      "auxiliary_loss_mlp": 0.01040064,
      "balance_loss_clip": 1.05686283,
      "balance_loss_mlp": 1.03090835,
      "epoch": 0.26285095893705285,
      "flos": 18327477720960.0,
      "grad_norm": 2.2549744231999247,
      "language_loss": 0.68717819,
      "learning_rate": 3.4578519331660464e-06,
      "loss": 0.70919561,
      "num_input_tokens_seen": 46844135,
      "step": 2186,
      "time_per_iteration": 2.556922674179077
    },
    {
      "auxiliary_loss_clip": 0.01184793,
      "auxiliary_loss_mlp": 0.01031914,
      "balance_loss_clip": 1.06004071,
      "balance_loss_mlp": 1.02288401,
      "epoch": 0.26297120182769196,
      "flos": 20193827466240.0,
      "grad_norm": 2.1806014955942987,
      "language_loss": 0.81868315,
      "learning_rate": 3.4573185444294426e-06,
      "loss": 0.84085023,
      "num_input_tokens_seen": 46862500,
      "step": 2187,
      "time_per_iteration": 2.482633113861084
    },
    {
      "auxiliary_loss_clip": 0.01175543,
      "auxiliary_loss_mlp": 0.00762761,
      "balance_loss_clip": 1.05780888,
      "balance_loss_mlp": 1.00025761,
      "epoch": 0.263091444718331,
      "flos": 22418421505920.0,
      "grad_norm": 1.7147972737176664,
      "language_loss": 0.78867865,
      "learning_rate": 3.456784934618774e-06,
      "loss": 0.80806172,
      "num_input_tokens_seen": 46883665,
      "step": 2188,
      "time_per_iteration": 2.550771713256836
    },
    {
      "auxiliary_loss_clip": 0.01168735,
      "auxiliary_loss_mlp": 0.01034199,
      "balance_loss_clip": 1.05375648,
      "balance_loss_mlp": 1.02525806,
      "epoch": 0.2632116876089701,
      "flos": 19024827338880.0,
      "grad_norm": 2.1884027256068688,
      "language_loss": 0.80210429,
      "learning_rate": 3.4562511038149897e-06,
      "loss": 0.82413363,
      "num_input_tokens_seen": 46899160,
      "step": 2189,
      "time_per_iteration": 2.5027947425842285
    },
    {
      "auxiliary_loss_clip": 0.01042226,
      "auxiliary_loss_mlp": 0.01009366,
      "balance_loss_clip": 1.01682734,
      "balance_loss_mlp": 1.0073632,
      "epoch": 0.26333193049960923,
      "flos": 67308054531840.0,
      "grad_norm": 0.9765564595558546,
      "language_loss": 0.57744515,
      "learning_rate": 3.4557170520990705e-06,
      "loss": 0.59796107,
      "num_input_tokens_seen": 46959835,
      "step": 2190,
      "time_per_iteration": 3.217667818069458
    },
    {
      "auxiliary_loss_clip": 0.01184914,
      "auxiliary_loss_mlp": 0.01033823,
      "balance_loss_clip": 1.05888677,
      "balance_loss_mlp": 1.02524626,
      "epoch": 0.2634521733902483,
      "flos": 25048806468480.0,
      "grad_norm": 1.4555542264737897,
      "language_loss": 0.8652705,
      "learning_rate": 3.4551827795520324e-06,
      "loss": 0.88745791,
      "num_input_tokens_seen": 46982720,
      "step": 2191,
      "time_per_iteration": 2.5665106773376465
    },
    {
      "auxiliary_loss_clip": 0.01185921,
      "auxiliary_loss_mlp": 0.01035909,
      "balance_loss_clip": 1.055668,
      "balance_loss_mlp": 1.02704597,
      "epoch": 0.2635724162808874,
      "flos": 20594985534720.0,
      "grad_norm": 1.759495401317626,
      "language_loss": 0.84780812,
      "learning_rate": 3.4546482862549226e-06,
      "loss": 0.87002647,
      "num_input_tokens_seen": 47003035,
      "step": 2192,
      "time_per_iteration": 2.511434555053711
    },
    {
      "auxiliary_loss_clip": 0.0115198,
      "auxiliary_loss_mlp": 0.01030346,
      "balance_loss_clip": 1.05241954,
      "balance_loss_mlp": 1.02182293,
      "epoch": 0.2636926591715265,
      "flos": 19244636616960.0,
      "grad_norm": 2.1794616290049347,
      "language_loss": 0.78806412,
      "learning_rate": 3.4541135722888253e-06,
      "loss": 0.80988735,
      "num_input_tokens_seen": 47019625,
      "step": 2193,
      "time_per_iteration": 2.5696299076080322
    },
    {
      "auxiliary_loss_clip": 0.01200368,
      "auxiliary_loss_mlp": 0.01029409,
      "balance_loss_clip": 1.05842614,
      "balance_loss_mlp": 1.02102304,
      "epoch": 0.26381290206216557,
      "flos": 28804882734720.0,
      "grad_norm": 1.6539957983333218,
      "language_loss": 0.80064142,
      "learning_rate": 3.453578637734854e-06,
      "loss": 0.82293916,
      "num_input_tokens_seen": 47040815,
      "step": 2194,
      "time_per_iteration": 2.5336151123046875
    },
    {
      "auxiliary_loss_clip": 0.01204618,
      "auxiliary_loss_mlp": 0.01037707,
      "balance_loss_clip": 1.06360078,
      "balance_loss_mlp": 1.02904081,
      "epoch": 0.2639331449528047,
      "flos": 25008909436800.0,
      "grad_norm": 1.7393670323153714,
      "language_loss": 0.78529447,
      "learning_rate": 3.4530434826741605e-06,
      "loss": 0.80771774,
      "num_input_tokens_seen": 47061755,
      "step": 2195,
      "time_per_iteration": 2.51389217376709
    },
    {
      "auxiliary_loss_clip": 0.01168917,
      "auxiliary_loss_mlp": 0.01031761,
      "balance_loss_clip": 1.05755568,
      "balance_loss_mlp": 1.02373266,
      "epoch": 0.26405338784344373,
      "flos": 46535775465600.0,
      "grad_norm": 1.61060983508782,
      "language_loss": 0.68563664,
      "learning_rate": 3.452508107187926e-06,
      "loss": 0.70764345,
      "num_input_tokens_seen": 47085130,
      "step": 2196,
      "time_per_iteration": 3.461779832839966
    },
    {
      "auxiliary_loss_clip": 0.01127693,
      "auxiliary_loss_mlp": 0.01037115,
      "balance_loss_clip": 1.04775918,
      "balance_loss_mlp": 1.02770305,
      "epoch": 0.26417363073408284,
      "flos": 21179467641600.0,
      "grad_norm": 1.9060793781494343,
      "language_loss": 0.7721808,
      "learning_rate": 3.451972511357366e-06,
      "loss": 0.79382885,
      "num_input_tokens_seen": 47104675,
      "step": 2197,
      "time_per_iteration": 2.637789487838745
    },
    {
      "auxiliary_loss_clip": 0.01181461,
      "auxiliary_loss_mlp": 0.01031407,
      "balance_loss_clip": 1.05806315,
      "balance_loss_mlp": 1.0226872,
      "epoch": 0.26429387362472195,
      "flos": 22674751937280.0,
      "grad_norm": 1.7579904093483376,
      "language_loss": 0.85398436,
      "learning_rate": 3.45143669526373e-06,
      "loss": 0.87611306,
      "num_input_tokens_seen": 47124435,
      "step": 2198,
      "time_per_iteration": 2.522179126739502
    },
    {
      "auxiliary_loss_clip": 0.0106985,
      "auxiliary_loss_mlp": 0.01002121,
      "balance_loss_clip": 1.01727867,
      "balance_loss_mlp": 1.00005913,
      "epoch": 0.264414116515361,
      "flos": 67180534272000.0,
      "grad_norm": 0.8164346032526584,
      "language_loss": 0.63189256,
      "learning_rate": 3.450900658988302e-06,
      "loss": 0.65261221,
      "num_input_tokens_seen": 47185985,
      "step": 2199,
      "time_per_iteration": 3.071333885192871
    },
    {
      "auxiliary_loss_clip": 0.01163374,
      "auxiliary_loss_mlp": 0.01035219,
      "balance_loss_clip": 1.05589855,
      "balance_loss_mlp": 1.02564692,
      "epoch": 0.2645343594060001,
      "flos": 25664709997440.0,
      "grad_norm": 1.9279970129618336,
      "language_loss": 0.77786541,
      "learning_rate": 3.450364402612397e-06,
      "loss": 0.79985136,
      "num_input_tokens_seen": 47203140,
      "step": 2200,
      "time_per_iteration": 2.57719349861145
    },
    {
      "auxiliary_loss_clip": 0.011683,
      "auxiliary_loss_mlp": 0.01038853,
      "balance_loss_clip": 1.05233419,
      "balance_loss_mlp": 1.03029335,
      "epoch": 0.26465460229663923,
      "flos": 22491822948480.0,
      "grad_norm": 1.875680228160289,
      "language_loss": 0.83933878,
      "learning_rate": 3.449827926217366e-06,
      "loss": 0.86141026,
      "num_input_tokens_seen": 47222575,
      "step": 2201,
      "time_per_iteration": 2.551110029220581
    },
    {
      "auxiliary_loss_clip": 0.0117148,
      "auxiliary_loss_mlp": 0.01032795,
      "balance_loss_clip": 1.05066216,
      "balance_loss_mlp": 1.02430153,
      "epoch": 0.2647748451872783,
      "flos": 29388036038400.0,
      "grad_norm": 3.563511531904457,
      "language_loss": 0.8042897,
      "learning_rate": 3.449291229884591e-06,
      "loss": 0.82633245,
      "num_input_tokens_seen": 47243815,
      "step": 2202,
      "time_per_iteration": 3.3992879390716553
    },
    {
      "auxiliary_loss_clip": 0.01159255,
      "auxiliary_loss_mlp": 0.01035775,
      "balance_loss_clip": 1.05328548,
      "balance_loss_mlp": 1.02681684,
      "epoch": 0.2648950880779174,
      "flos": 26797799502720.0,
      "grad_norm": 1.7124917921102993,
      "language_loss": 0.86763954,
      "learning_rate": 3.4487543136954887e-06,
      "loss": 0.88958985,
      "num_input_tokens_seen": 47263435,
      "step": 2203,
      "time_per_iteration": 3.268906354904175
    },
    {
      "auxiliary_loss_clip": 0.01155705,
      "auxiliary_loss_mlp": 0.01033708,
      "balance_loss_clip": 1.0537703,
      "balance_loss_mlp": 1.02491069,
      "epoch": 0.2650153309685565,
      "flos": 28841008838400.0,
      "grad_norm": 1.7005070816022452,
      "language_loss": 0.91251469,
      "learning_rate": 3.448217177731509e-06,
      "loss": 0.93440878,
      "num_input_tokens_seen": 47283920,
      "step": 2204,
      "time_per_iteration": 2.61950945854187
    },
    {
      "auxiliary_loss_clip": 0.01166272,
      "auxiliary_loss_mlp": 0.0103842,
      "balance_loss_clip": 1.05586767,
      "balance_loss_mlp": 1.02946115,
      "epoch": 0.26513557385919556,
      "flos": 20303247271680.0,
      "grad_norm": 2.7780153602186917,
      "language_loss": 0.77702719,
      "learning_rate": 3.4476798220741348e-06,
      "loss": 0.79907411,
      "num_input_tokens_seen": 47302800,
      "step": 2205,
      "time_per_iteration": 3.3384013175964355
    },
    {
      "auxiliary_loss_clip": 0.01202609,
      "auxiliary_loss_mlp": 0.01033691,
      "balance_loss_clip": 1.06305718,
      "balance_loss_mlp": 1.02554309,
      "epoch": 0.26525581674983467,
      "flos": 17676274101120.0,
      "grad_norm": 1.7720048109130149,
      "language_loss": 0.78223443,
      "learning_rate": 3.4471422468048826e-06,
      "loss": 0.8045975,
      "num_input_tokens_seen": 47321525,
      "step": 2206,
      "time_per_iteration": 2.451143264770508
    },
    {
      "auxiliary_loss_clip": 0.01177753,
      "auxiliary_loss_mlp": 0.01034461,
      "balance_loss_clip": 1.05757999,
      "balance_loss_mlp": 1.02479911,
      "epoch": 0.2653760596404738,
      "flos": 26833746038400.0,
      "grad_norm": 2.5366575595775043,
      "language_loss": 0.73327482,
      "learning_rate": 3.4466044520053022e-06,
      "loss": 0.75539696,
      "num_input_tokens_seen": 47340530,
      "step": 2207,
      "time_per_iteration": 2.539808511734009
    },
    {
      "auxiliary_loss_clip": 0.01156834,
      "auxiliary_loss_mlp": 0.01037486,
      "balance_loss_clip": 1.05093551,
      "balance_loss_mlp": 1.02886701,
      "epoch": 0.26549630253111284,
      "flos": 22782160581120.0,
      "grad_norm": 1.7545379625084028,
      "language_loss": 0.60538185,
      "learning_rate": 3.446066437756977e-06,
      "loss": 0.627325,
      "num_input_tokens_seen": 47359735,
      "step": 2208,
      "time_per_iteration": 2.5168755054473877
    },
    {
      "auxiliary_loss_clip": 0.01170691,
      "auxiliary_loss_mlp": 0.01027802,
      "balance_loss_clip": 1.05587602,
      "balance_loss_mlp": 1.01873577,
      "epoch": 0.26561654542175195,
      "flos": 23550002640000.0,
      "grad_norm": 2.279722453515776,
      "language_loss": 0.75347096,
      "learning_rate": 3.4455282041415224e-06,
      "loss": 0.77545589,
      "num_input_tokens_seen": 47378945,
      "step": 2209,
      "time_per_iteration": 2.5534069538116455
    },
    {
      "auxiliary_loss_clip": 0.01155693,
      "auxiliary_loss_mlp": 0.01034549,
      "balance_loss_clip": 1.05290401,
      "balance_loss_mlp": 1.02581668,
      "epoch": 0.265736788312391,
      "flos": 26906680604160.0,
      "grad_norm": 2.0280871493829205,
      "language_loss": 0.87160403,
      "learning_rate": 3.4449897512405894e-06,
      "loss": 0.89350647,
      "num_input_tokens_seen": 47398095,
      "step": 2210,
      "time_per_iteration": 2.598217725753784
    },
    {
      "auxiliary_loss_clip": 0.01120118,
      "auxiliary_loss_mlp": 0.00762592,
      "balance_loss_clip": 1.04640937,
      "balance_loss_mlp": 1.00027514,
      "epoch": 0.2658570312030301,
      "flos": 23477139901440.0,
      "grad_norm": 2.004827955253511,
      "language_loss": 0.75474727,
      "learning_rate": 3.444451079135859e-06,
      "loss": 0.77357441,
      "num_input_tokens_seen": 47417605,
      "step": 2211,
      "time_per_iteration": 2.6826331615448
    },
    {
      "auxiliary_loss_clip": 0.01134073,
      "auxiliary_loss_mlp": 0.00762596,
      "balance_loss_clip": 1.05016208,
      "balance_loss_mlp": 1.00026143,
      "epoch": 0.2659772740936692,
      "flos": 21866402315520.0,
      "grad_norm": 1.8568523219838553,
      "language_loss": 0.74154055,
      "learning_rate": 3.4439121879090493e-06,
      "loss": 0.76050729,
      "num_input_tokens_seen": 47435385,
      "step": 2212,
      "time_per_iteration": 2.6048130989074707
    },
    {
      "auxiliary_loss_clip": 0.01175681,
      "auxiliary_loss_mlp": 0.01035062,
      "balance_loss_clip": 1.05619442,
      "balance_loss_mlp": 1.02624631,
      "epoch": 0.2660975169843083,
      "flos": 19793100360960.0,
      "grad_norm": 2.072414500629902,
      "language_loss": 0.83193982,
      "learning_rate": 3.4433730776419082e-06,
      "loss": 0.85404718,
      "num_input_tokens_seen": 47454310,
      "step": 2213,
      "time_per_iteration": 2.559068202972412
    },
    {
      "auxiliary_loss_clip": 0.01188343,
      "auxiliary_loss_mlp": 0.00763351,
      "balance_loss_clip": 1.05639458,
      "balance_loss_mlp": 1.00027752,
      "epoch": 0.2662177598749474,
      "flos": 29018981750400.0,
      "grad_norm": 2.1545023116287756,
      "language_loss": 0.79959881,
      "learning_rate": 3.4428337484162183e-06,
      "loss": 0.81911576,
      "num_input_tokens_seen": 47475120,
      "step": 2214,
      "time_per_iteration": 2.5759618282318115
    },
    {
      "auxiliary_loss_clip": 0.01165226,
      "auxiliary_loss_mlp": 0.01036215,
      "balance_loss_clip": 1.05147755,
      "balance_loss_mlp": 1.02721488,
      "epoch": 0.2663380027655865,
      "flos": 21762549118080.0,
      "grad_norm": 1.726396583592112,
      "language_loss": 0.84172571,
      "learning_rate": 3.442294200313797e-06,
      "loss": 0.86374009,
      "num_input_tokens_seen": 47493150,
      "step": 2215,
      "time_per_iteration": 2.5511763095855713
    },
    {
      "auxiliary_loss_clip": 0.0109063,
      "auxiliary_loss_mlp": 0.0100209,
      "balance_loss_clip": 1.0204258,
      "balance_loss_mlp": 1.0001353,
      "epoch": 0.26645824565622556,
      "flos": 66980333819520.0,
      "grad_norm": 0.7662825368767422,
      "language_loss": 0.52735877,
      "learning_rate": 3.4417544334164916e-06,
      "loss": 0.54828596,
      "num_input_tokens_seen": 47557295,
      "step": 2216,
      "time_per_iteration": 3.1320981979370117
    },
    {
      "auxiliary_loss_clip": 0.01155642,
      "auxiliary_loss_mlp": 0.0103172,
      "balance_loss_clip": 1.05660963,
      "balance_loss_mlp": 1.02309561,
      "epoch": 0.26657848854686467,
      "flos": 25264198373760.0,
      "grad_norm": 1.7488988716850495,
      "language_loss": 0.77387643,
      "learning_rate": 3.4412144478061854e-06,
      "loss": 0.79575014,
      "num_input_tokens_seen": 47579705,
      "step": 2217,
      "time_per_iteration": 2.5900399684906006
    },
    {
      "auxiliary_loss_clip": 0.01097053,
      "auxiliary_loss_mlp": 0.01037391,
      "balance_loss_clip": 1.04836535,
      "balance_loss_mlp": 1.02795517,
      "epoch": 0.2666987314375038,
      "flos": 23696769611520.0,
      "grad_norm": 1.846033361114347,
      "language_loss": 0.75477493,
      "learning_rate": 3.4406742435647925e-06,
      "loss": 0.77611935,
      "num_input_tokens_seen": 47599770,
      "step": 2218,
      "time_per_iteration": 2.7240214347839355
    },
    {
      "auxiliary_loss_clip": 0.01185377,
      "auxiliary_loss_mlp": 0.01030184,
      "balance_loss_clip": 1.0600822,
      "balance_loss_mlp": 1.02222633,
      "epoch": 0.26681897432814283,
      "flos": 27048958375680.0,
      "grad_norm": 2.095684102872311,
      "language_loss": 0.79295325,
      "learning_rate": 3.440133820774263e-06,
      "loss": 0.8151089,
      "num_input_tokens_seen": 47619580,
      "step": 2219,
      "time_per_iteration": 2.5479087829589844
    },
    {
      "auxiliary_loss_clip": 0.01176551,
      "auxiliary_loss_mlp": 0.01037219,
      "balance_loss_clip": 1.05910158,
      "balance_loss_mlp": 1.02829003,
      "epoch": 0.26693921721878194,
      "flos": 28985944216320.0,
      "grad_norm": 3.523812715907441,
      "language_loss": 0.81593776,
      "learning_rate": 3.439593179516578e-06,
      "loss": 0.83807552,
      "num_input_tokens_seen": 47639490,
      "step": 2220,
      "time_per_iteration": 2.6019537448883057
    },
    {
      "auxiliary_loss_clip": 0.01176928,
      "auxiliary_loss_mlp": 0.01035626,
      "balance_loss_clip": 1.05759001,
      "balance_loss_mlp": 1.0264585,
      "epoch": 0.26705946010942105,
      "flos": 21507834798720.0,
      "grad_norm": 2.1459535063176416,
      "language_loss": 0.81082463,
      "learning_rate": 3.4390523198737524e-06,
      "loss": 0.83295017,
      "num_input_tokens_seen": 47658650,
      "step": 2221,
      "time_per_iteration": 2.5908679962158203
    },
    {
      "auxiliary_loss_clip": 0.01202081,
      "auxiliary_loss_mlp": 0.00762496,
      "balance_loss_clip": 1.05988002,
      "balance_loss_mlp": 1.00026941,
      "epoch": 0.2671797030000601,
      "flos": 21471277731840.0,
      "grad_norm": 2.7338498093676127,
      "language_loss": 0.73497987,
      "learning_rate": 3.4385112419278333e-06,
      "loss": 0.75462568,
      "num_input_tokens_seen": 47679875,
      "step": 2222,
      "time_per_iteration": 3.2499656677246094
    },
    {
      "auxiliary_loss_clip": 0.01080318,
      "auxiliary_loss_mlp": 0.01004256,
      "balance_loss_clip": 1.01966047,
      "balance_loss_mlp": 1.00222993,
      "epoch": 0.2672999458906992,
      "flos": 64189929767040.0,
      "grad_norm": 0.7942183606681902,
      "language_loss": 0.64862704,
      "learning_rate": 3.4379699457609033e-06,
      "loss": 0.66947275,
      "num_input_tokens_seen": 47737700,
      "step": 2223,
      "time_per_iteration": 2.9545304775238037
    },
    {
      "auxiliary_loss_clip": 0.01162159,
      "auxiliary_loss_mlp": 0.0103281,
      "balance_loss_clip": 1.05257583,
      "balance_loss_mlp": 1.02338028,
      "epoch": 0.26742018878133833,
      "flos": 16909042573440.0,
      "grad_norm": 1.7933627484966441,
      "language_loss": 0.90281963,
      "learning_rate": 3.4374284314550755e-06,
      "loss": 0.92476928,
      "num_input_tokens_seen": 47756740,
      "step": 2224,
      "time_per_iteration": 2.539372444152832
    },
    {
      "auxiliary_loss_clip": 0.01201072,
      "auxiliary_loss_mlp": 0.01036606,
      "balance_loss_clip": 1.06068206,
      "balance_loss_mlp": 1.02777791,
      "epoch": 0.2675404316719774,
      "flos": 20667560964480.0,
      "grad_norm": 1.8478705746315853,
      "language_loss": 0.81301486,
      "learning_rate": 3.436886699092498e-06,
      "loss": 0.83539164,
      "num_input_tokens_seen": 47775255,
      "step": 2225,
      "time_per_iteration": 2.4700374603271484
    },
    {
      "auxiliary_loss_clip": 0.01204834,
      "auxiliary_loss_mlp": 0.01041698,
      "balance_loss_clip": 1.06108308,
      "balance_loss_mlp": 1.03252482,
      "epoch": 0.2676606745626165,
      "flos": 17485013157120.0,
      "grad_norm": 7.787948792245968,
      "language_loss": 0.7133162,
      "learning_rate": 3.4363447487553502e-06,
      "loss": 0.73578155,
      "num_input_tokens_seen": 47788570,
      "step": 2226,
      "time_per_iteration": 2.432028293609619
    },
    {
      "auxiliary_loss_clip": 0.01168204,
      "auxiliary_loss_mlp": 0.0104118,
      "balance_loss_clip": 1.05713797,
      "balance_loss_mlp": 1.0320133,
      "epoch": 0.26778091745325555,
      "flos": 27852675143040.0,
      "grad_norm": 1.706422194811565,
      "language_loss": 0.78235567,
      "learning_rate": 3.4358025805258455e-06,
      "loss": 0.8044495,
      "num_input_tokens_seen": 47808275,
      "step": 2227,
      "time_per_iteration": 2.576878070831299
    },
    {
      "auxiliary_loss_clip": 0.01146194,
      "auxiliary_loss_mlp": 0.01030478,
      "balance_loss_clip": 1.05208302,
      "balance_loss_mlp": 1.02156126,
      "epoch": 0.26790116034389466,
      "flos": 20955995176320.0,
      "grad_norm": 2.512918619935285,
      "language_loss": 0.83910155,
      "learning_rate": 3.435260194486232e-06,
      "loss": 0.86086828,
      "num_input_tokens_seen": 47826245,
      "step": 2228,
      "time_per_iteration": 3.3695085048675537
    },
    {
      "auxiliary_loss_clip": 0.01171737,
      "auxiliary_loss_mlp": 0.01031217,
      "balance_loss_clip": 1.05812919,
      "balance_loss_mlp": 1.02218127,
      "epoch": 0.2680214032345338,
      "flos": 18040659621120.0,
      "grad_norm": 2.103098614691949,
      "language_loss": 0.82344294,
      "learning_rate": 3.4347175907187875e-06,
      "loss": 0.84547246,
      "num_input_tokens_seen": 47843235,
      "step": 2229,
      "time_per_iteration": 3.245657205581665
    },
    {
      "auxiliary_loss_clip": 0.01185765,
      "auxiliary_loss_mlp": 0.01036729,
      "balance_loss_clip": 1.05919456,
      "balance_loss_mlp": 1.02764511,
      "epoch": 0.26814164612517283,
      "flos": 22419427086720.0,
      "grad_norm": 1.9127866493973917,
      "language_loss": 0.87940794,
      "learning_rate": 3.4341747693058254e-06,
      "loss": 0.90163291,
      "num_input_tokens_seen": 47861710,
      "step": 2230,
      "time_per_iteration": 3.347287178039551
    },
    {
      "auxiliary_loss_clip": 0.01090477,
      "auxiliary_loss_mlp": 0.01035548,
      "balance_loss_clip": 1.04763794,
      "balance_loss_mlp": 1.02666128,
      "epoch": 0.26826188901581194,
      "flos": 35627371159680.0,
      "grad_norm": 1.9519237522537427,
      "language_loss": 0.7702868,
      "learning_rate": 3.4336317303296916e-06,
      "loss": 0.79154706,
      "num_input_tokens_seen": 47882685,
      "step": 2231,
      "time_per_iteration": 2.8360073566436768
    },
    {
      "auxiliary_loss_clip": 0.01182477,
      "auxiliary_loss_mlp": 0.01033374,
      "balance_loss_clip": 1.05707824,
      "balance_loss_mlp": 1.02437949,
      "epoch": 0.26838213190645105,
      "flos": 17639788861440.0,
      "grad_norm": 2.674670987534941,
      "language_loss": 0.75453848,
      "learning_rate": 3.4330884738727635e-06,
      "loss": 0.77669698,
      "num_input_tokens_seen": 47900860,
      "step": 2232,
      "time_per_iteration": 2.686034679412842
    },
    {
      "auxiliary_loss_clip": 0.01134557,
      "auxiliary_loss_mlp": 0.01035486,
      "balance_loss_clip": 1.05386961,
      "balance_loss_mlp": 1.0264914,
      "epoch": 0.2685023747970901,
      "flos": 22674823764480.0,
      "grad_norm": 1.8276510882690735,
      "language_loss": 0.70535713,
      "learning_rate": 3.4325450000174535e-06,
      "loss": 0.72705758,
      "num_input_tokens_seen": 47917500,
      "step": 2233,
      "time_per_iteration": 2.592045783996582
    },
    {
      "auxiliary_loss_clip": 0.01133252,
      "auxiliary_loss_mlp": 0.01035786,
      "balance_loss_clip": 1.05029309,
      "balance_loss_mlp": 1.02656555,
      "epoch": 0.2686226176877292,
      "flos": 20120533764480.0,
      "grad_norm": 1.6634343017392166,
      "language_loss": 0.73888707,
      "learning_rate": 3.4320013088462067e-06,
      "loss": 0.76057744,
      "num_input_tokens_seen": 47934860,
      "step": 2234,
      "time_per_iteration": 2.592363119125366
    },
    {
      "auxiliary_loss_clip": 0.01158895,
      "auxiliary_loss_mlp": 0.01034725,
      "balance_loss_clip": 1.05302835,
      "balance_loss_mlp": 1.02574217,
      "epoch": 0.2687428605783683,
      "flos": 21872040750720.0,
      "grad_norm": 1.436555645751939,
      "language_loss": 0.81479031,
      "learning_rate": 3.431457400441499e-06,
      "loss": 0.83672649,
      "num_input_tokens_seen": 47955255,
      "step": 2235,
      "time_per_iteration": 2.5742905139923096
    },
    {
      "auxiliary_loss_clip": 0.01020247,
      "auxiliary_loss_mlp": 0.01002051,
      "balance_loss_clip": 1.01335192,
      "balance_loss_mlp": 0.99990541,
      "epoch": 0.2688631034690074,
      "flos": 69943320766080.0,
      "grad_norm": 0.9227296692943235,
      "language_loss": 0.6093812,
      "learning_rate": 3.4309132748858424e-06,
      "loss": 0.62960422,
      "num_input_tokens_seen": 48016245,
      "step": 2236,
      "time_per_iteration": 3.2282798290252686
    },
    {
      "auxiliary_loss_clip": 0.0118394,
      "auxiliary_loss_mlp": 0.01034962,
      "balance_loss_clip": 1.05926514,
      "balance_loss_mlp": 1.02645683,
      "epoch": 0.2689833463596465,
      "flos": 22856639431680.0,
      "grad_norm": 1.6546448491307526,
      "language_loss": 0.83487093,
      "learning_rate": 3.430368932261779e-06,
      "loss": 0.85705996,
      "num_input_tokens_seen": 48036600,
      "step": 2237,
      "time_per_iteration": 2.540611743927002
    },
    {
      "auxiliary_loss_clip": 0.01170503,
      "auxiliary_loss_mlp": 0.01033786,
      "balance_loss_clip": 1.05815756,
      "balance_loss_mlp": 1.02480936,
      "epoch": 0.2691035892502856,
      "flos": 17200242132480.0,
      "grad_norm": 1.8312946505086265,
      "language_loss": 0.75157845,
      "learning_rate": 3.429824372651886e-06,
      "loss": 0.77362132,
      "num_input_tokens_seen": 48054750,
      "step": 2238,
      "time_per_iteration": 2.505087375640869
    },
    {
      "auxiliary_loss_clip": 0.01152107,
      "auxiliary_loss_mlp": 0.0102688,
      "balance_loss_clip": 1.05426788,
      "balance_loss_mlp": 1.01778483,
      "epoch": 0.26922383214092466,
      "flos": 17747484814080.0,
      "grad_norm": 2.878217044745344,
      "language_loss": 0.83077705,
      "learning_rate": 3.4292795961387732e-06,
      "loss": 0.8525669,
      "num_input_tokens_seen": 48072650,
      "step": 2239,
      "time_per_iteration": 2.6017048358917236
    },
    {
      "auxiliary_loss_clip": 0.01198014,
      "auxiliary_loss_mlp": 0.0103392,
      "balance_loss_clip": 1.05739641,
      "balance_loss_mlp": 1.0250268,
      "epoch": 0.26934407503156377,
      "flos": 16173376122240.0,
      "grad_norm": 2.3134215468665142,
      "language_loss": 0.87339002,
      "learning_rate": 3.4287346028050818e-06,
      "loss": 0.8957094,
      "num_input_tokens_seen": 48088720,
      "step": 2240,
      "time_per_iteration": 2.4366824626922607
    },
    {
      "auxiliary_loss_clip": 0.01172902,
      "auxiliary_loss_mlp": 0.01036305,
      "balance_loss_clip": 1.05716515,
      "balance_loss_mlp": 1.02790058,
      "epoch": 0.2694643179222028,
      "flos": 23732895715200.0,
      "grad_norm": 1.4729490659547153,
      "language_loss": 0.79646218,
      "learning_rate": 3.4281893927334866e-06,
      "loss": 0.81855428,
      "num_input_tokens_seen": 48108630,
      "step": 2241,
      "time_per_iteration": 2.5557057857513428
    },
    {
      "auxiliary_loss_clip": 0.01187268,
      "auxiliary_loss_mlp": 0.0103733,
      "balance_loss_clip": 1.05938363,
      "balance_loss_mlp": 1.02850246,
      "epoch": 0.26958456081284193,
      "flos": 24718140840960.0,
      "grad_norm": 1.9532074946406646,
      "language_loss": 0.75155002,
      "learning_rate": 3.4276439660066963e-06,
      "loss": 0.77379596,
      "num_input_tokens_seen": 48128330,
      "step": 2242,
      "time_per_iteration": 2.5369086265563965
    },
    {
      "auxiliary_loss_clip": 0.01196002,
      "auxiliary_loss_mlp": 0.01032312,
      "balance_loss_clip": 1.05810678,
      "balance_loss_mlp": 1.02357364,
      "epoch": 0.26970480370348104,
      "flos": 18112588606080.0,
      "grad_norm": 2.6956873230163003,
      "language_loss": 0.84165609,
      "learning_rate": 3.427098322707452e-06,
      "loss": 0.86393923,
      "num_input_tokens_seen": 48144295,
      "step": 2243,
      "time_per_iteration": 2.4258689880371094
    },
    {
      "auxiliary_loss_clip": 0.01184912,
      "auxiliary_loss_mlp": 0.01038653,
      "balance_loss_clip": 1.06041396,
      "balance_loss_mlp": 1.02965844,
      "epoch": 0.2698250465941201,
      "flos": 10816546250880.0,
      "grad_norm": 1.9891821945076495,
      "language_loss": 0.89422244,
      "learning_rate": 3.426552462918526e-06,
      "loss": 0.91645813,
      "num_input_tokens_seen": 48162230,
      "step": 2244,
      "time_per_iteration": 2.467433214187622
    },
    {
      "auxiliary_loss_clip": 0.01202326,
      "auxiliary_loss_mlp": 0.01038708,
      "balance_loss_clip": 1.06314254,
      "balance_loss_mlp": 1.02991068,
      "epoch": 0.2699452894847592,
      "flos": 17308117653120.0,
      "grad_norm": 2.320289609288705,
      "language_loss": 0.73065478,
      "learning_rate": 3.426006386722726e-06,
      "loss": 0.75306511,
      "num_input_tokens_seen": 48180290,
      "step": 2245,
      "time_per_iteration": 2.4437308311462402
    },
    {
      "auxiliary_loss_clip": 0.01159172,
      "auxiliary_loss_mlp": 0.01031721,
      "balance_loss_clip": 1.0578295,
      "balance_loss_mlp": 1.02324522,
      "epoch": 0.2700655323753983,
      "flos": 18078150441600.0,
      "grad_norm": 1.9242404232551598,
      "language_loss": 0.92233354,
      "learning_rate": 3.4254600942028914e-06,
      "loss": 0.94424254,
      "num_input_tokens_seen": 48198165,
      "step": 2246,
      "time_per_iteration": 2.563342809677124
    },
    {
      "auxiliary_loss_clip": 0.01166524,
      "auxiliary_loss_mlp": 0.01028229,
      "balance_loss_clip": 1.05758905,
      "balance_loss_mlp": 1.0200932,
      "epoch": 0.2701857752660374,
      "flos": 18186636493440.0,
      "grad_norm": 2.1349041809808127,
      "language_loss": 0.82637435,
      "learning_rate": 3.424913585441893e-06,
      "loss": 0.84832191,
      "num_input_tokens_seen": 48216000,
      "step": 2247,
      "time_per_iteration": 2.49953293800354
    },
    {
      "auxiliary_loss_clip": 0.01181778,
      "auxiliary_loss_mlp": 0.01039228,
      "balance_loss_clip": 1.05804133,
      "balance_loss_mlp": 1.0301739,
      "epoch": 0.2703060181566765,
      "flos": 16319496648960.0,
      "grad_norm": 1.8711787808860212,
      "language_loss": 0.87162197,
      "learning_rate": 3.4243668605226374e-06,
      "loss": 0.89383203,
      "num_input_tokens_seen": 48233025,
      "step": 2248,
      "time_per_iteration": 3.2924299240112305
    },
    {
      "auxiliary_loss_clip": 0.01157603,
      "auxiliary_loss_mlp": 0.01035154,
      "balance_loss_clip": 1.05695117,
      "balance_loss_mlp": 1.02496755,
      "epoch": 0.2704262610473156,
      "flos": 19572357329280.0,
      "grad_norm": 2.252389206973943,
      "language_loss": 0.82687432,
      "learning_rate": 3.423819919528061e-06,
      "loss": 0.84880185,
      "num_input_tokens_seen": 48251110,
      "step": 2249,
      "time_per_iteration": 2.545948028564453
    },
    {
      "auxiliary_loss_clip": 0.01145636,
      "auxiliary_loss_mlp": 0.01028299,
      "balance_loss_clip": 1.05055404,
      "balance_loss_mlp": 1.01962662,
      "epoch": 0.27054650393795465,
      "flos": 20740746925440.0,
      "grad_norm": 1.7786542082656838,
      "language_loss": 0.77917504,
      "learning_rate": 3.4232727625411355e-06,
      "loss": 0.80091441,
      "num_input_tokens_seen": 48270215,
      "step": 2250,
      "time_per_iteration": 2.62650203704834
    },
    {
      "auxiliary_loss_clip": 0.01122464,
      "auxiliary_loss_mlp": 0.01033659,
      "balance_loss_clip": 1.05063248,
      "balance_loss_mlp": 1.02586234,
      "epoch": 0.27066674682859376,
      "flos": 18658322916480.0,
      "grad_norm": 1.6614578282037553,
      "language_loss": 0.86242998,
      "learning_rate": 3.4227253896448626e-06,
      "loss": 0.88399124,
      "num_input_tokens_seen": 48288075,
      "step": 2251,
      "time_per_iteration": 2.6431922912597656
    },
    {
      "auxiliary_loss_clip": 0.01194589,
      "auxiliary_loss_mlp": 0.01031475,
      "balance_loss_clip": 1.05645943,
      "balance_loss_mlp": 1.02307057,
      "epoch": 0.2707869897192329,
      "flos": 23002759958400.0,
      "grad_norm": 3.866183862387396,
      "language_loss": 0.82192552,
      "learning_rate": 3.42217780092228e-06,
      "loss": 0.84418619,
      "num_input_tokens_seen": 48306415,
      "step": 2252,
      "time_per_iteration": 2.540240526199341
    },
    {
      "auxiliary_loss_clip": 0.01063102,
      "auxiliary_loss_mlp": 0.01006106,
      "balance_loss_clip": 1.02039075,
      "balance_loss_mlp": 1.00398433,
      "epoch": 0.27090723260987193,
      "flos": 58323240293760.0,
      "grad_norm": 1.075973138957363,
      "language_loss": 0.60361457,
      "learning_rate": 3.421629996456456e-06,
      "loss": 0.62430668,
      "num_input_tokens_seen": 48365035,
      "step": 2253,
      "time_per_iteration": 3.0880115032196045
    },
    {
      "auxiliary_loss_clip": 0.01185656,
      "auxiliary_loss_mlp": 0.01028135,
      "balance_loss_clip": 1.05892622,
      "balance_loss_mlp": 1.01928425,
      "epoch": 0.27102747550051104,
      "flos": 11984540797440.0,
      "grad_norm": 2.510374713282056,
      "language_loss": 0.82581669,
      "learning_rate": 3.421081976330491e-06,
      "loss": 0.84795463,
      "num_input_tokens_seen": 48383550,
      "step": 2254,
      "time_per_iteration": 3.3169963359832764
    },
    {
      "auxiliary_loss_clip": 0.01167592,
      "auxiliary_loss_mlp": 0.01030642,
      "balance_loss_clip": 1.05479491,
      "balance_loss_mlp": 1.02164793,
      "epoch": 0.27114771839115015,
      "flos": 19900401264000.0,
      "grad_norm": 1.8386960874784468,
      "language_loss": 0.87921578,
      "learning_rate": 3.4205337406275207e-06,
      "loss": 0.90119809,
      "num_input_tokens_seen": 48403670,
      "step": 2255,
      "time_per_iteration": 3.5977959632873535
    },
    {
      "auxiliary_loss_clip": 0.01196911,
      "auxiliary_loss_mlp": 0.01029099,
      "balance_loss_clip": 1.05784917,
      "balance_loss_mlp": 1.02028942,
      "epoch": 0.2712679612817892,
      "flos": 18331966920960.0,
      "grad_norm": 2.4170129637831548,
      "language_loss": 0.75643092,
      "learning_rate": 3.4199852894307114e-06,
      "loss": 0.77869105,
      "num_input_tokens_seen": 48420420,
      "step": 2256,
      "time_per_iteration": 3.2933542728424072
    },
    {
      "auxiliary_loss_clip": 0.01125728,
      "auxiliary_loss_mlp": 0.01036609,
      "balance_loss_clip": 1.05161619,
      "balance_loss_mlp": 1.02765667,
      "epoch": 0.2713882041724283,
      "flos": 24460302038400.0,
      "grad_norm": 2.0643096880949163,
      "language_loss": 0.78901768,
      "learning_rate": 3.419436622823262e-06,
      "loss": 0.81064099,
      "num_input_tokens_seen": 48441140,
      "step": 2257,
      "time_per_iteration": 2.7175052165985107
    },
    {
      "auxiliary_loss_clip": 0.01168734,
      "auxiliary_loss_mlp": 0.0103346,
      "balance_loss_clip": 1.05622458,
      "balance_loss_mlp": 1.02514529,
      "epoch": 0.27150844706306737,
      "flos": 23039317025280.0,
      "grad_norm": 1.5982451878459463,
      "language_loss": 0.7428233,
      "learning_rate": 3.4188877408884063e-06,
      "loss": 0.76484525,
      "num_input_tokens_seen": 48461845,
      "step": 2258,
      "time_per_iteration": 2.575711250305176
    },
    {
      "auxiliary_loss_clip": 0.01162188,
      "auxiliary_loss_mlp": 0.01033811,
      "balance_loss_clip": 1.05368352,
      "balance_loss_mlp": 1.02482224,
      "epoch": 0.2716286899537065,
      "flos": 22563644192640.0,
      "grad_norm": 2.447584437659268,
      "language_loss": 0.65511882,
      "learning_rate": 3.4183386437094088e-06,
      "loss": 0.67707884,
      "num_input_tokens_seen": 48478510,
      "step": 2259,
      "time_per_iteration": 2.551903009414673
    },
    {
      "auxiliary_loss_clip": 0.01167976,
      "auxiliary_loss_mlp": 0.01031545,
      "balance_loss_clip": 1.05380964,
      "balance_loss_mlp": 1.02256894,
      "epoch": 0.2717489328443456,
      "flos": 13115044523520.0,
      "grad_norm": 2.355237988514992,
      "language_loss": 0.82321292,
      "learning_rate": 3.417789331369565e-06,
      "loss": 0.84520811,
      "num_input_tokens_seen": 48494300,
      "step": 2260,
      "time_per_iteration": 2.5333566665649414
    },
    {
      "auxiliary_loss_clip": 0.01203283,
      "auxiliary_loss_mlp": 0.01035272,
      "balance_loss_clip": 1.06136799,
      "balance_loss_mlp": 1.02594376,
      "epoch": 0.27186917573498465,
      "flos": 29278688060160.0,
      "grad_norm": 2.027773563450376,
      "language_loss": 0.91154671,
      "learning_rate": 3.4172398039522088e-06,
      "loss": 0.9339323,
      "num_input_tokens_seen": 48515585,
      "step": 2261,
      "time_per_iteration": 2.5839219093322754
    },
    {
      "auxiliary_loss_clip": 0.01183076,
      "auxiliary_loss_mlp": 0.0103808,
      "balance_loss_clip": 1.0546602,
      "balance_loss_mlp": 1.0291574,
      "epoch": 0.27198941862562376,
      "flos": 26032220000640.0,
      "grad_norm": 1.602194166732769,
      "language_loss": 0.79815626,
      "learning_rate": 3.4166900615407e-06,
      "loss": 0.82036781,
      "num_input_tokens_seen": 48533500,
      "step": 2262,
      "time_per_iteration": 2.5713634490966797
    },
    {
      "auxiliary_loss_clip": 0.01181644,
      "auxiliary_loss_mlp": 0.01035002,
      "balance_loss_clip": 1.05782342,
      "balance_loss_mlp": 1.02604365,
      "epoch": 0.27210966151626287,
      "flos": 32780983760640.0,
      "grad_norm": 3.7549789484977882,
      "language_loss": 0.74882424,
      "learning_rate": 3.416140104218436e-06,
      "loss": 0.77099073,
      "num_input_tokens_seen": 48552865,
      "step": 2263,
      "time_per_iteration": 2.602962017059326
    },
    {
      "auxiliary_loss_clip": 0.01068012,
      "auxiliary_loss_mlp": 0.00753123,
      "balance_loss_clip": 1.01964021,
      "balance_loss_mlp": 1.00008857,
      "epoch": 0.2722299044069019,
      "flos": 65471043219840.0,
      "grad_norm": 0.8481393731686929,
      "language_loss": 0.69674075,
      "learning_rate": 3.4155899320688437e-06,
      "loss": 0.71495211,
      "num_input_tokens_seen": 48618940,
      "step": 2264,
      "time_per_iteration": 3.2006149291992188
    },
    {
      "auxiliary_loss_clip": 0.01131867,
      "auxiliary_loss_mlp": 0.01036602,
      "balance_loss_clip": 1.05524397,
      "balance_loss_mlp": 1.02785814,
      "epoch": 0.27235014729754103,
      "flos": 15334143782400.0,
      "grad_norm": 2.1999569119108955,
      "language_loss": 0.74411392,
      "learning_rate": 3.415039545175384e-06,
      "loss": 0.76579857,
      "num_input_tokens_seen": 48634665,
      "step": 2265,
      "time_per_iteration": 2.59149169921875
    },
    {
      "auxiliary_loss_clip": 0.01183574,
      "auxiliary_loss_mlp": 0.01031228,
      "balance_loss_clip": 1.05618286,
      "balance_loss_mlp": 1.02316928,
      "epoch": 0.27247039018818014,
      "flos": 21872363973120.0,
      "grad_norm": 1.9811182495103772,
      "language_loss": 0.65338802,
      "learning_rate": 3.414488943621551e-06,
      "loss": 0.67553604,
      "num_input_tokens_seen": 48653330,
      "step": 2266,
      "time_per_iteration": 2.648681163787842
    },
    {
      "auxiliary_loss_clip": 0.01180206,
      "auxiliary_loss_mlp": 0.01029991,
      "balance_loss_clip": 1.05708086,
      "balance_loss_mlp": 1.02144992,
      "epoch": 0.2725906330788192,
      "flos": 18695490514560.0,
      "grad_norm": 3.436472819122573,
      "language_loss": 0.73715234,
      "learning_rate": 3.41393812749087e-06,
      "loss": 0.75925434,
      "num_input_tokens_seen": 48671375,
      "step": 2267,
      "time_per_iteration": 2.480748414993286
    },
    {
      "auxiliary_loss_clip": 0.01171464,
      "auxiliary_loss_mlp": 0.01032447,
      "balance_loss_clip": 1.05973148,
      "balance_loss_mlp": 1.02409697,
      "epoch": 0.2727108759694583,
      "flos": 17886099398400.0,
      "grad_norm": 2.1550656314549985,
      "language_loss": 0.71902907,
      "learning_rate": 3.4133870968668984e-06,
      "loss": 0.74106818,
      "num_input_tokens_seen": 48686175,
      "step": 2268,
      "time_per_iteration": 2.572136878967285
    },
    {
      "auxiliary_loss_clip": 0.0117183,
      "auxiliary_loss_mlp": 0.01031925,
      "balance_loss_clip": 1.05667353,
      "balance_loss_mlp": 1.02279353,
      "epoch": 0.2728311188600974,
      "flos": 24461666755200.0,
      "grad_norm": 1.788150724339478,
      "language_loss": 0.786641,
      "learning_rate": 3.412835851833229e-06,
      "loss": 0.80867851,
      "num_input_tokens_seen": 48708370,
      "step": 2269,
      "time_per_iteration": 2.5876986980438232
    },
    {
      "auxiliary_loss_clip": 0.01179797,
      "auxiliary_loss_mlp": 0.01030713,
      "balance_loss_clip": 1.05809331,
      "balance_loss_mlp": 1.0227077,
      "epoch": 0.2729513617507365,
      "flos": 30993314757120.0,
      "grad_norm": 1.8001756783304097,
      "language_loss": 0.77820289,
      "learning_rate": 3.4122843924734834e-06,
      "loss": 0.80030799,
      "num_input_tokens_seen": 48730670,
      "step": 2270,
      "time_per_iteration": 2.6624083518981934
    },
    {
      "auxiliary_loss_clip": 0.01164215,
      "auxiliary_loss_mlp": 0.01033999,
      "balance_loss_clip": 1.05291176,
      "balance_loss_mlp": 1.02509463,
      "epoch": 0.2730716046413756,
      "flos": 19094637421440.0,
      "grad_norm": 3.3996512260842264,
      "language_loss": 0.87893951,
      "learning_rate": 3.411732718871319e-06,
      "loss": 0.9009217,
      "num_input_tokens_seen": 48746510,
      "step": 2271,
      "time_per_iteration": 2.516752004623413
    },
    {
      "auxiliary_loss_clip": 0.01200705,
      "auxiliary_loss_mlp": 0.01043313,
      "balance_loss_clip": 1.06394768,
      "balance_loss_mlp": 1.03538251,
      "epoch": 0.27319184753201464,
      "flos": 26944566474240.0,
      "grad_norm": 1.5351493955583282,
      "language_loss": 0.78631657,
      "learning_rate": 3.4111808311104227e-06,
      "loss": 0.80875671,
      "num_input_tokens_seen": 48768825,
      "step": 2272,
      "time_per_iteration": 2.6020734310150146
    },
    {
      "auxiliary_loss_clip": 0.0117568,
      "auxiliary_loss_mlp": 0.0103138,
      "balance_loss_clip": 1.05491078,
      "balance_loss_mlp": 1.02276075,
      "epoch": 0.27331209042265375,
      "flos": 31759828012800.0,
      "grad_norm": 1.6908205418561253,
      "language_loss": 0.69405758,
      "learning_rate": 3.410628729274517e-06,
      "loss": 0.71612823,
      "num_input_tokens_seen": 48790345,
      "step": 2273,
      "time_per_iteration": 2.6533362865448
    },
    {
      "auxiliary_loss_clip": 0.01169293,
      "auxiliary_loss_mlp": 0.00762278,
      "balance_loss_clip": 1.05876446,
      "balance_loss_mlp": 1.00030541,
      "epoch": 0.27343233331329286,
      "flos": 25739081107200.0,
      "grad_norm": 1.8678985382687405,
      "language_loss": 0.8258878,
      "learning_rate": 3.4100764134473546e-06,
      "loss": 0.84520352,
      "num_input_tokens_seen": 48809630,
      "step": 2274,
      "time_per_iteration": 3.484149932861328
    },
    {
      "auxiliary_loss_clip": 0.01199705,
      "auxiliary_loss_mlp": 0.01036378,
      "balance_loss_clip": 1.06190908,
      "balance_loss_mlp": 1.02710927,
      "epoch": 0.2735525762039319,
      "flos": 24389414547840.0,
      "grad_norm": 2.440916894271007,
      "language_loss": 0.8498624,
      "learning_rate": 3.4095238837127215e-06,
      "loss": 0.8722232,
      "num_input_tokens_seen": 48828770,
      "step": 2275,
      "time_per_iteration": 2.523191452026367
    },
    {
      "auxiliary_loss_clip": 0.01151603,
      "auxiliary_loss_mlp": 0.0103484,
      "balance_loss_clip": 1.05229747,
      "balance_loss_mlp": 1.02612615,
      "epoch": 0.27367281909457103,
      "flos": 14465357527680.0,
      "grad_norm": 1.9720022296458541,
      "language_loss": 0.79697514,
      "learning_rate": 3.4089711401544355e-06,
      "loss": 0.81883955,
      "num_input_tokens_seen": 48846365,
      "step": 2276,
      "time_per_iteration": 2.6317927837371826
    },
    {
      "auxiliary_loss_clip": 0.01181717,
      "auxiliary_loss_mlp": 0.01036244,
      "balance_loss_clip": 1.05582213,
      "balance_loss_mlp": 1.02717209,
      "epoch": 0.27379306198521014,
      "flos": 23476996247040.0,
      "grad_norm": 2.1710850209138504,
      "language_loss": 0.67650348,
      "learning_rate": 3.4084181828563486e-06,
      "loss": 0.69868308,
      "num_input_tokens_seen": 48863085,
      "step": 2277,
      "time_per_iteration": 2.5354247093200684
    },
    {
      "auxiliary_loss_clip": 0.01138721,
      "auxiliary_loss_mlp": 0.01035209,
      "balance_loss_clip": 1.05287623,
      "balance_loss_mlp": 1.02623296,
      "epoch": 0.2739133048758492,
      "flos": 17458152762240.0,
      "grad_norm": 1.6218890834247193,
      "language_loss": 0.70605528,
      "learning_rate": 3.4078650119023428e-06,
      "loss": 0.72779459,
      "num_input_tokens_seen": 48881400,
      "step": 2278,
      "time_per_iteration": 2.6225030422210693
    },
    {
      "auxiliary_loss_clip": 0.01128205,
      "auxiliary_loss_mlp": 0.01029558,
      "balance_loss_clip": 1.04796946,
      "balance_loss_mlp": 1.02124929,
      "epoch": 0.2740335477664883,
      "flos": 19273113123840.0,
      "grad_norm": 2.036580690976605,
      "language_loss": 0.74273491,
      "learning_rate": 3.4073116273763337e-06,
      "loss": 0.76431257,
      "num_input_tokens_seen": 48895845,
      "step": 2279,
      "time_per_iteration": 2.6097829341888428
    },
    {
      "auxiliary_loss_clip": 0.01172522,
      "auxiliary_loss_mlp": 0.01029652,
      "balance_loss_clip": 1.05498958,
      "balance_loss_mlp": 1.020908,
      "epoch": 0.2741537906571274,
      "flos": 26104723603200.0,
      "grad_norm": 3.888509359255447,
      "language_loss": 0.81337821,
      "learning_rate": 3.40675802936227e-06,
      "loss": 0.83539987,
      "num_input_tokens_seen": 48916630,
      "step": 2280,
      "time_per_iteration": 3.3913345336914062
    },
    {
      "auxiliary_loss_clip": 0.01161393,
      "auxiliary_loss_mlp": 0.01034777,
      "balance_loss_clip": 1.05448782,
      "balance_loss_mlp": 1.02647996,
      "epoch": 0.27427403354776647,
      "flos": 34164190644480.0,
      "grad_norm": 2.2894344403407896,
      "language_loss": 0.71678078,
      "learning_rate": 3.4062042179441318e-06,
      "loss": 0.73874247,
      "num_input_tokens_seen": 48937100,
      "step": 2281,
      "time_per_iteration": 3.3291873931884766
    },
    {
      "auxiliary_loss_clip": 0.01183796,
      "auxiliary_loss_mlp": 0.01033019,
      "balance_loss_clip": 1.06016493,
      "balance_loss_mlp": 1.0252527,
      "epoch": 0.2743942764384056,
      "flos": 18766988536320.0,
      "grad_norm": 2.0746420146189757,
      "language_loss": 0.80552638,
      "learning_rate": 3.4056501932059314e-06,
      "loss": 0.82769454,
      "num_input_tokens_seen": 48955175,
      "step": 2282,
      "time_per_iteration": 2.5367634296417236
    },
    {
      "auxiliary_loss_clip": 0.01095811,
      "auxiliary_loss_mlp": 0.01005538,
      "balance_loss_clip": 1.02536678,
      "balance_loss_mlp": 1.00328541,
      "epoch": 0.2745145193290447,
      "flos": 64904048058240.0,
      "grad_norm": 0.780079428552975,
      "language_loss": 0.58127439,
      "learning_rate": 3.405095955231715e-06,
      "loss": 0.60228789,
      "num_input_tokens_seen": 49006830,
      "step": 2283,
      "time_per_iteration": 3.843421220779419
    },
    {
      "auxiliary_loss_clip": 0.01185487,
      "auxiliary_loss_mlp": 0.01029124,
      "balance_loss_clip": 1.05570102,
      "balance_loss_mlp": 1.02040982,
      "epoch": 0.27463476221968375,
      "flos": 16136926796160.0,
      "grad_norm": 2.1543327281224935,
      "language_loss": 0.94339073,
      "learning_rate": 3.4045415041055585e-06,
      "loss": 0.96553689,
      "num_input_tokens_seen": 49022470,
      "step": 2284,
      "time_per_iteration": 2.4744014739990234
    },
    {
      "auxiliary_loss_clip": 0.01175467,
      "auxiliary_loss_mlp": 0.01031281,
      "balance_loss_clip": 1.05705309,
      "balance_loss_mlp": 1.02267432,
      "epoch": 0.27475500511032286,
      "flos": 10376712213120.0,
      "grad_norm": 2.2405872840752363,
      "language_loss": 0.78392631,
      "learning_rate": 3.4039868399115728e-06,
      "loss": 0.80599374,
      "num_input_tokens_seen": 49037110,
      "step": 2285,
      "time_per_iteration": 2.509169578552246
    },
    {
      "auxiliary_loss_clip": 0.01133063,
      "auxiliary_loss_mlp": 0.01034206,
      "balance_loss_clip": 1.05610323,
      "balance_loss_mlp": 1.02559328,
      "epoch": 0.27487524800096197,
      "flos": 17311062568320.0,
      "grad_norm": 1.8219634809793295,
      "language_loss": 0.80644256,
      "learning_rate": 3.4034319627339003e-06,
      "loss": 0.82811522,
      "num_input_tokens_seen": 49053975,
      "step": 2286,
      "time_per_iteration": 2.5945608615875244
    },
    {
      "auxiliary_loss_clip": 0.0117601,
      "auxiliary_loss_mlp": 0.0103255,
      "balance_loss_clip": 1.05919862,
      "balance_loss_mlp": 1.02302563,
      "epoch": 0.274995490891601,
      "flos": 27120205002240.0,
      "grad_norm": 2.447530144356101,
      "language_loss": 0.69527757,
      "learning_rate": 3.402876872656715e-06,
      "loss": 0.71736318,
      "num_input_tokens_seen": 49072295,
      "step": 2287,
      "time_per_iteration": 2.5797178745269775
    },
    {
      "auxiliary_loss_clip": 0.01170615,
      "auxiliary_loss_mlp": 0.01034501,
      "balance_loss_clip": 1.06035018,
      "balance_loss_mlp": 1.02588856,
      "epoch": 0.27511573378224013,
      "flos": 23436093634560.0,
      "grad_norm": 1.8859791511865065,
      "language_loss": 0.89366668,
      "learning_rate": 3.402321569764223e-06,
      "loss": 0.91571784,
      "num_input_tokens_seen": 49091600,
      "step": 2288,
      "time_per_iteration": 2.5705463886260986
    },
    {
      "auxiliary_loss_clip": 0.01143996,
      "auxiliary_loss_mlp": 0.00763037,
      "balance_loss_clip": 1.05294728,
      "balance_loss_mlp": 1.0003767,
      "epoch": 0.2752359766728792,
      "flos": 16722019434240.0,
      "grad_norm": 1.750682033908901,
      "language_loss": 0.83623439,
      "learning_rate": 3.4017660541406635e-06,
      "loss": 0.85530472,
      "num_input_tokens_seen": 49107665,
      "step": 2289,
      "time_per_iteration": 2.564131259918213
    },
    {
      "auxiliary_loss_clip": 0.01176769,
      "auxiliary_loss_mlp": 0.01033275,
      "balance_loss_clip": 1.05448079,
      "balance_loss_mlp": 1.02477527,
      "epoch": 0.2753562195635183,
      "flos": 25297738698240.0,
      "grad_norm": 1.6274894987992146,
      "language_loss": 0.74114525,
      "learning_rate": 3.4012103258703092e-06,
      "loss": 0.7632457,
      "num_input_tokens_seen": 49126420,
      "step": 2290,
      "time_per_iteration": 2.583829641342163
    },
    {
      "auxiliary_loss_clip": 0.01155931,
      "auxiliary_loss_mlp": 0.01027624,
      "balance_loss_clip": 1.05345964,
      "balance_loss_mlp": 1.01913071,
      "epoch": 0.2754764624541574,
      "flos": 27338972785920.0,
      "grad_norm": 1.9034205286247927,
      "language_loss": 0.83004677,
      "learning_rate": 3.4006543850374616e-06,
      "loss": 0.85188234,
      "num_input_tokens_seen": 49141470,
      "step": 2291,
      "time_per_iteration": 2.6084015369415283
    },
    {
      "auxiliary_loss_clip": 0.01185322,
      "auxiliary_loss_mlp": 0.01029381,
      "balance_loss_clip": 1.05609512,
      "balance_loss_mlp": 1.02111387,
      "epoch": 0.27559670534479647,
      "flos": 17238379397760.0,
      "grad_norm": 3.452400852496113,
      "language_loss": 0.74744749,
      "learning_rate": 3.400098231726458e-06,
      "loss": 0.76959455,
      "num_input_tokens_seen": 49158570,
      "step": 2292,
      "time_per_iteration": 2.496499538421631
    },
    {
      "auxiliary_loss_clip": 0.01160475,
      "auxiliary_loss_mlp": 0.01033232,
      "balance_loss_clip": 1.05353534,
      "balance_loss_mlp": 1.02452922,
      "epoch": 0.2757169482354356,
      "flos": 21939085486080.0,
      "grad_norm": 1.9006559471212872,
      "language_loss": 0.86941755,
      "learning_rate": 3.3995418660216657e-06,
      "loss": 0.89135456,
      "num_input_tokens_seen": 49176025,
      "step": 2293,
      "time_per_iteration": 2.569826602935791
    },
    {
      "auxiliary_loss_clip": 0.01206133,
      "auxiliary_loss_mlp": 0.01032502,
      "balance_loss_clip": 1.06126571,
      "balance_loss_mlp": 1.02295303,
      "epoch": 0.2758371911260747,
      "flos": 20850669521280.0,
      "grad_norm": 2.4926548021187394,
      "language_loss": 0.80531693,
      "learning_rate": 3.3989852880074848e-06,
      "loss": 0.82770324,
      "num_input_tokens_seen": 49197455,
      "step": 2294,
      "time_per_iteration": 2.5286197662353516
    },
    {
      "auxiliary_loss_clip": 0.01072357,
      "auxiliary_loss_mlp": 0.01002806,
      "balance_loss_clip": 1.02571154,
      "balance_loss_mlp": 1.00073195,
      "epoch": 0.27595743401671374,
      "flos": 69269063592960.0,
      "grad_norm": 0.7477702102275624,
      "language_loss": 0.60645652,
      "learning_rate": 3.398428497768348e-06,
      "loss": 0.62720811,
      "num_input_tokens_seen": 49262625,
      "step": 2295,
      "time_per_iteration": 3.259068250656128
    },
    {
      "auxiliary_loss_clip": 0.01164687,
      "auxiliary_loss_mlp": 0.01033976,
      "balance_loss_clip": 1.05507493,
      "balance_loss_mlp": 1.02591777,
      "epoch": 0.27607767690735285,
      "flos": 21215019127680.0,
      "grad_norm": 1.6307159120361676,
      "language_loss": 0.71894181,
      "learning_rate": 3.3978714953887205e-06,
      "loss": 0.74092841,
      "num_input_tokens_seen": 49282380,
      "step": 2296,
      "time_per_iteration": 2.5908689498901367
    },
    {
      "auxiliary_loss_clip": 0.01127996,
      "auxiliary_loss_mlp": 0.01032992,
      "balance_loss_clip": 1.04803848,
      "balance_loss_mlp": 1.02377665,
      "epoch": 0.27619791979799196,
      "flos": 24825334003200.0,
      "grad_norm": 1.6951924245105532,
      "language_loss": 0.85909843,
      "learning_rate": 3.397314280953098e-06,
      "loss": 0.88070828,
      "num_input_tokens_seen": 49303205,
      "step": 2297,
      "time_per_iteration": 2.646383762359619
    },
    {
      "auxiliary_loss_clip": 0.01165186,
      "auxiliary_loss_mlp": 0.01031782,
      "balance_loss_clip": 1.05380154,
      "balance_loss_mlp": 1.02311552,
      "epoch": 0.276318162688631,
      "flos": 24753548672640.0,
      "grad_norm": 1.9609257505174187,
      "language_loss": 0.79776949,
      "learning_rate": 3.3967568545460108e-06,
      "loss": 0.81973922,
      "num_input_tokens_seen": 49322745,
      "step": 2298,
      "time_per_iteration": 2.5776305198669434
    },
    {
      "auxiliary_loss_clip": 0.01187015,
      "auxiliary_loss_mlp": 0.01031575,
      "balance_loss_clip": 1.061674,
      "balance_loss_mlp": 1.02259278,
      "epoch": 0.27643840557927013,
      "flos": 18150007599360.0,
      "grad_norm": 1.94470900507694,
      "language_loss": 0.80802852,
      "learning_rate": 3.3961992162520185e-06,
      "loss": 0.83021438,
      "num_input_tokens_seen": 49341370,
      "step": 2299,
      "time_per_iteration": 2.496711015701294
    },
    {
      "auxiliary_loss_clip": 0.0118406,
      "auxiliary_loss_mlp": 0.01030525,
      "balance_loss_clip": 1.0559057,
      "balance_loss_mlp": 1.02144158,
      "epoch": 0.27655864846990924,
      "flos": 24823933372800.0,
      "grad_norm": 2.0237109343743804,
      "language_loss": 0.71622896,
      "learning_rate": 3.3956413661557156e-06,
      "loss": 0.73837483,
      "num_input_tokens_seen": 49361545,
      "step": 2300,
      "time_per_iteration": 3.2322211265563965
    },
    {
      "auxiliary_loss_clip": 0.01161384,
      "auxiliary_loss_mlp": 0.01033318,
      "balance_loss_clip": 1.05302405,
      "balance_loss_mlp": 1.0242939,
      "epoch": 0.2766788913605483,
      "flos": 20266582464000.0,
      "grad_norm": 2.1263552932332885,
      "language_loss": 0.65990627,
      "learning_rate": 3.3950833043417273e-06,
      "loss": 0.68185329,
      "num_input_tokens_seen": 49379690,
      "step": 2301,
      "time_per_iteration": 2.573422431945801
    },
    {
      "auxiliary_loss_clip": 0.01189766,
      "auxiliary_loss_mlp": 0.01035358,
      "balance_loss_clip": 1.0613606,
      "balance_loss_mlp": 1.02589869,
      "epoch": 0.2767991342511874,
      "flos": 21470272151040.0,
      "grad_norm": 2.4389957663547936,
      "language_loss": 0.73478973,
      "learning_rate": 3.3945250308947105e-06,
      "loss": 0.75704098,
      "num_input_tokens_seen": 49395995,
      "step": 2302,
      "time_per_iteration": 2.495997905731201
    },
    {
      "auxiliary_loss_clip": 0.01084349,
      "auxiliary_loss_mlp": 0.01002615,
      "balance_loss_clip": 1.02532744,
      "balance_loss_mlp": 1.00048161,
      "epoch": 0.2769193771418265,
      "flos": 66002627571840.0,
      "grad_norm": 1.284581056719423,
      "language_loss": 0.68331122,
      "learning_rate": 3.3939665458993556e-06,
      "loss": 0.70418084,
      "num_input_tokens_seen": 49450415,
      "step": 2303,
      "time_per_iteration": 3.056732177734375
    },
    {
      "auxiliary_loss_clip": 0.01160896,
      "auxiliary_loss_mlp": 0.01027349,
      "balance_loss_clip": 1.05259442,
      "balance_loss_mlp": 1.01846802,
      "epoch": 0.27703962003246557,
      "flos": 20704441253760.0,
      "grad_norm": 2.3994453602002253,
      "language_loss": 0.76605582,
      "learning_rate": 3.3934078494403843e-06,
      "loss": 0.7879383,
      "num_input_tokens_seen": 49469990,
      "step": 2304,
      "time_per_iteration": 2.6541948318481445
    },
    {
      "auxiliary_loss_clip": 0.01108309,
      "auxiliary_loss_mlp": 0.00763169,
      "balance_loss_clip": 1.04957771,
      "balance_loss_mlp": 1.00035763,
      "epoch": 0.2771598629231047,
      "flos": 22929897219840.0,
      "grad_norm": 1.7117782930569578,
      "language_loss": 0.81321514,
      "learning_rate": 3.3928489416025495e-06,
      "loss": 0.83192986,
      "num_input_tokens_seen": 49490835,
      "step": 2305,
      "time_per_iteration": 2.699719190597534
    },
    {
      "auxiliary_loss_clip": 0.01169345,
      "auxiliary_loss_mlp": 0.0103371,
      "balance_loss_clip": 1.0562526,
      "balance_loss_mlp": 1.02475715,
      "epoch": 0.27728010581374374,
      "flos": 18369457741440.0,
      "grad_norm": 2.3706953056481535,
      "language_loss": 0.79159749,
      "learning_rate": 3.392289822470638e-06,
      "loss": 0.81362808,
      "num_input_tokens_seen": 49508815,
      "step": 2306,
      "time_per_iteration": 4.017672300338745
    },
    {
      "auxiliary_loss_clip": 0.01170894,
      "auxiliary_loss_mlp": 0.01032487,
      "balance_loss_clip": 1.05587888,
      "balance_loss_mlp": 1.0233078,
      "epoch": 0.27740034870438285,
      "flos": 19427637432960.0,
      "grad_norm": 3.0339486039165546,
      "language_loss": 0.76025456,
      "learning_rate": 3.3917304921294674e-06,
      "loss": 0.78228843,
      "num_input_tokens_seen": 49526980,
      "step": 2307,
      "time_per_iteration": 2.5429069995880127
    },
    {
      "auxiliary_loss_clip": 0.01186361,
      "auxiliary_loss_mlp": 0.01031375,
      "balance_loss_clip": 1.05800498,
      "balance_loss_mlp": 1.02219605,
      "epoch": 0.27752059159502196,
      "flos": 21614776565760.0,
      "grad_norm": 9.416918470282337,
      "language_loss": 0.80699545,
      "learning_rate": 3.3911709506638876e-06,
      "loss": 0.82917285,
      "num_input_tokens_seen": 49546290,
      "step": 2308,
      "time_per_iteration": 2.5394418239593506
    },
    {
      "auxiliary_loss_clip": 0.01145806,
      "auxiliary_loss_mlp": 0.00762606,
      "balance_loss_clip": 1.04922771,
      "balance_loss_mlp": 1.00033951,
      "epoch": 0.277640834485661,
      "flos": 26608011016320.0,
      "grad_norm": 2.4037091303076115,
      "language_loss": 0.81399131,
      "learning_rate": 3.390611198158781e-06,
      "loss": 0.8330754,
      "num_input_tokens_seen": 49564165,
      "step": 2309,
      "time_per_iteration": 3.4194347858428955
    },
    {
      "auxiliary_loss_clip": 0.01201752,
      "auxiliary_loss_mlp": 0.01038655,
      "balance_loss_clip": 1.06038666,
      "balance_loss_mlp": 1.02965486,
      "epoch": 0.2777610773763001,
      "flos": 19492814661120.0,
      "grad_norm": 2.060080663708529,
      "language_loss": 0.89733112,
      "learning_rate": 3.3900512346990612e-06,
      "loss": 0.91973519,
      "num_input_tokens_seen": 49580155,
      "step": 2310,
      "time_per_iteration": 2.483231782913208
    },
    {
      "auxiliary_loss_clip": 0.01141226,
      "auxiliary_loss_mlp": 0.01035675,
      "balance_loss_clip": 1.04985797,
      "balance_loss_mlp": 1.02600145,
      "epoch": 0.27788132026693924,
      "flos": 38290650001920.0,
      "grad_norm": 1.5529337323387418,
      "language_loss": 0.6577422,
      "learning_rate": 3.389491060369674e-06,
      "loss": 0.67951119,
      "num_input_tokens_seen": 49605830,
      "step": 2311,
      "time_per_iteration": 2.753255605697632
    },
    {
      "auxiliary_loss_clip": 0.01136092,
      "auxiliary_loss_mlp": 0.01029333,
      "balance_loss_clip": 1.05210257,
      "balance_loss_mlp": 1.02030838,
      "epoch": 0.2780015631575783,
      "flos": 22382546797440.0,
      "grad_norm": 1.8922797477492919,
      "language_loss": 0.89742136,
      "learning_rate": 3.388930675255598e-06,
      "loss": 0.91907561,
      "num_input_tokens_seen": 49625680,
      "step": 2312,
      "time_per_iteration": 2.6003127098083496
    },
    {
      "auxiliary_loss_clip": 0.01176891,
      "auxiliary_loss_mlp": 0.01037093,
      "balance_loss_clip": 1.05801845,
      "balance_loss_mlp": 1.02878451,
      "epoch": 0.2781218060482174,
      "flos": 12203200840320.0,
      "grad_norm": 2.9341162273159656,
      "language_loss": 0.79254663,
      "learning_rate": 3.388370079441843e-06,
      "loss": 0.81468654,
      "num_input_tokens_seen": 49641195,
      "step": 2313,
      "time_per_iteration": 2.5437889099121094
    },
    {
      "auxiliary_loss_clip": 0.01165516,
      "auxiliary_loss_mlp": 0.01025489,
      "balance_loss_clip": 1.06128693,
      "balance_loss_mlp": 1.01664984,
      "epoch": 0.2782420489388565,
      "flos": 18107632529280.0,
      "grad_norm": 2.7830308280423215,
      "language_loss": 0.92708522,
      "learning_rate": 3.3878092730134505e-06,
      "loss": 0.94899529,
      "num_input_tokens_seen": 49659180,
      "step": 2314,
      "time_per_iteration": 2.5944576263427734
    },
    {
      "auxiliary_loss_clip": 0.01178343,
      "auxiliary_loss_mlp": 0.01033178,
      "balance_loss_clip": 1.05664587,
      "balance_loss_mlp": 1.02424335,
      "epoch": 0.27836229182949557,
      "flos": 18514752255360.0,
      "grad_norm": 1.6877238273571884,
      "language_loss": 0.80810231,
      "learning_rate": 3.3872482560554947e-06,
      "loss": 0.83021748,
      "num_input_tokens_seen": 49677955,
      "step": 2315,
      "time_per_iteration": 2.5167667865753174
    },
    {
      "auxiliary_loss_clip": 0.01078365,
      "auxiliary_loss_mlp": 0.01001793,
      "balance_loss_clip": 1.01988482,
      "balance_loss_mlp": 0.99956381,
      "epoch": 0.2784825347201347,
      "flos": 67079230940160.0,
      "grad_norm": 0.8015104850419964,
      "language_loss": 0.57017863,
      "learning_rate": 3.386687028653082e-06,
      "loss": 0.59098023,
      "num_input_tokens_seen": 49740800,
      "step": 2316,
      "time_per_iteration": 3.133065700531006
    },
    {
      "auxiliary_loss_clip": 0.01144245,
      "auxiliary_loss_mlp": 0.01035397,
      "balance_loss_clip": 1.05539536,
      "balance_loss_mlp": 1.02654564,
      "epoch": 0.2786027776107738,
      "flos": 22631119891200.0,
      "grad_norm": 1.806315407295578,
      "language_loss": 0.85074472,
      "learning_rate": 3.386125590891349e-06,
      "loss": 0.87254119,
      "num_input_tokens_seen": 49757675,
      "step": 2317,
      "time_per_iteration": 2.62072491645813
    },
    {
      "auxiliary_loss_clip": 0.0115682,
      "auxiliary_loss_mlp": 0.01029952,
      "balance_loss_clip": 1.05097592,
      "balance_loss_mlp": 1.02168489,
      "epoch": 0.27872302050141284,
      "flos": 15778826156160.0,
      "grad_norm": 1.9837047466976963,
      "language_loss": 0.82905364,
      "learning_rate": 3.3855639428554657e-06,
      "loss": 0.85092139,
      "num_input_tokens_seen": 49775205,
      "step": 2318,
      "time_per_iteration": 2.546367883682251
    },
    {
      "auxiliary_loss_clip": 0.01146372,
      "auxiliary_loss_mlp": 0.01032292,
      "balance_loss_clip": 1.0535562,
      "balance_loss_mlp": 1.02405477,
      "epoch": 0.27884326339205195,
      "flos": 22126970551680.0,
      "grad_norm": 1.753956472785239,
      "language_loss": 0.80786771,
      "learning_rate": 3.385002084630635e-06,
      "loss": 0.82965434,
      "num_input_tokens_seen": 49794175,
      "step": 2319,
      "time_per_iteration": 2.57684063911438
    },
    {
      "auxiliary_loss_clip": 0.01188095,
      "auxiliary_loss_mlp": 0.01032736,
      "balance_loss_clip": 1.05644846,
      "balance_loss_mlp": 1.02374172,
      "epoch": 0.278963506282691,
      "flos": 20558715776640.0,
      "grad_norm": 1.8549229915251515,
      "language_loss": 0.85078871,
      "learning_rate": 3.384440016302088e-06,
      "loss": 0.87299699,
      "num_input_tokens_seen": 49812850,
      "step": 2320,
      "time_per_iteration": 2.500657796859741
    },
    {
      "auxiliary_loss_clip": 0.01179033,
      "auxiliary_loss_mlp": 0.01025053,
      "balance_loss_clip": 1.05489326,
      "balance_loss_mlp": 1.01639235,
      "epoch": 0.2790837491733301,
      "flos": 21942928241280.0,
      "grad_norm": 2.045960203096301,
      "language_loss": 0.61753112,
      "learning_rate": 3.3838777379550923e-06,
      "loss": 0.63957202,
      "num_input_tokens_seen": 49832295,
      "step": 2321,
      "time_per_iteration": 2.5185043811798096
    },
    {
      "auxiliary_loss_clip": 0.01172967,
      "auxiliary_loss_mlp": 0.01038195,
      "balance_loss_clip": 1.05738866,
      "balance_loss_mlp": 1.02950466,
      "epoch": 0.27920399206396923,
      "flos": 26286790665600.0,
      "grad_norm": 5.06074148636033,
      "language_loss": 0.7857995,
      "learning_rate": 3.383315249674944e-06,
      "loss": 0.80791122,
      "num_input_tokens_seen": 49850860,
      "step": 2322,
      "time_per_iteration": 2.5764336585998535
    },
    {
      "auxiliary_loss_clip": 0.01161928,
      "auxiliary_loss_mlp": 0.01036245,
      "balance_loss_clip": 1.05752814,
      "balance_loss_mlp": 1.02765036,
      "epoch": 0.2793242349546083,
      "flos": 25400981364480.0,
      "grad_norm": 1.8712862256211764,
      "language_loss": 0.85873115,
      "learning_rate": 3.3827525515469715e-06,
      "loss": 0.88071281,
      "num_input_tokens_seen": 49865765,
      "step": 2323,
      "time_per_iteration": 2.5874247550964355
    },
    {
      "auxiliary_loss_clip": 0.01151026,
      "auxiliary_loss_mlp": 0.0103359,
      "balance_loss_clip": 1.05195189,
      "balance_loss_mlp": 1.02407146,
      "epoch": 0.2794444778452474,
      "flos": 20850346298880.0,
      "grad_norm": 2.1852777591156536,
      "language_loss": 0.70642447,
      "learning_rate": 3.3821896436565367e-06,
      "loss": 0.72827065,
      "num_input_tokens_seen": 49885425,
      "step": 2324,
      "time_per_iteration": 2.5841286182403564
    },
    {
      "auxiliary_loss_clip": 0.01190584,
      "auxiliary_loss_mlp": 0.01035146,
      "balance_loss_clip": 1.06318879,
      "balance_loss_mlp": 1.02670014,
      "epoch": 0.2795647207358865,
      "flos": 21576244250880.0,
      "grad_norm": 1.6838341951319196,
      "language_loss": 0.70392448,
      "learning_rate": 3.381626526089032e-06,
      "loss": 0.72618175,
      "num_input_tokens_seen": 49904990,
      "step": 2325,
      "time_per_iteration": 3.3648955821990967
    },
    {
      "auxiliary_loss_clip": 0.01167087,
      "auxiliary_loss_mlp": 0.01032214,
      "balance_loss_clip": 1.05392408,
      "balance_loss_mlp": 1.02344632,
      "epoch": 0.27968496362652556,
      "flos": 21471744608640.0,
      "grad_norm": 1.9927110247272322,
      "language_loss": 0.79081529,
      "learning_rate": 3.3810631989298815e-06,
      "loss": 0.81280828,
      "num_input_tokens_seen": 49924600,
      "step": 2326,
      "time_per_iteration": 2.543272018432617
    },
    {
      "auxiliary_loss_clip": 0.01150093,
      "auxiliary_loss_mlp": 0.01029562,
      "balance_loss_clip": 1.05717731,
      "balance_loss_mlp": 1.02026916,
      "epoch": 0.2798052065171647,
      "flos": 23258695340160.0,
      "grad_norm": 1.974147088696905,
      "language_loss": 0.83774936,
      "learning_rate": 3.3804996622645423e-06,
      "loss": 0.85954589,
      "num_input_tokens_seen": 49942600,
      "step": 2327,
      "time_per_iteration": 2.626181125640869
    },
    {
      "auxiliary_loss_clip": 0.01200029,
      "auxiliary_loss_mlp": 0.01032782,
      "balance_loss_clip": 1.05928755,
      "balance_loss_mlp": 1.02386498,
      "epoch": 0.2799254494078038,
      "flos": 21539328048000.0,
      "grad_norm": 1.8668572346391497,
      "language_loss": 0.89678979,
      "learning_rate": 3.3799359161785015e-06,
      "loss": 0.91911787,
      "num_input_tokens_seen": 49962250,
      "step": 2328,
      "time_per_iteration": 2.508880615234375
    },
    {
      "auxiliary_loss_clip": 0.01186627,
      "auxiliary_loss_mlp": 0.01026439,
      "balance_loss_clip": 1.05880737,
      "balance_loss_mlp": 1.01773667,
      "epoch": 0.28004569229844284,
      "flos": 26393912000640.0,
      "grad_norm": 1.5293267680037377,
      "language_loss": 0.85518348,
      "learning_rate": 3.3793719607572798e-06,
      "loss": 0.87731409,
      "num_input_tokens_seen": 49983215,
      "step": 2329,
      "time_per_iteration": 2.582002639770508
    },
    {
      "auxiliary_loss_clip": 0.01154672,
      "auxiliary_loss_mlp": 0.01033533,
      "balance_loss_clip": 1.05176592,
      "balance_loss_mlp": 1.02453828,
      "epoch": 0.28016593518908195,
      "flos": 33547676584320.0,
      "grad_norm": 1.883495065171994,
      "language_loss": 0.76823884,
      "learning_rate": 3.378807796086428e-06,
      "loss": 0.7901209,
      "num_input_tokens_seen": 50006075,
      "step": 2330,
      "time_per_iteration": 2.649073839187622
    },
    {
      "auxiliary_loss_clip": 0.01203426,
      "auxiliary_loss_mlp": 0.01036393,
      "balance_loss_clip": 1.06290126,
      "balance_loss_mlp": 1.02768493,
      "epoch": 0.28028617807972106,
      "flos": 15340823712000.0,
      "grad_norm": 3.5502929597227815,
      "language_loss": 0.76534289,
      "learning_rate": 3.37824342225153e-06,
      "loss": 0.78774107,
      "num_input_tokens_seen": 50022495,
      "step": 2331,
      "time_per_iteration": 2.4641940593719482
    },
    {
      "auxiliary_loss_clip": 0.01146351,
      "auxiliary_loss_mlp": 0.01027906,
      "balance_loss_clip": 1.05596852,
      "balance_loss_mlp": 1.01950204,
      "epoch": 0.2804064209703601,
      "flos": 25520277409920.0,
      "grad_norm": 1.8384849859055858,
      "language_loss": 0.77740169,
      "learning_rate": 3.3776788393382006e-06,
      "loss": 0.79914427,
      "num_input_tokens_seen": 50041975,
      "step": 2332,
      "time_per_iteration": 4.097648620605469
    },
    {
      "auxiliary_loss_clip": 0.01204909,
      "auxiliary_loss_mlp": 0.01035276,
      "balance_loss_clip": 1.06305301,
      "balance_loss_mlp": 1.02567995,
      "epoch": 0.2805266638609992,
      "flos": 29351766280320.0,
      "grad_norm": 2.3644137433015375,
      "language_loss": 0.76232588,
      "learning_rate": 3.3771140474320872e-06,
      "loss": 0.78472775,
      "num_input_tokens_seen": 50061925,
      "step": 2333,
      "time_per_iteration": 2.5466911792755127
    },
    {
      "auxiliary_loss_clip": 0.01166638,
      "auxiliary_loss_mlp": 0.010351,
      "balance_loss_clip": 1.05780804,
      "balance_loss_mlp": 1.02610588,
      "epoch": 0.28064690675163834,
      "flos": 21463735875840.0,
      "grad_norm": 2.075772375120525,
      "language_loss": 0.79563433,
      "learning_rate": 3.3765490466188664e-06,
      "loss": 0.81765175,
      "num_input_tokens_seen": 50079325,
      "step": 2334,
      "time_per_iteration": 2.5882232189178467
    },
    {
      "auxiliary_loss_clip": 0.01159415,
      "auxiliary_loss_mlp": 0.01038455,
      "balance_loss_clip": 1.0573988,
      "balance_loss_mlp": 1.02964568,
      "epoch": 0.2807671496422774,
      "flos": 20995640812800.0,
      "grad_norm": 2.465579958995106,
      "language_loss": 0.73542249,
      "learning_rate": 3.3759838369842508e-06,
      "loss": 0.75740123,
      "num_input_tokens_seen": 50097400,
      "step": 2335,
      "time_per_iteration": 3.3503365516662598
    },
    {
      "auxiliary_loss_clip": 0.01156852,
      "auxiliary_loss_mlp": 0.01030001,
      "balance_loss_clip": 1.05525935,
      "balance_loss_mlp": 1.02116799,
      "epoch": 0.2808873925329165,
      "flos": 21506577822720.0,
      "grad_norm": 2.0364294958930533,
      "language_loss": 0.72825623,
      "learning_rate": 3.375418418613981e-06,
      "loss": 0.75012481,
      "num_input_tokens_seen": 50116425,
      "step": 2336,
      "time_per_iteration": 2.5756640434265137
    },
    {
      "auxiliary_loss_clip": 0.01178136,
      "auxiliary_loss_mlp": 0.01036338,
      "balance_loss_clip": 1.05980349,
      "balance_loss_mlp": 1.02750492,
      "epoch": 0.28100763542355556,
      "flos": 16070815814400.0,
      "grad_norm": 2.236579214357391,
      "language_loss": 0.82947671,
      "learning_rate": 3.374852791593831e-06,
      "loss": 0.85162145,
      "num_input_tokens_seen": 50132625,
      "step": 2337,
      "time_per_iteration": 2.5386619567871094
    },
    {
      "auxiliary_loss_clip": 0.01155276,
      "auxiliary_loss_mlp": 0.01035501,
      "balance_loss_clip": 1.05408335,
      "balance_loss_mlp": 1.02631593,
      "epoch": 0.28112787831419467,
      "flos": 19062605468160.0,
      "grad_norm": 2.2226510703510614,
      "language_loss": 0.5382067,
      "learning_rate": 3.374286956009605e-06,
      "loss": 0.5601145,
      "num_input_tokens_seen": 50151190,
      "step": 2338,
      "time_per_iteration": 2.5844810009002686
    },
    {
      "auxiliary_loss_clip": 0.01192518,
      "auxiliary_loss_mlp": 0.0103878,
      "balance_loss_clip": 1.06561363,
      "balance_loss_mlp": 1.02943969,
      "epoch": 0.2812481212048338,
      "flos": 12823629482880.0,
      "grad_norm": 2.108964411793755,
      "language_loss": 0.75374472,
      "learning_rate": 3.3737209119471405e-06,
      "loss": 0.77605766,
      "num_input_tokens_seen": 50167700,
      "step": 2339,
      "time_per_iteration": 2.497767210006714
    },
    {
      "auxiliary_loss_clip": 0.01195843,
      "auxiliary_loss_mlp": 0.01037108,
      "balance_loss_clip": 1.06292367,
      "balance_loss_mlp": 1.02721965,
      "epoch": 0.28136836409547283,
      "flos": 15633064765440.0,
      "grad_norm": 2.3740089726549316,
      "language_loss": 0.63649601,
      "learning_rate": 3.373154659492306e-06,
      "loss": 0.65882552,
      "num_input_tokens_seen": 50185840,
      "step": 2340,
      "time_per_iteration": 2.506227493286133
    },
    {
      "auxiliary_loss_clip": 0.01179232,
      "auxiliary_loss_mlp": 0.01030277,
      "balance_loss_clip": 1.06135356,
      "balance_loss_mlp": 1.02130616,
      "epoch": 0.28148860698611194,
      "flos": 19933726106880.0,
      "grad_norm": 2.6452116813264626,
      "language_loss": 0.85453004,
      "learning_rate": 3.3725881987310016e-06,
      "loss": 0.87662512,
      "num_input_tokens_seen": 50203375,
      "step": 2341,
      "time_per_iteration": 2.553468942642212
    },
    {
      "auxiliary_loss_clip": 0.01173539,
      "auxiliary_loss_mlp": 0.01033578,
      "balance_loss_clip": 1.05762947,
      "balance_loss_mlp": 1.02451849,
      "epoch": 0.28160884987675106,
      "flos": 17457219008640.0,
      "grad_norm": 1.7948437171279177,
      "language_loss": 0.87716126,
      "learning_rate": 3.372021529749159e-06,
      "loss": 0.89923245,
      "num_input_tokens_seen": 50222435,
      "step": 2342,
      "time_per_iteration": 2.563180685043335
    },
    {
      "auxiliary_loss_clip": 0.01139057,
      "auxiliary_loss_mlp": 0.01027116,
      "balance_loss_clip": 1.05819237,
      "balance_loss_mlp": 1.01814556,
      "epoch": 0.2817290927673901,
      "flos": 16834743290880.0,
      "grad_norm": 1.9741260920870247,
      "language_loss": 0.92215884,
      "learning_rate": 3.3714546526327405e-06,
      "loss": 0.94382048,
      "num_input_tokens_seen": 50240435,
      "step": 2343,
      "time_per_iteration": 2.679696798324585
    },
    {
      "auxiliary_loss_clip": 0.01166002,
      "auxiliary_loss_mlp": 0.01038705,
      "balance_loss_clip": 1.05743122,
      "balance_loss_mlp": 1.02863216,
      "epoch": 0.2818493356580292,
      "flos": 15414081500160.0,
      "grad_norm": 2.4276310243663763,
      "language_loss": 0.87668025,
      "learning_rate": 3.3708875674677423e-06,
      "loss": 0.8987273,
      "num_input_tokens_seen": 50258410,
      "step": 2344,
      "time_per_iteration": 2.62688946723938
    },
    {
      "auxiliary_loss_clip": 0.01184987,
      "auxiliary_loss_mlp": 0.01031287,
      "balance_loss_clip": 1.06151104,
      "balance_loss_mlp": 1.02087379,
      "epoch": 0.28196957854866833,
      "flos": 20412451595520.0,
      "grad_norm": 2.0105405331850967,
      "language_loss": 0.8328501,
      "learning_rate": 3.37032027434019e-06,
      "loss": 0.85501289,
      "num_input_tokens_seen": 50277930,
      "step": 2345,
      "time_per_iteration": 2.6030330657958984
    },
    {
      "auxiliary_loss_clip": 0.01199203,
      "auxiliary_loss_mlp": 0.01049297,
      "balance_loss_clip": 1.06237841,
      "balance_loss_mlp": 1.03896761,
      "epoch": 0.2820898214393074,
      "flos": 19973120348160.0,
      "grad_norm": 1.6571212460476064,
      "language_loss": 0.82804948,
      "learning_rate": 3.369752773336141e-06,
      "loss": 0.8505345,
      "num_input_tokens_seen": 50297410,
      "step": 2346,
      "time_per_iteration": 2.5261080265045166
    },
    {
      "auxiliary_loss_clip": 0.01176417,
      "auxiliary_loss_mlp": 0.0103417,
      "balance_loss_clip": 1.05879474,
      "balance_loss_mlp": 1.02444863,
      "epoch": 0.2822100643299465,
      "flos": 22528308188160.0,
      "grad_norm": 1.5752937694689921,
      "language_loss": 0.77824962,
      "learning_rate": 3.3691850645416864e-06,
      "loss": 0.80035549,
      "num_input_tokens_seen": 50317120,
      "step": 2347,
      "time_per_iteration": 2.5922775268554688
    },
    {
      "auxiliary_loss_clip": 0.01193897,
      "auxiliary_loss_mlp": 0.01038396,
      "balance_loss_clip": 1.06050467,
      "balance_loss_mlp": 1.02915764,
      "epoch": 0.2823303072205856,
      "flos": 11546682007680.0,
      "grad_norm": 4.3372757036243215,
      "language_loss": 0.83092844,
      "learning_rate": 3.368617148042945e-06,
      "loss": 0.8532514,
      "num_input_tokens_seen": 50334790,
      "step": 2348,
      "time_per_iteration": 2.4736721515655518
    },
    {
      "auxiliary_loss_clip": 0.01168754,
      "auxiliary_loss_mlp": 0.01037254,
      "balance_loss_clip": 1.05486274,
      "balance_loss_mlp": 1.02809334,
      "epoch": 0.28245055011122466,
      "flos": 18259894281600.0,
      "grad_norm": 2.236359607953175,
      "language_loss": 0.8492946,
      "learning_rate": 3.368049023926071e-06,
      "loss": 0.87135464,
      "num_input_tokens_seen": 50353785,
      "step": 2349,
      "time_per_iteration": 2.5265262126922607
    },
    {
      "auxiliary_loss_clip": 0.01187788,
      "auxiliary_loss_mlp": 0.01034188,
      "balance_loss_clip": 1.0608263,
      "balance_loss_mlp": 1.0255456,
      "epoch": 0.2825707930018638,
      "flos": 24608110504320.0,
      "grad_norm": 1.601790612410944,
      "language_loss": 0.84018081,
      "learning_rate": 3.3674806922772476e-06,
      "loss": 0.86240053,
      "num_input_tokens_seen": 50374670,
      "step": 2350,
      "time_per_iteration": 2.560978889465332
    },
    {
      "auxiliary_loss_clip": 0.0116557,
      "auxiliary_loss_mlp": 0.0103794,
      "balance_loss_clip": 1.05657578,
      "balance_loss_mlp": 1.02911222,
      "epoch": 0.28269103589250283,
      "flos": 25226994862080.0,
      "grad_norm": 2.1271662017492545,
      "language_loss": 0.75183511,
      "learning_rate": 3.3669121531826904e-06,
      "loss": 0.77387017,
      "num_input_tokens_seen": 50395650,
      "step": 2351,
      "time_per_iteration": 3.417557954788208
    },
    {
      "auxiliary_loss_clip": 0.01154916,
      "auxiliary_loss_mlp": 0.01036231,
      "balance_loss_clip": 1.05951262,
      "balance_loss_mlp": 1.02664709,
      "epoch": 0.28281127878314194,
      "flos": 19281552819840.0,
      "grad_norm": 2.2179443730072914,
      "language_loss": 0.83052421,
      "learning_rate": 3.366343406728647e-06,
      "loss": 0.85243565,
      "num_input_tokens_seen": 50415100,
      "step": 2352,
      "time_per_iteration": 2.5498199462890625
    },
    {
      "auxiliary_loss_clip": 0.0117959,
      "auxiliary_loss_mlp": 0.01034225,
      "balance_loss_clip": 1.05512118,
      "balance_loss_mlp": 1.02495015,
      "epoch": 0.28293152167378105,
      "flos": 23878405710720.0,
      "grad_norm": 1.8212585966856105,
      "language_loss": 0.68572551,
      "learning_rate": 3.3657744530013946e-06,
      "loss": 0.70786363,
      "num_input_tokens_seen": 50434335,
      "step": 2353,
      "time_per_iteration": 2.5268144607543945
    },
    {
      "auxiliary_loss_clip": 0.01195889,
      "auxiliary_loss_mlp": 0.01041043,
      "balance_loss_clip": 1.06276,
      "balance_loss_mlp": 1.03218603,
      "epoch": 0.2830517645644201,
      "flos": 43866965928960.0,
      "grad_norm": 1.854523478834294,
      "language_loss": 0.71479404,
      "learning_rate": 3.3652052920872437e-06,
      "loss": 0.73716342,
      "num_input_tokens_seen": 50457200,
      "step": 2354,
      "time_per_iteration": 2.704134225845337
    },
    {
      "auxiliary_loss_clip": 0.011781,
      "auxiliary_loss_mlp": 0.01028989,
      "balance_loss_clip": 1.05972373,
      "balance_loss_mlp": 1.01967824,
      "epoch": 0.2831720074550592,
      "flos": 26651750803200.0,
      "grad_norm": 2.1982389545300576,
      "language_loss": 0.85546333,
      "learning_rate": 3.3646359240725355e-06,
      "loss": 0.87753421,
      "num_input_tokens_seen": 50476390,
      "step": 2355,
      "time_per_iteration": 2.5805399417877197
    },
    {
      "auxiliary_loss_clip": 0.01184779,
      "auxiliary_loss_mlp": 0.00763648,
      "balance_loss_clip": 1.05962467,
      "balance_loss_mlp": 1.00037503,
      "epoch": 0.2832922503456983,
      "flos": 31029979564800.0,
      "grad_norm": 1.8878251582563985,
      "language_loss": 0.67490673,
      "learning_rate": 3.364066349043643e-06,
      "loss": 0.69439101,
      "num_input_tokens_seen": 50497595,
      "step": 2356,
      "time_per_iteration": 2.59031343460083
    },
    {
      "auxiliary_loss_clip": 0.01169022,
      "auxiliary_loss_mlp": 0.01027325,
      "balance_loss_clip": 1.05657482,
      "balance_loss_mlp": 1.01900434,
      "epoch": 0.2834124932363374,
      "flos": 20405699838720.0,
      "grad_norm": 1.7408698155109856,
      "language_loss": 0.8215242,
      "learning_rate": 3.363496567086969e-06,
      "loss": 0.84348768,
      "num_input_tokens_seen": 50514690,
      "step": 2357,
      "time_per_iteration": 2.5392723083496094
    },
    {
      "auxiliary_loss_clip": 0.01202552,
      "auxiliary_loss_mlp": 0.01040598,
      "balance_loss_clip": 1.06109214,
      "balance_loss_mlp": 1.03141856,
      "epoch": 0.2835327361269765,
      "flos": 39384848056320.0,
      "grad_norm": 1.8409664747447745,
      "language_loss": 0.7579217,
      "learning_rate": 3.3629265782889506e-06,
      "loss": 0.78035325,
      "num_input_tokens_seen": 50536515,
      "step": 2358,
      "time_per_iteration": 3.4181575775146484
    },
    {
      "auxiliary_loss_clip": 0.01154553,
      "auxiliary_loss_mlp": 0.01039057,
      "balance_loss_clip": 1.05246568,
      "balance_loss_mlp": 1.02992535,
      "epoch": 0.2836529790176156,
      "flos": 30261598801920.0,
      "grad_norm": 1.758168405456009,
      "language_loss": 0.71616125,
      "learning_rate": 3.362356382736054e-06,
      "loss": 0.73809731,
      "num_input_tokens_seen": 50557120,
      "step": 2359,
      "time_per_iteration": 2.641368865966797
    },
    {
      "auxiliary_loss_clip": 0.01157354,
      "auxiliary_loss_mlp": 0.01030335,
      "balance_loss_clip": 1.05092144,
      "balance_loss_mlp": 1.02192473,
      "epoch": 0.28377322190825466,
      "flos": 12677796264960.0,
      "grad_norm": 1.9150518708386102,
      "language_loss": 0.90632254,
      "learning_rate": 3.361785980514777e-06,
      "loss": 0.92819941,
      "num_input_tokens_seen": 50573320,
      "step": 2360,
      "time_per_iteration": 3.309847593307495
    },
    {
      "auxiliary_loss_clip": 0.01126675,
      "auxiliary_loss_mlp": 0.01032611,
      "balance_loss_clip": 1.05270445,
      "balance_loss_mlp": 1.02299106,
      "epoch": 0.28389346479889377,
      "flos": 18296666830080.0,
      "grad_norm": 2.2472897167114882,
      "language_loss": 0.76732624,
      "learning_rate": 3.361215371711649e-06,
      "loss": 0.78891909,
      "num_input_tokens_seen": 50592415,
      "step": 2361,
      "time_per_iteration": 2.6080105304718018
    },
    {
      "auxiliary_loss_clip": 0.01150617,
      "auxiliary_loss_mlp": 0.01036189,
      "balance_loss_clip": 1.05474639,
      "balance_loss_mlp": 1.02740288,
      "epoch": 0.2840137076895329,
      "flos": 20406992728320.0,
      "grad_norm": 1.727049474427333,
      "language_loss": 0.83204579,
      "learning_rate": 3.3606445564132326e-06,
      "loss": 0.85391378,
      "num_input_tokens_seen": 50609710,
      "step": 2362,
      "time_per_iteration": 2.562739372253418
    },
    {
      "auxiliary_loss_clip": 0.0120641,
      "auxiliary_loss_mlp": 0.00762966,
      "balance_loss_clip": 1.06317496,
      "balance_loss_mlp": 1.00036168,
      "epoch": 0.28413395058017193,
      "flos": 20048030161920.0,
      "grad_norm": 2.3058760286327233,
      "language_loss": 0.82320356,
      "learning_rate": 3.360073534706118e-06,
      "loss": 0.8428973,
      "num_input_tokens_seen": 50626865,
      "step": 2363,
      "time_per_iteration": 2.4814774990081787
    },
    {
      "auxiliary_loss_clip": 0.01175418,
      "auxiliary_loss_mlp": 0.010362,
      "balance_loss_clip": 1.05887008,
      "balance_loss_mlp": 1.02600169,
      "epoch": 0.28425419347081105,
      "flos": 37663613256960.0,
      "grad_norm": 1.9772328012963118,
      "language_loss": 0.75779009,
      "learning_rate": 3.35950230667693e-06,
      "loss": 0.77990627,
      "num_input_tokens_seen": 50648560,
      "step": 2364,
      "time_per_iteration": 2.6726913452148438
    },
    {
      "auxiliary_loss_clip": 0.01188882,
      "auxiliary_loss_mlp": 0.01036656,
      "balance_loss_clip": 1.05864322,
      "balance_loss_mlp": 1.02774477,
      "epoch": 0.28437443636145016,
      "flos": 13845072539520.0,
      "grad_norm": 2.2543759259073575,
      "language_loss": 0.86099052,
      "learning_rate": 3.358930872412323e-06,
      "loss": 0.88324583,
      "num_input_tokens_seen": 50665725,
      "step": 2365,
      "time_per_iteration": 2.47017502784729
    },
    {
      "auxiliary_loss_clip": 0.01185447,
      "auxiliary_loss_mlp": 0.0103826,
      "balance_loss_clip": 1.05976987,
      "balance_loss_mlp": 1.02911663,
      "epoch": 0.2844946792520892,
      "flos": 22747794243840.0,
      "grad_norm": 1.5921565102871085,
      "language_loss": 0.80955541,
      "learning_rate": 3.3583592319989825e-06,
      "loss": 0.83179247,
      "num_input_tokens_seen": 50685095,
      "step": 2366,
      "time_per_iteration": 2.5105650424957275
    },
    {
      "auxiliary_loss_clip": 0.01194523,
      "auxiliary_loss_mlp": 0.01034731,
      "balance_loss_clip": 1.06066811,
      "balance_loss_mlp": 1.02436018,
      "epoch": 0.2846149221427283,
      "flos": 32415987709440.0,
      "grad_norm": 2.1606507421283534,
      "language_loss": 0.68655193,
      "learning_rate": 3.357787385523627e-06,
      "loss": 0.70884442,
      "num_input_tokens_seen": 50706500,
      "step": 2367,
      "time_per_iteration": 2.5969314575195312
    },
    {
      "auxiliary_loss_clip": 0.01138095,
      "auxiliary_loss_mlp": 0.01039012,
      "balance_loss_clip": 1.05304301,
      "balance_loss_mlp": 1.02979755,
      "epoch": 0.2847351650333674,
      "flos": 28475976873600.0,
      "grad_norm": 2.8961190017883447,
      "language_loss": 0.82635599,
      "learning_rate": 3.3572153330730048e-06,
      "loss": 0.84812713,
      "num_input_tokens_seen": 50727595,
      "step": 2368,
      "time_per_iteration": 2.687941789627075
    },
    {
      "auxiliary_loss_clip": 0.01062609,
      "auxiliary_loss_mlp": 0.01004592,
      "balance_loss_clip": 1.02489674,
      "balance_loss_mlp": 1.00268495,
      "epoch": 0.2848554079240065,
      "flos": 55753399704960.0,
      "grad_norm": 0.832670656776317,
      "language_loss": 0.64721942,
      "learning_rate": 3.3566430747338956e-06,
      "loss": 0.66789144,
      "num_input_tokens_seen": 50782800,
      "step": 2369,
      "time_per_iteration": 3.0093696117401123
    },
    {
      "auxiliary_loss_clip": 0.01189633,
      "auxiliary_loss_mlp": 0.0103662,
      "balance_loss_clip": 1.05733109,
      "balance_loss_mlp": 1.02780485,
      "epoch": 0.2849756508146456,
      "flos": 11836875985920.0,
      "grad_norm": 1.9115927239543933,
      "language_loss": 0.86490762,
      "learning_rate": 3.35607061059311e-06,
      "loss": 0.8871702,
      "num_input_tokens_seen": 50797730,
      "step": 2370,
      "time_per_iteration": 2.4733998775482178
    },
    {
      "auxiliary_loss_clip": 0.01198364,
      "auxiliary_loss_mlp": 0.01029106,
      "balance_loss_clip": 1.05984664,
      "balance_loss_mlp": 1.01975393,
      "epoch": 0.28509589370528465,
      "flos": 25155209531520.0,
      "grad_norm": 1.7484409438440027,
      "language_loss": 0.74846464,
      "learning_rate": 3.3554979407374917e-06,
      "loss": 0.77073932,
      "num_input_tokens_seen": 50819840,
      "step": 2371,
      "time_per_iteration": 2.5399396419525146
    },
    {
      "auxiliary_loss_clip": 0.0118602,
      "auxiliary_loss_mlp": 0.01036131,
      "balance_loss_clip": 1.05640125,
      "balance_loss_mlp": 1.02691638,
      "epoch": 0.28521613659592376,
      "flos": 19974808287360.0,
      "grad_norm": 1.4832880198675957,
      "language_loss": 0.73315066,
      "learning_rate": 3.3549250652539134e-06,
      "loss": 0.75537217,
      "num_input_tokens_seen": 50838935,
      "step": 2372,
      "time_per_iteration": 2.510124921798706
    },
    {
      "auxiliary_loss_clip": 0.0117129,
      "auxiliary_loss_mlp": 0.01035296,
      "balance_loss_clip": 1.05431974,
      "balance_loss_mlp": 1.02597356,
      "epoch": 0.2853363794865629,
      "flos": 23367971491200.0,
      "grad_norm": 2.3378263483364305,
      "language_loss": 0.8161844,
      "learning_rate": 3.3543519842292794e-06,
      "loss": 0.83825022,
      "num_input_tokens_seen": 50858590,
      "step": 2373,
      "time_per_iteration": 2.5563879013061523
    },
    {
      "auxiliary_loss_clip": 0.01202957,
      "auxiliary_loss_mlp": 0.00762767,
      "balance_loss_clip": 1.06213975,
      "balance_loss_mlp": 1.00037646,
      "epoch": 0.28545662237720193,
      "flos": 19861940776320.0,
      "grad_norm": 1.938395407520442,
      "language_loss": 0.83594716,
      "learning_rate": 3.353778697750527e-06,
      "loss": 0.85560429,
      "num_input_tokens_seen": 50876995,
      "step": 2374,
      "time_per_iteration": 2.470747709274292
    },
    {
      "auxiliary_loss_clip": 0.01168328,
      "auxiliary_loss_mlp": 0.01035845,
      "balance_loss_clip": 1.05692816,
      "balance_loss_mlp": 1.02667761,
      "epoch": 0.28557686526784104,
      "flos": 23879016241920.0,
      "grad_norm": 1.547567901401587,
      "language_loss": 0.89322853,
      "learning_rate": 3.353205205904622e-06,
      "loss": 0.91527021,
      "num_input_tokens_seen": 50896105,
      "step": 2375,
      "time_per_iteration": 2.5622775554656982
    },
    {
      "auxiliary_loss_clip": 0.01171637,
      "auxiliary_loss_mlp": 0.01030874,
      "balance_loss_clip": 1.05634058,
      "balance_loss_mlp": 1.02152228,
      "epoch": 0.28569710815848015,
      "flos": 44890384233600.0,
      "grad_norm": 2.392648938616002,
      "language_loss": 0.71414751,
      "learning_rate": 3.3526315087785637e-06,
      "loss": 0.73617262,
      "num_input_tokens_seen": 50917220,
      "step": 2376,
      "time_per_iteration": 2.7424986362457275
    },
    {
      "auxiliary_loss_clip": 0.01126984,
      "auxiliary_loss_mlp": 0.01026086,
      "balance_loss_clip": 1.05321229,
      "balance_loss_mlp": 1.01690638,
      "epoch": 0.2858173510491192,
      "flos": 26829759628800.0,
      "grad_norm": 1.9162436880566267,
      "language_loss": 0.80860561,
      "learning_rate": 3.3520576064593805e-06,
      "loss": 0.8301363,
      "num_input_tokens_seen": 50937175,
      "step": 2377,
      "time_per_iteration": 3.4274961948394775
    },
    {
      "auxiliary_loss_clip": 0.01195215,
      "auxiliary_loss_mlp": 0.01038367,
      "balance_loss_clip": 1.06215811,
      "balance_loss_mlp": 1.02896094,
      "epoch": 0.2859375939397583,
      "flos": 23148916398720.0,
      "grad_norm": 1.5007515788121881,
      "language_loss": 0.81920648,
      "learning_rate": 3.3514834990341337e-06,
      "loss": 0.84154236,
      "num_input_tokens_seen": 50957500,
      "step": 2378,
      "time_per_iteration": 2.523139238357544
    },
    {
      "auxiliary_loss_clip": 0.01180616,
      "auxiliary_loss_mlp": 0.0103072,
      "balance_loss_clip": 1.05916286,
      "balance_loss_mlp": 1.02169561,
      "epoch": 0.2860578368303974,
      "flos": 12129799397760.0,
      "grad_norm": 2.3154240960621304,
      "language_loss": 0.92843497,
      "learning_rate": 3.3509091865899144e-06,
      "loss": 0.95054829,
      "num_input_tokens_seen": 50972690,
      "step": 2379,
      "time_per_iteration": 2.5333428382873535
    },
    {
      "auxiliary_loss_clip": 0.01201899,
      "auxiliary_loss_mlp": 0.01038295,
      "balance_loss_clip": 1.05971718,
      "balance_loss_mlp": 1.02817988,
      "epoch": 0.2861780797210365,
      "flos": 19938035738880.0,
      "grad_norm": 1.8821793016245674,
      "language_loss": 0.7055462,
      "learning_rate": 3.350334669213846e-06,
      "loss": 0.72794819,
      "num_input_tokens_seen": 50990095,
      "step": 2380,
      "time_per_iteration": 2.471461057662964
    },
    {
      "auxiliary_loss_clip": 0.01188812,
      "auxiliary_loss_mlp": 0.01032562,
      "balance_loss_clip": 1.06231141,
      "balance_loss_mlp": 1.02362692,
      "epoch": 0.2862983226116756,
      "flos": 27563127609600.0,
      "grad_norm": 2.165408972904066,
      "language_loss": 0.75717354,
      "learning_rate": 3.3497599469930816e-06,
      "loss": 0.77938724,
      "num_input_tokens_seen": 51008305,
      "step": 2381,
      "time_per_iteration": 2.532580852508545
    },
    {
      "auxiliary_loss_clip": 0.01203972,
      "auxiliary_loss_mlp": 0.01029673,
      "balance_loss_clip": 1.06198025,
      "balance_loss_mlp": 1.01948667,
      "epoch": 0.28641856550231465,
      "flos": 22053964158720.0,
      "grad_norm": 2.400974629201936,
      "language_loss": 0.82978415,
      "learning_rate": 3.349185020014807e-06,
      "loss": 0.85212064,
      "num_input_tokens_seen": 51025570,
      "step": 2382,
      "time_per_iteration": 2.466705799102783
    },
    {
      "auxiliary_loss_clip": 0.0119085,
      "auxiliary_loss_mlp": 0.01035732,
      "balance_loss_clip": 1.05953276,
      "balance_loss_mlp": 1.02580822,
      "epoch": 0.28653880839295376,
      "flos": 22378775869440.0,
      "grad_norm": 1.9465267917010782,
      "language_loss": 0.74525058,
      "learning_rate": 3.348609888366237e-06,
      "loss": 0.76751637,
      "num_input_tokens_seen": 51044585,
      "step": 2383,
      "time_per_iteration": 2.5319015979766846
    },
    {
      "auxiliary_loss_clip": 0.01124293,
      "auxiliary_loss_mlp": 0.01028558,
      "balance_loss_clip": 1.05058908,
      "balance_loss_mlp": 1.01952195,
      "epoch": 0.28665905128359287,
      "flos": 23367971491200.0,
      "grad_norm": 2.0865753703418295,
      "language_loss": 0.62901264,
      "learning_rate": 3.348034552134619e-06,
      "loss": 0.65054113,
      "num_input_tokens_seen": 51063990,
      "step": 2384,
      "time_per_iteration": 4.19303297996521
    },
    {
      "auxiliary_loss_clip": 0.01138827,
      "auxiliary_loss_mlp": 0.01036387,
      "balance_loss_clip": 1.0536809,
      "balance_loss_mlp": 1.02671933,
      "epoch": 0.2867792941742319,
      "flos": 20881695893760.0,
      "grad_norm": 3.3574807857846376,
      "language_loss": 0.84212196,
      "learning_rate": 3.3474590114072316e-06,
      "loss": 0.86387414,
      "num_input_tokens_seen": 51081990,
      "step": 2385,
      "time_per_iteration": 2.5792152881622314
    },
    {
      "auxiliary_loss_clip": 0.01157049,
      "auxiliary_loss_mlp": 0.010335,
      "balance_loss_clip": 1.05846095,
      "balance_loss_mlp": 1.02374887,
      "epoch": 0.28689953706487104,
      "flos": 20664005518080.0,
      "grad_norm": 1.9521252083613763,
      "language_loss": 0.82797921,
      "learning_rate": 3.3468832662713836e-06,
      "loss": 0.84988463,
      "num_input_tokens_seen": 51100235,
      "step": 2386,
      "time_per_iteration": 3.395561695098877
    },
    {
      "auxiliary_loss_clip": 0.0115961,
      "auxiliary_loss_mlp": 0.01035869,
      "balance_loss_clip": 1.05763149,
      "balance_loss_mlp": 1.02719617,
      "epoch": 0.28701977995551015,
      "flos": 12675533708160.0,
      "grad_norm": 2.4467970638605743,
      "language_loss": 0.8390286,
      "learning_rate": 3.346307316814415e-06,
      "loss": 0.86098337,
      "num_input_tokens_seen": 51115405,
      "step": 2387,
      "time_per_iteration": 2.5318331718444824
    },
    {
      "auxiliary_loss_clip": 0.01184453,
      "auxiliary_loss_mlp": 0.01034529,
      "balance_loss_clip": 1.05925167,
      "balance_loss_mlp": 1.02499795,
      "epoch": 0.2871400228461492,
      "flos": 21252366293760.0,
      "grad_norm": 1.981796763167291,
      "language_loss": 0.75437748,
      "learning_rate": 3.3457311631236965e-06,
      "loss": 0.77656734,
      "num_input_tokens_seen": 51136390,
      "step": 2388,
      "time_per_iteration": 2.519674301147461
    },
    {
      "auxiliary_loss_clip": 0.01159724,
      "auxiliary_loss_mlp": 0.01036267,
      "balance_loss_clip": 1.05405605,
      "balance_loss_mlp": 1.02671838,
      "epoch": 0.2872602657367883,
      "flos": 25119262995840.0,
      "grad_norm": 1.6585515317572712,
      "language_loss": 0.84537327,
      "learning_rate": 3.345154805286631e-06,
      "loss": 0.86733323,
      "num_input_tokens_seen": 51156650,
      "step": 2389,
      "time_per_iteration": 2.5627686977386475
    },
    {
      "auxiliary_loss_clip": 0.01181423,
      "auxiliary_loss_mlp": 0.01047823,
      "balance_loss_clip": 1.05673647,
      "balance_loss_mlp": 1.0378989,
      "epoch": 0.2873805086274274,
      "flos": 16646606830080.0,
      "grad_norm": 3.229318789910953,
      "language_loss": 0.7604515,
      "learning_rate": 3.344578243390651e-06,
      "loss": 0.78274405,
      "num_input_tokens_seen": 51172210,
      "step": 2390,
      "time_per_iteration": 2.4615721702575684
    },
    {
      "auxiliary_loss_clip": 0.01172161,
      "auxiliary_loss_mlp": 0.01031914,
      "balance_loss_clip": 1.05871987,
      "balance_loss_mlp": 1.02263975,
      "epoch": 0.2875007515180665,
      "flos": 17420123237760.0,
      "grad_norm": 2.3616347103555557,
      "language_loss": 0.78692114,
      "learning_rate": 3.3440014775232206e-06,
      "loss": 0.80896187,
      "num_input_tokens_seen": 51190265,
      "step": 2391,
      "time_per_iteration": 2.5250093936920166
    },
    {
      "auxiliary_loss_clip": 0.01162949,
      "auxiliary_loss_mlp": 0.01031933,
      "balance_loss_clip": 1.05780399,
      "balance_loss_mlp": 1.02329612,
      "epoch": 0.2876209944087056,
      "flos": 23434190213760.0,
      "grad_norm": 2.0477771162669205,
      "language_loss": 0.71003515,
      "learning_rate": 3.343424507771834e-06,
      "loss": 0.73198402,
      "num_input_tokens_seen": 51208475,
      "step": 2392,
      "time_per_iteration": 2.5756216049194336
    },
    {
      "auxiliary_loss_clip": 0.011604,
      "auxiliary_loss_mlp": 0.01035492,
      "balance_loss_clip": 1.05760288,
      "balance_loss_mlp": 1.02659321,
      "epoch": 0.2877412372993447,
      "flos": 13735509079680.0,
      "grad_norm": 1.727992258144669,
      "language_loss": 0.86510241,
      "learning_rate": 3.342847334224018e-06,
      "loss": 0.8870613,
      "num_input_tokens_seen": 51225875,
      "step": 2393,
      "time_per_iteration": 2.5388665199279785
    },
    {
      "auxiliary_loss_clip": 0.01083563,
      "auxiliary_loss_mlp": 0.01007444,
      "balance_loss_clip": 1.02665806,
      "balance_loss_mlp": 1.00554848,
      "epoch": 0.28786148018998375,
      "flos": 58079695104000.0,
      "grad_norm": 0.9435571300786746,
      "language_loss": 0.62343264,
      "learning_rate": 3.342269956967329e-06,
      "loss": 0.64434278,
      "num_input_tokens_seen": 51287780,
      "step": 2394,
      "time_per_iteration": 3.1629865169525146
    },
    {
      "auxiliary_loss_clip": 0.01188591,
      "auxiliary_loss_mlp": 0.01041098,
      "balance_loss_clip": 1.05799174,
      "balance_loss_mlp": 1.03106642,
      "epoch": 0.28798172308062286,
      "flos": 23435052140160.0,
      "grad_norm": 3.0317380246266143,
      "language_loss": 0.7150296,
      "learning_rate": 3.341692376089355e-06,
      "loss": 0.7373265,
      "num_input_tokens_seen": 51303335,
      "step": 2395,
      "time_per_iteration": 2.5881645679473877
    },
    {
      "auxiliary_loss_clip": 0.01184814,
      "auxiliary_loss_mlp": 0.01033841,
      "balance_loss_clip": 1.05934167,
      "balance_loss_mlp": 1.02463818,
      "epoch": 0.288101965971262,
      "flos": 25110033200640.0,
      "grad_norm": 4.542591623399206,
      "language_loss": 0.83471435,
      "learning_rate": 3.3411145916777146e-06,
      "loss": 0.85690093,
      "num_input_tokens_seen": 51317495,
      "step": 2396,
      "time_per_iteration": 2.528890609741211
    },
    {
      "auxiliary_loss_clip": 0.01165468,
      "auxiliary_loss_mlp": 0.0103917,
      "balance_loss_clip": 1.05615056,
      "balance_loss_mlp": 1.02934766,
      "epoch": 0.28822220886190103,
      "flos": 16252559654400.0,
      "grad_norm": 2.991114731114391,
      "language_loss": 0.90647137,
      "learning_rate": 3.3405366038200566e-06,
      "loss": 0.92851782,
      "num_input_tokens_seen": 51336430,
      "step": 2397,
      "time_per_iteration": 2.529747247695923
    },
    {
      "auxiliary_loss_clip": 0.01176874,
      "auxiliary_loss_mlp": 0.01036416,
      "balance_loss_clip": 1.06057322,
      "balance_loss_mlp": 1.02661109,
      "epoch": 0.28834245175254014,
      "flos": 24535642815360.0,
      "grad_norm": 2.3936783733360767,
      "language_loss": 0.85164142,
      "learning_rate": 3.3399584126040617e-06,
      "loss": 0.87377429,
      "num_input_tokens_seen": 51355930,
      "step": 2398,
      "time_per_iteration": 2.568298578262329
    },
    {
      "auxiliary_loss_clip": 0.01201693,
      "auxiliary_loss_mlp": 0.00762693,
      "balance_loss_clip": 1.06080699,
      "balance_loss_mlp": 1.00033689,
      "epoch": 0.2884626946431792,
      "flos": 24571445696640.0,
      "grad_norm": 3.489197519283409,
      "language_loss": 0.90500599,
      "learning_rate": 3.339380018117441e-06,
      "loss": 0.92464983,
      "num_input_tokens_seen": 51376765,
      "step": 2399,
      "time_per_iteration": 2.5237443447113037
    },
    {
      "auxiliary_loss_clip": 0.01187878,
      "auxiliary_loss_mlp": 0.01030123,
      "balance_loss_clip": 1.0638566,
      "balance_loss_mlp": 1.02148676,
      "epoch": 0.2885829375338183,
      "flos": 16544657053440.0,
      "grad_norm": 2.3679014595306516,
      "language_loss": 0.78726918,
      "learning_rate": 3.3388014204479366e-06,
      "loss": 0.8094492,
      "num_input_tokens_seen": 51394570,
      "step": 2400,
      "time_per_iteration": 2.4822442531585693
    },
    {
      "auxiliary_loss_clip": 0.0120252,
      "auxiliary_loss_mlp": 0.01037009,
      "balance_loss_clip": 1.06066263,
      "balance_loss_mlp": 1.02738905,
      "epoch": 0.2887031804244574,
      "flos": 24061226958720.0,
      "grad_norm": 1.96814232368608,
      "language_loss": 0.9164784,
      "learning_rate": 3.338222619683321e-06,
      "loss": 0.93887377,
      "num_input_tokens_seen": 51414535,
      "step": 2401,
      "time_per_iteration": 2.5004332065582275
    },
    {
      "auxiliary_loss_clip": 0.01174755,
      "auxiliary_loss_mlp": 0.01035832,
      "balance_loss_clip": 1.05746579,
      "balance_loss_mlp": 1.0267539,
      "epoch": 0.2888234233150965,
      "flos": 23330696152320.0,
      "grad_norm": 2.4464112861387064,
      "language_loss": 0.72994709,
      "learning_rate": 3.337643615911398e-06,
      "loss": 0.75205296,
      "num_input_tokens_seen": 51434160,
      "step": 2402,
      "time_per_iteration": 2.542466163635254
    },
    {
      "auxiliary_loss_clip": 0.0118521,
      "auxiliary_loss_mlp": 0.01039341,
      "balance_loss_clip": 1.0557189,
      "balance_loss_mlp": 1.02969742,
      "epoch": 0.2889436662057356,
      "flos": 22272767856000.0,
      "grad_norm": 1.9784324965254685,
      "language_loss": 0.78910029,
      "learning_rate": 3.3370644092200026e-06,
      "loss": 0.81134582,
      "num_input_tokens_seen": 51451435,
      "step": 2403,
      "time_per_iteration": 3.314467668533325
    },
    {
      "auxiliary_loss_clip": 0.01144274,
      "auxiliary_loss_mlp": 0.01033493,
      "balance_loss_clip": 1.0510242,
      "balance_loss_mlp": 1.02460027,
      "epoch": 0.2890639090963747,
      "flos": 21616931381760.0,
      "grad_norm": 4.122704915882568,
      "language_loss": 0.78268206,
      "learning_rate": 3.3364849996969985e-06,
      "loss": 0.80445969,
      "num_input_tokens_seen": 51471455,
      "step": 2404,
      "time_per_iteration": 2.568740129470825
    },
    {
      "auxiliary_loss_clip": 0.01184798,
      "auxiliary_loss_mlp": 0.01035061,
      "balance_loss_clip": 1.05916619,
      "balance_loss_mlp": 1.02557826,
      "epoch": 0.28918415198701375,
      "flos": 28585540333440.0,
      "grad_norm": 1.900027722799323,
      "language_loss": 0.85315907,
      "learning_rate": 3.335905387430283e-06,
      "loss": 0.87535769,
      "num_input_tokens_seen": 51492890,
      "step": 2405,
      "time_per_iteration": 2.572610855102539
    },
    {
      "auxiliary_loss_clip": 0.01175205,
      "auxiliary_loss_mlp": 0.01034917,
      "balance_loss_clip": 1.05535316,
      "balance_loss_mlp": 1.02660751,
      "epoch": 0.28930439487765286,
      "flos": 21944688007680.0,
      "grad_norm": 1.9446456717219711,
      "language_loss": 0.83019471,
      "learning_rate": 3.335325572507782e-06,
      "loss": 0.85229599,
      "num_input_tokens_seen": 51513390,
      "step": 2406,
      "time_per_iteration": 2.565513849258423
    },
    {
      "auxiliary_loss_clip": 0.01204574,
      "auxiliary_loss_mlp": 0.00762784,
      "balance_loss_clip": 1.06504011,
      "balance_loss_mlp": 1.00036597,
      "epoch": 0.28942463776829197,
      "flos": 19281911955840.0,
      "grad_norm": 1.703996587734889,
      "language_loss": 0.74078274,
      "learning_rate": 3.3347455550174537e-06,
      "loss": 0.76045632,
      "num_input_tokens_seen": 51532730,
      "step": 2407,
      "time_per_iteration": 2.49052357673645
    },
    {
      "auxiliary_loss_clip": 0.01152303,
      "auxiliary_loss_mlp": 0.01031247,
      "balance_loss_clip": 1.05135429,
      "balance_loss_mlp": 1.02257395,
      "epoch": 0.289544880658931,
      "flos": 14645700737280.0,
      "grad_norm": 1.6652870265105486,
      "language_loss": 0.67954171,
      "learning_rate": 3.3341653350472864e-06,
      "loss": 0.70137727,
      "num_input_tokens_seen": 51549560,
      "step": 2408,
      "time_per_iteration": 2.517117738723755
    },
    {
      "auxiliary_loss_clip": 0.0120579,
      "auxiliary_loss_mlp": 0.01034307,
      "balance_loss_clip": 1.0603354,
      "balance_loss_mlp": 1.02379906,
      "epoch": 0.28966512354957014,
      "flos": 28621881918720.0,
      "grad_norm": 2.4811104187625452,
      "language_loss": 0.6946069,
      "learning_rate": 3.333584912685298e-06,
      "loss": 0.71700788,
      "num_input_tokens_seen": 51568180,
      "step": 2409,
      "time_per_iteration": 2.5228824615478516
    },
    {
      "auxiliary_loss_clip": 0.01059413,
      "auxiliary_loss_mlp": 0.01006232,
      "balance_loss_clip": 1.02534842,
      "balance_loss_mlp": 1.0043366,
      "epoch": 0.28978536644020925,
      "flos": 64711784511360.0,
      "grad_norm": 0.8970236442131861,
      "language_loss": 0.55562925,
      "learning_rate": 3.3330042880195385e-06,
      "loss": 0.57628572,
      "num_input_tokens_seen": 51622530,
      "step": 2410,
      "time_per_iteration": 3.801926612854004
    },
    {
      "auxiliary_loss_clip": 0.01170732,
      "auxiliary_loss_mlp": 0.01028572,
      "balance_loss_clip": 1.05524683,
      "balance_loss_mlp": 1.01957738,
      "epoch": 0.2899056093308483,
      "flos": 18624638937600.0,
      "grad_norm": 2.131850535188781,
      "language_loss": 0.78816617,
      "learning_rate": 3.3324234611380888e-06,
      "loss": 0.81015921,
      "num_input_tokens_seen": 51641260,
      "step": 2411,
      "time_per_iteration": 3.307101249694824
    },
    {
      "auxiliary_loss_clip": 0.01149658,
      "auxiliary_loss_mlp": 0.01036056,
      "balance_loss_clip": 1.05336714,
      "balance_loss_mlp": 1.02729964,
      "epoch": 0.2900258522214874,
      "flos": 22893735202560.0,
      "grad_norm": 2.312450210385715,
      "language_loss": 0.81642991,
      "learning_rate": 3.3318424321290596e-06,
      "loss": 0.838287,
      "num_input_tokens_seen": 51660975,
      "step": 2412,
      "time_per_iteration": 3.31197190284729
    },
    {
      "auxiliary_loss_clip": 0.01050656,
      "auxiliary_loss_mlp": 0.01000991,
      "balance_loss_clip": 1.01744866,
      "balance_loss_mlp": 0.99914283,
      "epoch": 0.2901460951121265,
      "flos": 71106036013440.0,
      "grad_norm": 0.8273849352883264,
      "language_loss": 0.59949553,
      "learning_rate": 3.3312612010805917e-06,
      "loss": 0.62001193,
      "num_input_tokens_seen": 51720550,
      "step": 2413,
      "time_per_iteration": 3.201558828353882
    },
    {
      "auxiliary_loss_clip": 0.01158828,
      "auxiliary_loss_mlp": 0.01038101,
      "balance_loss_clip": 1.05455792,
      "balance_loss_mlp": 1.02876127,
      "epoch": 0.2902663380027656,
      "flos": 32160986081280.0,
      "grad_norm": 1.7020718499068668,
      "language_loss": 0.7019453,
      "learning_rate": 3.330679768080858e-06,
      "loss": 0.72391456,
      "num_input_tokens_seen": 51744435,
      "step": 2414,
      "time_per_iteration": 2.600551128387451
    },
    {
      "auxiliary_loss_clip": 0.01185823,
      "auxiliary_loss_mlp": 0.01042939,
      "balance_loss_clip": 1.06016445,
      "balance_loss_mlp": 1.03324091,
      "epoch": 0.2903865808934047,
      "flos": 29351658539520.0,
      "grad_norm": 2.3670046855639844,
      "language_loss": 0.83202106,
      "learning_rate": 3.3300981332180627e-06,
      "loss": 0.85430872,
      "num_input_tokens_seen": 51763640,
      "step": 2415,
      "time_per_iteration": 2.5840582847595215
    },
    {
      "auxiliary_loss_clip": 0.01161934,
      "auxiliary_loss_mlp": 0.01036163,
      "balance_loss_clip": 1.05605626,
      "balance_loss_mlp": 1.02699614,
      "epoch": 0.29050682378404374,
      "flos": 17089026647040.0,
      "grad_norm": 1.9505864914524436,
      "language_loss": 0.79842639,
      "learning_rate": 3.3295162965804373e-06,
      "loss": 0.82040739,
      "num_input_tokens_seen": 51782135,
      "step": 2416,
      "time_per_iteration": 2.5302350521087646
    },
    {
      "auxiliary_loss_clip": 0.0115445,
      "auxiliary_loss_mlp": 0.01034533,
      "balance_loss_clip": 1.05459321,
      "balance_loss_mlp": 1.02594972,
      "epoch": 0.29062706667468285,
      "flos": 17858233422720.0,
      "grad_norm": 2.2295140351066265,
      "language_loss": 0.78410506,
      "learning_rate": 3.328934258256247e-06,
      "loss": 0.80599493,
      "num_input_tokens_seen": 51800200,
      "step": 2417,
      "time_per_iteration": 2.5521957874298096
    },
    {
      "auxiliary_loss_clip": 0.01182792,
      "auxiliary_loss_mlp": 0.01035066,
      "balance_loss_clip": 1.05612218,
      "balance_loss_mlp": 1.02612495,
      "epoch": 0.29074730956532197,
      "flos": 24279815174400.0,
      "grad_norm": 3.079278819489508,
      "language_loss": 0.67013514,
      "learning_rate": 3.3283520183337856e-06,
      "loss": 0.69231373,
      "num_input_tokens_seen": 51819905,
      "step": 2418,
      "time_per_iteration": 2.5344526767730713
    },
    {
      "auxiliary_loss_clip": 0.01168346,
      "auxiliary_loss_mlp": 0.01031048,
      "balance_loss_clip": 1.05508685,
      "balance_loss_mlp": 1.02204204,
      "epoch": 0.290867552455961,
      "flos": 22340961826560.0,
      "grad_norm": 1.7730347837345348,
      "language_loss": 0.68783128,
      "learning_rate": 3.3277695769013797e-06,
      "loss": 0.70982516,
      "num_input_tokens_seen": 51839350,
      "step": 2419,
      "time_per_iteration": 2.574254274368286
    },
    {
      "auxiliary_loss_clip": 0.01186153,
      "auxiliary_loss_mlp": 0.01036302,
      "balance_loss_clip": 1.05865359,
      "balance_loss_mlp": 1.02687287,
      "epoch": 0.29098779534660013,
      "flos": 23186155824000.0,
      "grad_norm": 1.9148056965129299,
      "language_loss": 0.77428234,
      "learning_rate": 3.327186934047385e-06,
      "loss": 0.79650688,
      "num_input_tokens_seen": 51858045,
      "step": 2420,
      "time_per_iteration": 2.5046346187591553
    },
    {
      "auxiliary_loss_clip": 0.01158492,
      "auxiliary_loss_mlp": 0.01038072,
      "balance_loss_clip": 1.05109048,
      "balance_loss_mlp": 1.02910709,
      "epoch": 0.29110803823723924,
      "flos": 15304194817920.0,
      "grad_norm": 3.04833949111504,
      "language_loss": 0.655424,
      "learning_rate": 3.3266040898601877e-06,
      "loss": 0.67738962,
      "num_input_tokens_seen": 51875880,
      "step": 2421,
      "time_per_iteration": 2.5122509002685547
    },
    {
      "auxiliary_loss_clip": 0.0113248,
      "auxiliary_loss_mlp": 0.01035251,
      "balance_loss_clip": 1.04723775,
      "balance_loss_mlp": 1.0270257,
      "epoch": 0.2912282811278783,
      "flos": 22595352923520.0,
      "grad_norm": 1.8914826358752002,
      "language_loss": 0.77827018,
      "learning_rate": 3.3260210444282045e-06,
      "loss": 0.7999475,
      "num_input_tokens_seen": 51893835,
      "step": 2422,
      "time_per_iteration": 2.6015543937683105
    },
    {
      "auxiliary_loss_clip": 0.01177379,
      "auxiliary_loss_mlp": 0.01030086,
      "balance_loss_clip": 1.05566704,
      "balance_loss_mlp": 1.02141345,
      "epoch": 0.2913485240185174,
      "flos": 24497900599680.0,
      "grad_norm": 2.1318419610532513,
      "language_loss": 0.73121077,
      "learning_rate": 3.325437797839883e-06,
      "loss": 0.75328541,
      "num_input_tokens_seen": 51912205,
      "step": 2423,
      "time_per_iteration": 2.529358148574829
    },
    {
      "auxiliary_loss_clip": 0.01200246,
      "auxiliary_loss_mlp": 0.0103535,
      "balance_loss_clip": 1.05909157,
      "balance_loss_mlp": 1.02655888,
      "epoch": 0.2914687669091565,
      "flos": 17931024334080.0,
      "grad_norm": 2.418907675710475,
      "language_loss": 0.75137031,
      "learning_rate": 3.3248543501837015e-06,
      "loss": 0.77372622,
      "num_input_tokens_seen": 51929410,
      "step": 2424,
      "time_per_iteration": 2.52130126953125
    },
    {
      "auxiliary_loss_clip": 0.01151372,
      "auxiliary_loss_mlp": 0.01034171,
      "balance_loss_clip": 1.05811715,
      "balance_loss_mlp": 1.02617812,
      "epoch": 0.2915890097997956,
      "flos": 22529313768960.0,
      "grad_norm": 1.8058898493029816,
      "language_loss": 0.77121162,
      "learning_rate": 3.3242707015481684e-06,
      "loss": 0.7930671,
      "num_input_tokens_seen": 51949345,
      "step": 2425,
      "time_per_iteration": 2.6164917945861816
    },
    {
      "auxiliary_loss_clip": 0.01166409,
      "auxiliary_loss_mlp": 0.0103099,
      "balance_loss_clip": 1.04963052,
      "balance_loss_mlp": 1.02228224,
      "epoch": 0.2917092526904347,
      "flos": 13845216193920.0,
      "grad_norm": 1.635185620823894,
      "language_loss": 0.80908477,
      "learning_rate": 3.323686852021823e-06,
      "loss": 0.83105874,
      "num_input_tokens_seen": 51966855,
      "step": 2426,
      "time_per_iteration": 2.5039713382720947
    },
    {
      "auxiliary_loss_clip": 0.01159832,
      "auxiliary_loss_mlp": 0.01037532,
      "balance_loss_clip": 1.05212998,
      "balance_loss_mlp": 1.02793002,
      "epoch": 0.2918294955810738,
      "flos": 22674859678080.0,
      "grad_norm": 1.9354214409264983,
      "language_loss": 0.79396296,
      "learning_rate": 3.323102801693235e-06,
      "loss": 0.81593657,
      "num_input_tokens_seen": 51985620,
      "step": 2427,
      "time_per_iteration": 2.5924267768859863
    },
    {
      "auxiliary_loss_clip": 0.01176676,
      "auxiliary_loss_mlp": 0.01032829,
      "balance_loss_clip": 1.05561376,
      "balance_loss_mlp": 1.0235368,
      "epoch": 0.29194973847171285,
      "flos": 23438284364160.0,
      "grad_norm": 2.2368522512028135,
      "language_loss": 0.80944872,
      "learning_rate": 3.322518550651003e-06,
      "loss": 0.83154368,
      "num_input_tokens_seen": 52004930,
      "step": 2428,
      "time_per_iteration": 2.535911798477173
    },
    {
      "auxiliary_loss_clip": 0.01174391,
      "auxiliary_loss_mlp": 0.0103453,
      "balance_loss_clip": 1.053864,
      "balance_loss_mlp": 1.02587557,
      "epoch": 0.29206998136235196,
      "flos": 21909064694400.0,
      "grad_norm": 1.6596149562710818,
      "language_loss": 0.81304467,
      "learning_rate": 3.3219340989837586e-06,
      "loss": 0.83513391,
      "num_input_tokens_seen": 52024920,
      "step": 2429,
      "time_per_iteration": 3.3419485092163086
    },
    {
      "auxiliary_loss_clip": 0.01169841,
      "auxiliary_loss_mlp": 0.01034004,
      "balance_loss_clip": 1.05479693,
      "balance_loss_mlp": 1.02622008,
      "epoch": 0.292190224252991,
      "flos": 23215925220480.0,
      "grad_norm": 1.698795395695219,
      "language_loss": 0.80147791,
      "learning_rate": 3.3213494467801625e-06,
      "loss": 0.82351631,
      "num_input_tokens_seen": 52044095,
      "step": 2430,
      "time_per_iteration": 2.5549981594085693
    },
    {
      "auxiliary_loss_clip": 0.01111268,
      "auxiliary_loss_mlp": 0.01033147,
      "balance_loss_clip": 1.04521585,
      "balance_loss_mlp": 1.02389598,
      "epoch": 0.2923104671436301,
      "flos": 20740818752640.0,
      "grad_norm": 1.992233760305085,
      "language_loss": 0.71436083,
      "learning_rate": 3.3207645941289063e-06,
      "loss": 0.73580498,
      "num_input_tokens_seen": 52062440,
      "step": 2431,
      "time_per_iteration": 2.715721368789673
    },
    {
      "auxiliary_loss_clip": 0.01182978,
      "auxiliary_loss_mlp": 0.00762449,
      "balance_loss_clip": 1.05692124,
      "balance_loss_mlp": 1.0003469,
      "epoch": 0.29243071003426924,
      "flos": 35809114999680.0,
      "grad_norm": 1.7835221353485522,
      "language_loss": 0.80036533,
      "learning_rate": 3.320179541118711e-06,
      "loss": 0.81981957,
      "num_input_tokens_seen": 52084940,
      "step": 2432,
      "time_per_iteration": 2.715223550796509
    },
    {
      "auxiliary_loss_clip": 0.01074643,
      "auxiliary_loss_mlp": 0.01003369,
      "balance_loss_clip": 1.01757157,
      "balance_loss_mlp": 1.00152159,
      "epoch": 0.2925509529249083,
      "flos": 58081598524800.0,
      "grad_norm": 1.00345004022608,
      "language_loss": 0.60321307,
      "learning_rate": 3.3195942878383293e-06,
      "loss": 0.62399316,
      "num_input_tokens_seen": 52141040,
      "step": 2433,
      "time_per_iteration": 3.1321558952331543
    },
    {
      "auxiliary_loss_clip": 0.01186527,
      "auxiliary_loss_mlp": 0.01037681,
      "balance_loss_clip": 1.06054354,
      "balance_loss_mlp": 1.02857351,
      "epoch": 0.2926711958155474,
      "flos": 21397122103680.0,
      "grad_norm": 3.814591903988737,
      "language_loss": 0.777349,
      "learning_rate": 3.319008834376543e-06,
      "loss": 0.79959106,
      "num_input_tokens_seen": 52160730,
      "step": 2434,
      "time_per_iteration": 2.5396580696105957
    },
    {
      "auxiliary_loss_clip": 0.01158888,
      "auxiliary_loss_mlp": 0.01028754,
      "balance_loss_clip": 1.04928637,
      "balance_loss_mlp": 1.01960456,
      "epoch": 0.2927914387061865,
      "flos": 23185796688000.0,
      "grad_norm": 2.6140770324049103,
      "language_loss": 0.8864547,
      "learning_rate": 3.3184231808221654e-06,
      "loss": 0.90833116,
      "num_input_tokens_seen": 52175055,
      "step": 2435,
      "time_per_iteration": 2.5725255012512207
    },
    {
      "auxiliary_loss_clip": 0.01156316,
      "auxiliary_loss_mlp": 0.0103427,
      "balance_loss_clip": 1.05384183,
      "balance_loss_mlp": 1.02479327,
      "epoch": 0.29291168159682557,
      "flos": 22455553190400.0,
      "grad_norm": 2.278607087779395,
      "language_loss": 0.62422669,
      "learning_rate": 3.3178373272640394e-06,
      "loss": 0.64613259,
      "num_input_tokens_seen": 52194150,
      "step": 2436,
      "time_per_iteration": 4.107103109359741
    },
    {
      "auxiliary_loss_clip": 0.01197018,
      "auxiliary_loss_mlp": 0.01030435,
      "balance_loss_clip": 1.05929971,
      "balance_loss_mlp": 1.02194166,
      "epoch": 0.2930319244874647,
      "flos": 21170632896000.0,
      "grad_norm": 2.402536829158671,
      "language_loss": 0.84483981,
      "learning_rate": 3.3172512737910387e-06,
      "loss": 0.86711442,
      "num_input_tokens_seen": 52211660,
      "step": 2437,
      "time_per_iteration": 2.4915659427642822
    },
    {
      "auxiliary_loss_clip": 0.01183612,
      "auxiliary_loss_mlp": 0.01037392,
      "balance_loss_clip": 1.05524182,
      "balance_loss_mlp": 1.02842748,
      "epoch": 0.2931521673781038,
      "flos": 31357843931520.0,
      "grad_norm": 1.9839286344219806,
      "language_loss": 0.87983799,
      "learning_rate": 3.3166650204920674e-06,
      "loss": 0.90204799,
      "num_input_tokens_seen": 52232830,
      "step": 2438,
      "time_per_iteration": 3.404224157333374
    },
    {
      "auxiliary_loss_clip": 0.01185194,
      "auxiliary_loss_mlp": 0.01031314,
      "balance_loss_clip": 1.05863845,
      "balance_loss_mlp": 1.02194989,
      "epoch": 0.29327241026874284,
      "flos": 24200990778240.0,
      "grad_norm": 1.699245774308353,
      "language_loss": 0.81571299,
      "learning_rate": 3.316078567456059e-06,
      "loss": 0.83787799,
      "num_input_tokens_seen": 52250670,
      "step": 2439,
      "time_per_iteration": 2.5440163612365723
    },
    {
      "auxiliary_loss_clip": 0.01132734,
      "auxiliary_loss_mlp": 0.01040677,
      "balance_loss_clip": 1.05131006,
      "balance_loss_mlp": 1.03196275,
      "epoch": 0.29339265315938196,
      "flos": 24242611662720.0,
      "grad_norm": 2.1008481464579516,
      "language_loss": 0.75592458,
      "learning_rate": 3.3154919147719786e-06,
      "loss": 0.7776587,
      "num_input_tokens_seen": 52271685,
      "step": 2440,
      "time_per_iteration": 2.6706416606903076
    },
    {
      "auxiliary_loss_clip": 0.01183594,
      "auxiliary_loss_mlp": 0.01032588,
      "balance_loss_clip": 1.05680084,
      "balance_loss_mlp": 1.02371836,
      "epoch": 0.29351289605002107,
      "flos": 16946641134720.0,
      "grad_norm": 2.503933893656644,
      "language_loss": 0.85954106,
      "learning_rate": 3.31490506252882e-06,
      "loss": 0.88170284,
      "num_input_tokens_seen": 52291065,
      "step": 2441,
      "time_per_iteration": 2.5020763874053955
    },
    {
      "auxiliary_loss_clip": 0.01146298,
      "auxiliary_loss_mlp": 0.01038992,
      "balance_loss_clip": 1.04910517,
      "balance_loss_mlp": 1.03021216,
      "epoch": 0.2936331389406601,
      "flos": 19829082810240.0,
      "grad_norm": 1.92249498911603,
      "language_loss": 0.83540368,
      "learning_rate": 3.31431801081561e-06,
      "loss": 0.85725653,
      "num_input_tokens_seen": 52310000,
      "step": 2442,
      "time_per_iteration": 2.5887837409973145
    },
    {
      "auxiliary_loss_clip": 0.01075814,
      "auxiliary_loss_mlp": 0.01002401,
      "balance_loss_clip": 1.03314924,
      "balance_loss_mlp": 1.00074387,
      "epoch": 0.29375338183129923,
      "flos": 71416844398080.0,
      "grad_norm": 0.9082784831042658,
      "language_loss": 0.67917764,
      "learning_rate": 3.313730759721402e-06,
      "loss": 0.69995981,
      "num_input_tokens_seen": 52372930,
      "step": 2443,
      "time_per_iteration": 3.2620534896850586
    },
    {
      "auxiliary_loss_clip": 0.01167475,
      "auxiliary_loss_mlp": 0.01040118,
      "balance_loss_clip": 1.05678678,
      "balance_loss_mlp": 1.03167796,
      "epoch": 0.29387362472193834,
      "flos": 22054502862720.0,
      "grad_norm": 1.8678653541792958,
      "language_loss": 0.86396754,
      "learning_rate": 3.313143309335282e-06,
      "loss": 0.88604349,
      "num_input_tokens_seen": 52391420,
      "step": 2444,
      "time_per_iteration": 2.530712604522705
    },
    {
      "auxiliary_loss_clip": 0.01152829,
      "auxiliary_loss_mlp": 0.01030546,
      "balance_loss_clip": 1.05355835,
      "balance_loss_mlp": 1.02137852,
      "epoch": 0.2939938676125774,
      "flos": 22966418373120.0,
      "grad_norm": 1.8895841324958564,
      "language_loss": 0.84930515,
      "learning_rate": 3.3125556597463665e-06,
      "loss": 0.87113881,
      "num_input_tokens_seen": 52410725,
      "step": 2445,
      "time_per_iteration": 2.5981037616729736
    },
    {
      "auxiliary_loss_clip": 0.01185379,
      "auxiliary_loss_mlp": 0.01039106,
      "balance_loss_clip": 1.05999565,
      "balance_loss_mlp": 1.03084469,
      "epoch": 0.2941141105032165,
      "flos": 31358705857920.0,
      "grad_norm": 2.155739762763418,
      "language_loss": 0.66615808,
      "learning_rate": 3.311967811043801e-06,
      "loss": 0.68840295,
      "num_input_tokens_seen": 52432645,
      "step": 2446,
      "time_per_iteration": 2.58964467048645
    },
    {
      "auxiliary_loss_clip": 0.01184652,
      "auxiliary_loss_mlp": 0.01037896,
      "balance_loss_clip": 1.0597589,
      "balance_loss_mlp": 1.02924728,
      "epoch": 0.29423435339385556,
      "flos": 23222138273280.0,
      "grad_norm": 2.0215629468298064,
      "language_loss": 0.82002008,
      "learning_rate": 3.3113797633167617e-06,
      "loss": 0.84224552,
      "num_input_tokens_seen": 52450940,
      "step": 2447,
      "time_per_iteration": 2.5618627071380615
    },
    {
      "auxiliary_loss_clip": 0.01198472,
      "auxiliary_loss_mlp": 0.01031819,
      "balance_loss_clip": 1.05867612,
      "balance_loss_mlp": 1.02313459,
      "epoch": 0.2943545962844947,
      "flos": 26864054138880.0,
      "grad_norm": 2.0338429723827662,
      "language_loss": 0.68589652,
      "learning_rate": 3.310791516654455e-06,
      "loss": 0.70819944,
      "num_input_tokens_seen": 52468000,
      "step": 2448,
      "time_per_iteration": 2.510129928588867
    },
    {
      "auxiliary_loss_clip": 0.01158785,
      "auxiliary_loss_mlp": 0.0103798,
      "balance_loss_clip": 1.05226517,
      "balance_loss_mlp": 1.02893829,
      "epoch": 0.2944748391751338,
      "flos": 20231677422720.0,
      "grad_norm": 1.9200950876062715,
      "language_loss": 0.79376876,
      "learning_rate": 3.3102030711461177e-06,
      "loss": 0.81573641,
      "num_input_tokens_seen": 52487575,
      "step": 2449,
      "time_per_iteration": 2.5894017219543457
    },
    {
      "auxiliary_loss_clip": 0.01157068,
      "auxiliary_loss_mlp": 0.01039375,
      "balance_loss_clip": 1.05295348,
      "balance_loss_mlp": 1.02993369,
      "epoch": 0.29459508206577284,
      "flos": 15960965045760.0,
      "grad_norm": 2.122811860071899,
      "language_loss": 0.67900801,
      "learning_rate": 3.3096144268810156e-06,
      "loss": 0.70097244,
      "num_input_tokens_seen": 52506335,
      "step": 2450,
      "time_per_iteration": 2.561272382736206
    },
    {
      "auxiliary_loss_clip": 0.01171769,
      "auxiliary_loss_mlp": 0.01036701,
      "balance_loss_clip": 1.05361807,
      "balance_loss_mlp": 1.02752209,
      "epoch": 0.29471532495641195,
      "flos": 20412882558720.0,
      "grad_norm": 1.8571043711220467,
      "language_loss": 0.72882128,
      "learning_rate": 3.3090255839484462e-06,
      "loss": 0.75090605,
      "num_input_tokens_seen": 52524330,
      "step": 2451,
      "time_per_iteration": 2.5463149547576904
    },
    {
      "auxiliary_loss_clip": 0.01167173,
      "auxiliary_loss_mlp": 0.01032145,
      "balance_loss_clip": 1.05220604,
      "balance_loss_mlp": 1.02353811,
      "epoch": 0.29483556784705106,
      "flos": 20376576887040.0,
      "grad_norm": 1.8691923021265149,
      "language_loss": 0.85358196,
      "learning_rate": 3.3084365424377366e-06,
      "loss": 0.87557518,
      "num_input_tokens_seen": 52543095,
      "step": 2452,
      "time_per_iteration": 2.5425808429718018
    },
    {
      "auxiliary_loss_clip": 0.01048809,
      "auxiliary_loss_mlp": 0.01002979,
      "balance_loss_clip": 1.02432251,
      "balance_loss_mlp": 1.00144124,
      "epoch": 0.2949558107376901,
      "flos": 68555660595840.0,
      "grad_norm": 1.2123634580202236,
      "language_loss": 0.56001246,
      "learning_rate": 3.307847302438245e-06,
      "loss": 0.58053029,
      "num_input_tokens_seen": 52597075,
      "step": 2453,
      "time_per_iteration": 3.0884931087493896
    },
    {
      "auxiliary_loss_clip": 0.01122085,
      "auxiliary_loss_mlp": 0.01037675,
      "balance_loss_clip": 1.04489481,
      "balance_loss_mlp": 1.02823925,
      "epoch": 0.2950760536283292,
      "flos": 16107085572480.0,
      "grad_norm": 2.088588009301697,
      "language_loss": 0.77563035,
      "learning_rate": 3.3072578640393562e-06,
      "loss": 0.79722792,
      "num_input_tokens_seen": 52614410,
      "step": 2454,
      "time_per_iteration": 2.594249725341797
    },
    {
      "auxiliary_loss_clip": 0.01170135,
      "auxiliary_loss_mlp": 0.01031695,
      "balance_loss_clip": 1.05491805,
      "balance_loss_mlp": 1.02315974,
      "epoch": 0.29519629651896834,
      "flos": 20483626394880.0,
      "grad_norm": 2.1030331415416788,
      "language_loss": 0.79631305,
      "learning_rate": 3.3066682273304886e-06,
      "loss": 0.81833136,
      "num_input_tokens_seen": 52632055,
      "step": 2455,
      "time_per_iteration": 3.252758026123047
    },
    {
      "auxiliary_loss_clip": 0.01187842,
      "auxiliary_loss_mlp": 0.007628,
      "balance_loss_clip": 1.05767989,
      "balance_loss_mlp": 1.00032735,
      "epoch": 0.2953165394096074,
      "flos": 18916484941440.0,
      "grad_norm": 1.8979001851550605,
      "language_loss": 0.78561115,
      "learning_rate": 3.3060783924010904e-06,
      "loss": 0.80511755,
      "num_input_tokens_seen": 52649980,
      "step": 2456,
      "time_per_iteration": 2.5073962211608887
    },
    {
      "auxiliary_loss_clip": 0.0115496,
      "auxiliary_loss_mlp": 0.01032481,
      "balance_loss_clip": 1.05314028,
      "balance_loss_mlp": 1.02370071,
      "epoch": 0.2954367823002465,
      "flos": 20624467622400.0,
      "grad_norm": 2.0805814026230975,
      "language_loss": 0.85043514,
      "learning_rate": 3.3054883593406387e-06,
      "loss": 0.87230957,
      "num_input_tokens_seen": 52664730,
      "step": 2457,
      "time_per_iteration": 2.5773870944976807
    },
    {
      "auxiliary_loss_clip": 0.01173291,
      "auxiliary_loss_mlp": 0.01033991,
      "balance_loss_clip": 1.05480123,
      "balance_loss_mlp": 1.02513981,
      "epoch": 0.2955570251908856,
      "flos": 31175525473920.0,
      "grad_norm": 2.2764640571841888,
      "language_loss": 0.65365601,
      "learning_rate": 3.3048981282386404e-06,
      "loss": 0.67572874,
      "num_input_tokens_seen": 52686040,
      "step": 2458,
      "time_per_iteration": 2.6222801208496094
    },
    {
      "auxiliary_loss_clip": 0.01141152,
      "auxiliary_loss_mlp": 0.0103434,
      "balance_loss_clip": 1.05137539,
      "balance_loss_mlp": 1.0250771,
      "epoch": 0.29567726808152467,
      "flos": 21650328051840.0,
      "grad_norm": 3.719711829924088,
      "language_loss": 0.82628119,
      "learning_rate": 3.304307699184634e-06,
      "loss": 0.84803605,
      "num_input_tokens_seen": 52704630,
      "step": 2459,
      "time_per_iteration": 2.5909266471862793
    },
    {
      "auxiliary_loss_clip": 0.01171326,
      "auxiliary_loss_mlp": 0.01038541,
      "balance_loss_clip": 1.05718517,
      "balance_loss_mlp": 1.02958894,
      "epoch": 0.2957975109721638,
      "flos": 24243868638720.0,
      "grad_norm": 1.7349819985106545,
      "language_loss": 0.7889744,
      "learning_rate": 3.3037170722681866e-06,
      "loss": 0.81107306,
      "num_input_tokens_seen": 52725465,
      "step": 2460,
      "time_per_iteration": 2.590567111968994
    },
    {
      "auxiliary_loss_clip": 0.01148109,
      "auxiliary_loss_mlp": 0.01032396,
      "balance_loss_clip": 1.05330038,
      "balance_loss_mlp": 1.02321124,
      "epoch": 0.29591775386280283,
      "flos": 13479717352320.0,
      "grad_norm": 1.77757326079078,
      "language_loss": 0.67758799,
      "learning_rate": 3.3031262475788956e-06,
      "loss": 0.69939303,
      "num_input_tokens_seen": 52742405,
      "step": 2461,
      "time_per_iteration": 2.543487548828125
    },
    {
      "auxiliary_loss_clip": 0.01166042,
      "auxiliary_loss_mlp": 0.01025207,
      "balance_loss_clip": 1.05477631,
      "balance_loss_mlp": 1.01691031,
      "epoch": 0.29603799675344195,
      "flos": 17749783284480.0,
      "grad_norm": 1.634353352880499,
      "language_loss": 0.72798067,
      "learning_rate": 3.3025352252063897e-06,
      "loss": 0.74989319,
      "num_input_tokens_seen": 52761100,
      "step": 2462,
      "time_per_iteration": 3.3852343559265137
    },
    {
      "auxiliary_loss_clip": 0.01185023,
      "auxiliary_loss_mlp": 0.01029853,
      "balance_loss_clip": 1.05919933,
      "balance_loss_mlp": 1.02192569,
      "epoch": 0.29615823964408106,
      "flos": 22783920347520.0,
      "grad_norm": 1.9256520172860376,
      "language_loss": 0.74570453,
      "learning_rate": 3.3019440052403252e-06,
      "loss": 0.76785326,
      "num_input_tokens_seen": 52780965,
      "step": 2463,
      "time_per_iteration": 2.545198440551758
    },
    {
      "auxiliary_loss_clip": 0.0117092,
      "auxiliary_loss_mlp": 0.01033352,
      "balance_loss_clip": 1.05634141,
      "balance_loss_mlp": 1.02460766,
      "epoch": 0.2962784825347201,
      "flos": 23514199758720.0,
      "grad_norm": 2.0967880326168995,
      "language_loss": 0.70839697,
      "learning_rate": 3.30135258777039e-06,
      "loss": 0.73043966,
      "num_input_tokens_seen": 52800335,
      "step": 2464,
      "time_per_iteration": 3.280278444290161
    },
    {
      "auxiliary_loss_clip": 0.01186439,
      "auxiliary_loss_mlp": 0.00762591,
      "balance_loss_clip": 1.05541146,
      "balance_loss_mlp": 1.00031757,
      "epoch": 0.2963987254253592,
      "flos": 16362769559040.0,
      "grad_norm": 1.9750850652919807,
      "language_loss": 0.70132291,
      "learning_rate": 3.3007609728863024e-06,
      "loss": 0.72081327,
      "num_input_tokens_seen": 52818425,
      "step": 2465,
      "time_per_iteration": 2.514265775680542
    },
    {
      "auxiliary_loss_clip": 0.01123416,
      "auxiliary_loss_mlp": 0.01031433,
      "balance_loss_clip": 1.05566669,
      "balance_loss_mlp": 1.02278435,
      "epoch": 0.29651896831599833,
      "flos": 33472263980160.0,
      "grad_norm": 1.7932924768866525,
      "language_loss": 0.73105145,
      "learning_rate": 3.300169160677809e-06,
      "loss": 0.7525999,
      "num_input_tokens_seen": 52842340,
      "step": 2466,
      "time_per_iteration": 2.7565760612487793
    },
    {
      "auxiliary_loss_clip": 0.01165384,
      "auxiliary_loss_mlp": 0.01033375,
      "balance_loss_clip": 1.05672431,
      "balance_loss_mlp": 1.02387428,
      "epoch": 0.2966392112066374,
      "flos": 23805363404160.0,
      "grad_norm": 2.6386223810522362,
      "language_loss": 0.77660775,
      "learning_rate": 3.2995771512346878e-06,
      "loss": 0.79859537,
      "num_input_tokens_seen": 52860690,
      "step": 2467,
      "time_per_iteration": 2.5952529907226562
    },
    {
      "auxiliary_loss_clip": 0.01201767,
      "auxiliary_loss_mlp": 0.00762383,
      "balance_loss_clip": 1.06075978,
      "balance_loss_mlp": 1.0003413,
      "epoch": 0.2967594540972765,
      "flos": 19938466702080.0,
      "grad_norm": 2.3795122361842105,
      "language_loss": 0.73470116,
      "learning_rate": 3.298984944646746e-06,
      "loss": 0.75434268,
      "num_input_tokens_seen": 52879370,
      "step": 2468,
      "time_per_iteration": 2.4656336307525635
    },
    {
      "auxiliary_loss_clip": 0.01187394,
      "auxiliary_loss_mlp": 0.00761826,
      "balance_loss_clip": 1.0599314,
      "balance_loss_mlp": 1.00036609,
      "epoch": 0.2968796969879156,
      "flos": 23732823888000.0,
      "grad_norm": 1.859371354096997,
      "language_loss": 0.81915569,
      "learning_rate": 3.298392541003822e-06,
      "loss": 0.83864784,
      "num_input_tokens_seen": 52898775,
      "step": 2469,
      "time_per_iteration": 2.562732696533203
    },
    {
      "auxiliary_loss_clip": 0.01165706,
      "auxiliary_loss_mlp": 0.01034319,
      "balance_loss_clip": 1.05425835,
      "balance_loss_mlp": 1.02629662,
      "epoch": 0.29699993987855466,
      "flos": 22893699288960.0,
      "grad_norm": 1.7238899784871047,
      "language_loss": 0.89789903,
      "learning_rate": 3.2977999403957806e-06,
      "loss": 0.91989928,
      "num_input_tokens_seen": 52917535,
      "step": 2470,
      "time_per_iteration": 2.5340116024017334
    },
    {
      "auxiliary_loss_clip": 0.01200573,
      "auxiliary_loss_mlp": 0.01034482,
      "balance_loss_clip": 1.06088161,
      "balance_loss_mlp": 1.02548122,
      "epoch": 0.2971201827691938,
      "flos": 33832555349760.0,
      "grad_norm": 2.017282103090687,
      "language_loss": 0.67053419,
      "learning_rate": 3.2972071429125207e-06,
      "loss": 0.69288474,
      "num_input_tokens_seen": 52938755,
      "step": 2471,
      "time_per_iteration": 2.610945463180542
    },
    {
      "auxiliary_loss_clip": 0.01154294,
      "auxiliary_loss_mlp": 0.0103386,
      "balance_loss_clip": 1.05513227,
      "balance_loss_mlp": 1.02459741,
      "epoch": 0.2972404256598329,
      "flos": 22054359208320.0,
      "grad_norm": 2.0443086939334796,
      "language_loss": 0.88290483,
      "learning_rate": 3.2966141486439682e-06,
      "loss": 0.90478641,
      "num_input_tokens_seen": 52957945,
      "step": 2472,
      "time_per_iteration": 2.5556628704071045
    },
    {
      "auxiliary_loss_clip": 0.0112616,
      "auxiliary_loss_mlp": 0.01035418,
      "balance_loss_clip": 1.04613543,
      "balance_loss_mlp": 1.02529693,
      "epoch": 0.29736066855047194,
      "flos": 31978595796480.0,
      "grad_norm": 2.485067868808452,
      "language_loss": 0.63920689,
      "learning_rate": 3.29602095768008e-06,
      "loss": 0.66082269,
      "num_input_tokens_seen": 52978460,
      "step": 2473,
      "time_per_iteration": 2.7373788356781006
    },
    {
      "auxiliary_loss_clip": 0.01166087,
      "auxiliary_loss_mlp": 0.01032143,
      "balance_loss_clip": 1.05825901,
      "balance_loss_mlp": 1.02371454,
      "epoch": 0.29748091144111105,
      "flos": 33510401245440.0,
      "grad_norm": 1.8789819343648704,
      "language_loss": 0.63824511,
      "learning_rate": 3.2954275701108437e-06,
      "loss": 0.66022742,
      "num_input_tokens_seen": 52999640,
      "step": 2474,
      "time_per_iteration": 2.6426918506622314
    },
    {
      "auxiliary_loss_clip": 0.01134998,
      "auxiliary_loss_mlp": 0.01030363,
      "balance_loss_clip": 1.05010724,
      "balance_loss_mlp": 1.02127886,
      "epoch": 0.29760115433175016,
      "flos": 41283373409280.0,
      "grad_norm": 1.7477815406723394,
      "language_loss": 0.684232,
      "learning_rate": 3.294833986026275e-06,
      "loss": 0.70588565,
      "num_input_tokens_seen": 53022880,
      "step": 2475,
      "time_per_iteration": 2.7863945960998535
    },
    {
      "auxiliary_loss_clip": 0.01149097,
      "auxiliary_loss_mlp": 0.01036658,
      "balance_loss_clip": 1.05248344,
      "balance_loss_mlp": 1.02708507,
      "epoch": 0.2977213972223892,
      "flos": 24493339572480.0,
      "grad_norm": 2.028121219475763,
      "language_loss": 0.85000437,
      "learning_rate": 3.29424020551642e-06,
      "loss": 0.87186193,
      "num_input_tokens_seen": 53041515,
      "step": 2476,
      "time_per_iteration": 2.5883164405822754
    },
    {
      "auxiliary_loss_clip": 0.0120539,
      "auxiliary_loss_mlp": 0.01040047,
      "balance_loss_clip": 1.06282187,
      "balance_loss_mlp": 1.02922869,
      "epoch": 0.2978416401130283,
      "flos": 21285116519040.0,
      "grad_norm": 1.8455583514008382,
      "language_loss": 0.71600235,
      "learning_rate": 3.2936462286713546e-06,
      "loss": 0.73845673,
      "num_input_tokens_seen": 53059865,
      "step": 2477,
      "time_per_iteration": 2.5126779079437256
    },
    {
      "auxiliary_loss_clip": 0.01186866,
      "auxiliary_loss_mlp": 0.01038438,
      "balance_loss_clip": 1.05916047,
      "balance_loss_mlp": 1.02896106,
      "epoch": 0.2979618830036674,
      "flos": 25772154554880.0,
      "grad_norm": 2.168664004138645,
      "language_loss": 0.77701133,
      "learning_rate": 3.2930520555811846e-06,
      "loss": 0.79926431,
      "num_input_tokens_seen": 53079490,
      "step": 2478,
      "time_per_iteration": 2.5531508922576904
    },
    {
      "auxiliary_loss_clip": 0.01088009,
      "auxiliary_loss_mlp": 0.00762774,
      "balance_loss_clip": 1.04468942,
      "balance_loss_mlp": 1.00039458,
      "epoch": 0.2980821258943065,
      "flos": 23476996247040.0,
      "grad_norm": 1.8739427548844902,
      "language_loss": 0.79995179,
      "learning_rate": 3.292457686336046e-06,
      "loss": 0.81845963,
      "num_input_tokens_seen": 53098810,
      "step": 2479,
      "time_per_iteration": 2.7011053562164307
    },
    {
      "auxiliary_loss_clip": 0.01067681,
      "auxiliary_loss_mlp": 0.01004298,
      "balance_loss_clip": 1.01993132,
      "balance_loss_mlp": 1.00243878,
      "epoch": 0.2982023687849456,
      "flos": 69752314195200.0,
      "grad_norm": 0.9166928983129429,
      "language_loss": 0.61254323,
      "learning_rate": 3.291863121026105e-06,
      "loss": 0.63326299,
      "num_input_tokens_seen": 53162590,
      "step": 2480,
      "time_per_iteration": 3.223708391189575
    },
    {
      "auxiliary_loss_clip": 0.01184399,
      "auxiliary_loss_mlp": 0.01038763,
      "balance_loss_clip": 1.05859339,
      "balance_loss_mlp": 1.02901816,
      "epoch": 0.29832261167558466,
      "flos": 29825930741760.0,
      "grad_norm": 2.6047029110385047,
      "language_loss": 0.76770616,
      "learning_rate": 3.2912683597415547e-06,
      "loss": 0.78993779,
      "num_input_tokens_seen": 53186675,
      "step": 2481,
      "time_per_iteration": 3.4659886360168457
    },
    {
      "auxiliary_loss_clip": 0.01159052,
      "auxiliary_loss_mlp": 0.01034755,
      "balance_loss_clip": 1.05394769,
      "balance_loss_mlp": 1.02549219,
      "epoch": 0.29844285456622377,
      "flos": 33910158683520.0,
      "grad_norm": 2.5332789302447756,
      "language_loss": 0.78195775,
      "learning_rate": 3.2906734025726213e-06,
      "loss": 0.80389583,
      "num_input_tokens_seen": 53205940,
      "step": 2482,
      "time_per_iteration": 2.6735920906066895
    },
    {
      "auxiliary_loss_clip": 0.01191954,
      "auxiliary_loss_mlp": 0.01037516,
      "balance_loss_clip": 1.05945182,
      "balance_loss_mlp": 1.0280447,
      "epoch": 0.2985630974568629,
      "flos": 23876933253120.0,
      "grad_norm": 2.292629247867991,
      "language_loss": 0.88564754,
      "learning_rate": 3.290078249609559e-06,
      "loss": 0.9079423,
      "num_input_tokens_seen": 53225360,
      "step": 2483,
      "time_per_iteration": 2.526149034500122
    },
    {
      "auxiliary_loss_clip": 0.01181912,
      "auxiliary_loss_mlp": 0.01040473,
      "balance_loss_clip": 1.05900955,
      "balance_loss_mlp": 1.03156209,
      "epoch": 0.29868334034750194,
      "flos": 21799106184960.0,
      "grad_norm": 1.9866865635194229,
      "language_loss": 0.88038599,
      "learning_rate": 3.2894829009426514e-06,
      "loss": 0.90260988,
      "num_input_tokens_seen": 53243195,
      "step": 2484,
      "time_per_iteration": 2.550354242324829
    },
    {
      "auxiliary_loss_clip": 0.01181505,
      "auxiliary_loss_mlp": 0.01028485,
      "balance_loss_clip": 1.05770254,
      "balance_loss_mlp": 1.02009225,
      "epoch": 0.29880358323814105,
      "flos": 25666649331840.0,
      "grad_norm": 1.847372081028074,
      "language_loss": 0.77599275,
      "learning_rate": 3.288887356662213e-06,
      "loss": 0.79809266,
      "num_input_tokens_seen": 53264530,
      "step": 2485,
      "time_per_iteration": 2.545795440673828
    },
    {
      "auxiliary_loss_clip": 0.01069384,
      "auxiliary_loss_mlp": 0.01002037,
      "balance_loss_clip": 1.01728976,
      "balance_loss_mlp": 1.0003798,
      "epoch": 0.29892382612878016,
      "flos": 71005846003200.0,
      "grad_norm": 0.7731265833124629,
      "language_loss": 0.59729171,
      "learning_rate": 3.288291616858588e-06,
      "loss": 0.61800593,
      "num_input_tokens_seen": 53319920,
      "step": 2486,
      "time_per_iteration": 2.98641037940979
    },
    {
      "auxiliary_loss_clip": 0.01134106,
      "auxiliary_loss_mlp": 0.01033022,
      "balance_loss_clip": 1.05414724,
      "balance_loss_mlp": 1.02474892,
      "epoch": 0.2990440690194192,
      "flos": 25481134563840.0,
      "grad_norm": 1.8516763966653518,
      "language_loss": 0.76823449,
      "learning_rate": 3.287695681622149e-06,
      "loss": 0.78990579,
      "num_input_tokens_seen": 53339270,
      "step": 2487,
      "time_per_iteration": 2.6198558807373047
    },
    {
      "auxiliary_loss_clip": 0.01175706,
      "auxiliary_loss_mlp": 0.01035842,
      "balance_loss_clip": 1.05728245,
      "balance_loss_mlp": 1.02670467,
      "epoch": 0.2991643119100583,
      "flos": 23732357011200.0,
      "grad_norm": 1.9817982528549574,
      "language_loss": 0.807886,
      "learning_rate": 3.2870995510432982e-06,
      "loss": 0.83000147,
      "num_input_tokens_seen": 53357750,
      "step": 2488,
      "time_per_iteration": 4.035360336303711
    },
    {
      "auxiliary_loss_clip": 0.01174938,
      "auxiliary_loss_mlp": 0.01029387,
      "balance_loss_clip": 1.05469251,
      "balance_loss_mlp": 1.02095246,
      "epoch": 0.29928455480069743,
      "flos": 27417545786880.0,
      "grad_norm": 1.9164718326331367,
      "language_loss": 0.7698313,
      "learning_rate": 3.2865032252124697e-06,
      "loss": 0.79187447,
      "num_input_tokens_seen": 53378265,
      "step": 2489,
      "time_per_iteration": 2.575758695602417
    },
    {
      "auxiliary_loss_clip": 0.01165216,
      "auxiliary_loss_mlp": 0.01029461,
      "balance_loss_clip": 1.05098987,
      "balance_loss_mlp": 1.02100313,
      "epoch": 0.2994047976913365,
      "flos": 33692935184640.0,
      "grad_norm": 1.4667192511036746,
      "language_loss": 0.77692902,
      "learning_rate": 3.2859067042201243e-06,
      "loss": 0.79887581,
      "num_input_tokens_seen": 53400305,
      "step": 2490,
      "time_per_iteration": 3.4615354537963867
    },
    {
      "auxiliary_loss_clip": 0.01105599,
      "auxiliary_loss_mlp": 0.01028291,
      "balance_loss_clip": 1.04866779,
      "balance_loss_mlp": 1.01966047,
      "epoch": 0.2995250405819756,
      "flos": 16763963541120.0,
      "grad_norm": 2.114510171439667,
      "language_loss": 0.78557283,
      "learning_rate": 3.2853099881567544e-06,
      "loss": 0.80691177,
      "num_input_tokens_seen": 53418705,
      "step": 2491,
      "time_per_iteration": 2.616880178451538
    },
    {
      "auxiliary_loss_clip": 0.01193165,
      "auxiliary_loss_mlp": 0.01033258,
      "balance_loss_clip": 1.05818331,
      "balance_loss_mlp": 1.02488947,
      "epoch": 0.29964528347261465,
      "flos": 22963976248320.0,
      "grad_norm": 1.7478453414963975,
      "language_loss": 0.7895276,
      "learning_rate": 3.284713077112881e-06,
      "loss": 0.8117919,
      "num_input_tokens_seen": 53438135,
      "step": 2492,
      "time_per_iteration": 2.5164763927459717
    },
    {
      "auxiliary_loss_clip": 0.01162148,
      "auxiliary_loss_mlp": 0.01034184,
      "balance_loss_clip": 1.05704916,
      "balance_loss_mlp": 1.02520204,
      "epoch": 0.29976552636325376,
      "flos": 16938021870720.0,
      "grad_norm": 11.249026507481148,
      "language_loss": 0.86752284,
      "learning_rate": 3.284115971179056e-06,
      "loss": 0.88948613,
      "num_input_tokens_seen": 53452165,
      "step": 2493,
      "time_per_iteration": 2.4941508769989014
    },
    {
      "auxiliary_loss_clip": 0.01129419,
      "auxiliary_loss_mlp": 0.01033247,
      "balance_loss_clip": 1.05211735,
      "balance_loss_mlp": 1.02447343,
      "epoch": 0.2998857692538929,
      "flos": 17056455989760.0,
      "grad_norm": 1.8699477178189492,
      "language_loss": 0.78707057,
      "learning_rate": 3.283518670445859e-06,
      "loss": 0.80869722,
      "num_input_tokens_seen": 53470075,
      "step": 2494,
      "time_per_iteration": 2.6393706798553467
    },
    {
      "auxiliary_loss_clip": 0.01054034,
      "auxiliary_loss_mlp": 0.00752488,
      "balance_loss_clip": 1.01771498,
      "balance_loss_mlp": 1.0003413,
      "epoch": 0.30000601214453193,
      "flos": 68831528025600.0,
      "grad_norm": 0.6796238380858755,
      "language_loss": 0.5425874,
      "learning_rate": 3.2829211750038995e-06,
      "loss": 0.56065267,
      "num_input_tokens_seen": 53538705,
      "step": 2495,
      "time_per_iteration": 3.2677879333496094
    },
    {
      "auxiliary_loss_clip": 0.01146423,
      "auxiliary_loss_mlp": 0.01029227,
      "balance_loss_clip": 1.05059052,
      "balance_loss_mlp": 1.01972067,
      "epoch": 0.30012625503517104,
      "flos": 17603267708160.0,
      "grad_norm": 1.7644545747916844,
      "language_loss": 0.89033282,
      "learning_rate": 3.2823234849438183e-06,
      "loss": 0.91208935,
      "num_input_tokens_seen": 53556740,
      "step": 2496,
      "time_per_iteration": 2.5672407150268555
    },
    {
      "auxiliary_loss_clip": 0.01172927,
      "auxiliary_loss_mlp": 0.0103262,
      "balance_loss_clip": 1.05780172,
      "balance_loss_mlp": 1.02385163,
      "epoch": 0.30024649792581015,
      "flos": 21252581775360.0,
      "grad_norm": 1.9767833704363609,
      "language_loss": 0.75908834,
      "learning_rate": 3.2817256003562836e-06,
      "loss": 0.78114378,
      "num_input_tokens_seen": 53577115,
      "step": 2497,
      "time_per_iteration": 2.5657405853271484
    },
    {
      "auxiliary_loss_clip": 0.01127885,
      "auxiliary_loss_mlp": 0.01038501,
      "balance_loss_clip": 1.05136549,
      "balance_loss_mlp": 1.02963805,
      "epoch": 0.3003667408164492,
      "flos": 23003262748800.0,
      "grad_norm": 1.8830214502412808,
      "language_loss": 0.6592325,
      "learning_rate": 3.281127521331995e-06,
      "loss": 0.6808964,
      "num_input_tokens_seen": 53598295,
      "step": 2498,
      "time_per_iteration": 2.714897632598877
    },
    {
      "auxiliary_loss_clip": 0.010846,
      "auxiliary_loss_mlp": 0.01003251,
      "balance_loss_clip": 1.01983929,
      "balance_loss_mlp": 1.00155854,
      "epoch": 0.3004869837070883,
      "flos": 64232340750720.0,
      "grad_norm": 0.8829780417685199,
      "language_loss": 0.6064657,
      "learning_rate": 3.2805292479616798e-06,
      "loss": 0.62734425,
      "num_input_tokens_seen": 53657160,
      "step": 2499,
      "time_per_iteration": 3.001973867416382
    },
    {
      "auxiliary_loss_clip": 0.01172325,
      "auxiliary_loss_mlp": 0.01033925,
      "balance_loss_clip": 1.0560832,
      "balance_loss_mlp": 1.02504945,
      "epoch": 0.30060722659772743,
      "flos": 26248653400320.0,
      "grad_norm": 2.3060179622942303,
      "language_loss": 0.91577661,
      "learning_rate": 3.2799307803360955e-06,
      "loss": 0.93783909,
      "num_input_tokens_seen": 53673090,
      "step": 2500,
      "time_per_iteration": 2.599374294281006
    },
    {
      "auxiliary_loss_clip": 0.01195159,
      "auxiliary_loss_mlp": 0.01034171,
      "balance_loss_clip": 1.05834711,
      "balance_loss_mlp": 1.02541542,
      "epoch": 0.3007274694883665,
      "flos": 24970879912320.0,
      "grad_norm": 1.527378730121462,
      "language_loss": 0.81548417,
      "learning_rate": 3.27933211854603e-06,
      "loss": 0.83777744,
      "num_input_tokens_seen": 53692145,
      "step": 2501,
      "time_per_iteration": 2.512568950653076
    },
    {
      "auxiliary_loss_clip": 0.01167227,
      "auxiliary_loss_mlp": 0.01033772,
      "balance_loss_clip": 1.05417728,
      "balance_loss_mlp": 1.02463424,
      "epoch": 0.3008477123790056,
      "flos": 17055845458560.0,
      "grad_norm": 1.7185268514402356,
      "language_loss": 0.87280363,
      "learning_rate": 3.278733262682299e-06,
      "loss": 0.8948136,
      "num_input_tokens_seen": 53710000,
      "step": 2502,
      "time_per_iteration": 2.5493764877319336
    },
    {
      "auxiliary_loss_clip": 0.01199833,
      "auxiliary_loss_mlp": 0.01031592,
      "balance_loss_clip": 1.05951917,
      "balance_loss_mlp": 1.02304494,
      "epoch": 0.3009679552696447,
      "flos": 21506398254720.0,
      "grad_norm": 3.132165855924124,
      "language_loss": 0.82169616,
      "learning_rate": 3.2781342128357484e-06,
      "loss": 0.84401047,
      "num_input_tokens_seen": 53729355,
      "step": 2503,
      "time_per_iteration": 2.4934184551239014
    },
    {
      "auxiliary_loss_clip": 0.01153029,
      "auxiliary_loss_mlp": 0.01033517,
      "balance_loss_clip": 1.05246639,
      "balance_loss_mlp": 1.025226,
      "epoch": 0.30108819816028376,
      "flos": 21134004001920.0,
      "grad_norm": 6.082762872591491,
      "language_loss": 0.8049134,
      "learning_rate": 3.2775349690972547e-06,
      "loss": 0.82677889,
      "num_input_tokens_seen": 53743505,
      "step": 2504,
      "time_per_iteration": 2.5580382347106934
    },
    {
      "auxiliary_loss_clip": 0.01076298,
      "auxiliary_loss_mlp": 0.01002128,
      "balance_loss_clip": 1.02604365,
      "balance_loss_mlp": 1.0003165,
      "epoch": 0.30120844105092287,
      "flos": 71126434938240.0,
      "grad_norm": 0.7682175658819874,
      "language_loss": 0.51827741,
      "learning_rate": 3.276935531557722e-06,
      "loss": 0.53906167,
      "num_input_tokens_seen": 53808725,
      "step": 2505,
      "time_per_iteration": 3.192943811416626
    },
    {
      "auxiliary_loss_clip": 0.01143697,
      "auxiliary_loss_mlp": 0.01028278,
      "balance_loss_clip": 1.05334234,
      "balance_loss_mlp": 1.01952839,
      "epoch": 0.301328683941562,
      "flos": 20264571302400.0,
      "grad_norm": 2.4117771820199168,
      "language_loss": 0.79473758,
      "learning_rate": 3.2763359003080837e-06,
      "loss": 0.81645733,
      "num_input_tokens_seen": 53825680,
      "step": 2506,
      "time_per_iteration": 2.5741944313049316
    },
    {
      "auxiliary_loss_clip": 0.01060626,
      "auxiliary_loss_mlp": 0.01002012,
      "balance_loss_clip": 1.01608419,
      "balance_loss_mlp": 1.00024819,
      "epoch": 0.30144892683220104,
      "flos": 70648212240000.0,
      "grad_norm": 0.8035525642113569,
      "language_loss": 0.6243301,
      "learning_rate": 3.2757360754393047e-06,
      "loss": 0.64495653,
      "num_input_tokens_seen": 53889750,
      "step": 2507,
      "time_per_iteration": 3.229503870010376
    },
    {
      "auxiliary_loss_clip": 0.01181797,
      "auxiliary_loss_mlp": 0.01031206,
      "balance_loss_clip": 1.05653954,
      "balance_loss_mlp": 1.02274227,
      "epoch": 0.30156916972284015,
      "flos": 22820549241600.0,
      "grad_norm": 2.63479311260939,
      "language_loss": 0.63851678,
      "learning_rate": 3.2751360570423767e-06,
      "loss": 0.6606468,
      "num_input_tokens_seen": 53908135,
      "step": 2508,
      "time_per_iteration": 3.2388031482696533
    },
    {
      "auxiliary_loss_clip": 0.0116737,
      "auxiliary_loss_mlp": 0.01039814,
      "balance_loss_clip": 1.05689895,
      "balance_loss_mlp": 1.03058088,
      "epoch": 0.3016894126134792,
      "flos": 29899188529920.0,
      "grad_norm": 1.9913099687783427,
      "language_loss": 0.75955772,
      "learning_rate": 3.2745358452083236e-06,
      "loss": 0.78162956,
      "num_input_tokens_seen": 53931035,
      "step": 2509,
      "time_per_iteration": 2.5982401371002197
    },
    {
      "auxiliary_loss_clip": 0.01182833,
      "auxiliary_loss_mlp": 0.01032738,
      "balance_loss_clip": 1.05768633,
      "balance_loss_mlp": 1.02494764,
      "epoch": 0.3018096555041183,
      "flos": 21546331200000.0,
      "grad_norm": 1.3697335034048275,
      "language_loss": 0.82241249,
      "learning_rate": 3.2739354400281955e-06,
      "loss": 0.84456819,
      "num_input_tokens_seen": 53952255,
      "step": 2510,
      "time_per_iteration": 2.539440631866455
    },
    {
      "auxiliary_loss_clip": 0.01052113,
      "auxiliary_loss_mlp": 0.00752646,
      "balance_loss_clip": 1.01704717,
      "balance_loss_mlp": 1.00036645,
      "epoch": 0.3019298983947574,
      "flos": 59136294597120.0,
      "grad_norm": 0.8661578115998271,
      "language_loss": 0.63716054,
      "learning_rate": 3.2733348415930744e-06,
      "loss": 0.65520811,
      "num_input_tokens_seen": 54014125,
      "step": 2511,
      "time_per_iteration": 3.191734790802002
    },
    {
      "auxiliary_loss_clip": 0.01148245,
      "auxiliary_loss_mlp": 0.01032909,
      "balance_loss_clip": 1.05212831,
      "balance_loss_mlp": 1.02471375,
      "epoch": 0.3020501412853965,
      "flos": 34423070941440.0,
      "grad_norm": 1.801940441230278,
      "language_loss": 0.79952371,
      "learning_rate": 3.27273404999407e-06,
      "loss": 0.82133526,
      "num_input_tokens_seen": 54036345,
      "step": 2512,
      "time_per_iteration": 2.6981899738311768
    },
    {
      "auxiliary_loss_clip": 0.01062004,
      "auxiliary_loss_mlp": 0.01011936,
      "balance_loss_clip": 1.01759267,
      "balance_loss_mlp": 1.01024365,
      "epoch": 0.3021703841760356,
      "flos": 71008288128000.0,
      "grad_norm": 0.7937731899838413,
      "language_loss": 0.60533571,
      "learning_rate": 3.272133065322322e-06,
      "loss": 0.62607515,
      "num_input_tokens_seen": 54094615,
      "step": 2513,
      "time_per_iteration": 3.8573601245880127
    },
    {
      "auxiliary_loss_clip": 0.01192817,
      "auxiliary_loss_mlp": 0.01027446,
      "balance_loss_clip": 1.05632186,
      "balance_loss_mlp": 1.01931643,
      "epoch": 0.3022906270666747,
      "flos": 21510528318720.0,
      "grad_norm": 1.8612740247388586,
      "language_loss": 0.79622924,
      "learning_rate": 3.271531887669e-06,
      "loss": 0.81843185,
      "num_input_tokens_seen": 54114675,
      "step": 2514,
      "time_per_iteration": 3.246549606323242
    },
    {
      "auxiliary_loss_clip": 0.01142308,
      "auxiliary_loss_mlp": 0.01033896,
      "balance_loss_clip": 1.04935098,
      "balance_loss_mlp": 1.02475858,
      "epoch": 0.30241086995731375,
      "flos": 31132001168640.0,
      "grad_norm": 2.185202647955437,
      "language_loss": 0.63061851,
      "learning_rate": 3.2709305171253015e-06,
      "loss": 0.65238059,
      "num_input_tokens_seen": 54134795,
      "step": 2515,
      "time_per_iteration": 3.432671070098877
    },
    {
      "auxiliary_loss_clip": 0.01182252,
      "auxiliary_loss_mlp": 0.01040488,
      "balance_loss_clip": 1.05780625,
      "balance_loss_mlp": 1.03173208,
      "epoch": 0.30253111284795287,
      "flos": 23511542152320.0,
      "grad_norm": 1.9078184259991826,
      "language_loss": 0.77801979,
      "learning_rate": 3.2703289537824536e-06,
      "loss": 0.80024719,
      "num_input_tokens_seen": 54154595,
      "step": 2516,
      "time_per_iteration": 2.528658151626587
    },
    {
      "auxiliary_loss_clip": 0.01143445,
      "auxiliary_loss_mlp": 0.01035326,
      "balance_loss_clip": 1.05332184,
      "balance_loss_mlp": 1.02689147,
      "epoch": 0.302651355738592,
      "flos": 18725367651840.0,
      "grad_norm": 2.4548310164834413,
      "language_loss": 0.78751862,
      "learning_rate": 3.269727197731714e-06,
      "loss": 0.80930638,
      "num_input_tokens_seen": 54167360,
      "step": 2517,
      "time_per_iteration": 2.5638952255249023
    },
    {
      "auxiliary_loss_clip": 0.01136357,
      "auxiliary_loss_mlp": 0.01033775,
      "balance_loss_clip": 1.05473781,
      "balance_loss_mlp": 1.02509093,
      "epoch": 0.30277159862923103,
      "flos": 22418888382720.0,
      "grad_norm": 1.5790593283914038,
      "language_loss": 0.78082377,
      "learning_rate": 3.269125249064367e-06,
      "loss": 0.80252516,
      "num_input_tokens_seen": 54187055,
      "step": 2518,
      "time_per_iteration": 2.5984466075897217
    },
    {
      "auxiliary_loss_clip": 0.01199686,
      "auxiliary_loss_mlp": 0.01036093,
      "balance_loss_clip": 1.05958676,
      "balance_loss_mlp": 1.02689052,
      "epoch": 0.30289184151987014,
      "flos": 22273126992000.0,
      "grad_norm": 1.5576144307137136,
      "language_loss": 0.83020765,
      "learning_rate": 3.2685231078717297e-06,
      "loss": 0.85256541,
      "num_input_tokens_seen": 54207245,
      "step": 2519,
      "time_per_iteration": 2.4865667819976807
    },
    {
      "auxiliary_loss_clip": 0.01148391,
      "auxiliary_loss_mlp": 0.00762641,
      "balance_loss_clip": 1.05544639,
      "balance_loss_mlp": 1.00035441,
      "epoch": 0.30301208441050925,
      "flos": 25225594231680.0,
      "grad_norm": 1.9020986290103246,
      "language_loss": 0.7547785,
      "learning_rate": 3.267920774245145e-06,
      "loss": 0.77388883,
      "num_input_tokens_seen": 54226650,
      "step": 2520,
      "time_per_iteration": 2.6166739463806152
    },
    {
      "auxiliary_loss_clip": 0.01188565,
      "auxiliary_loss_mlp": 0.01036526,
      "balance_loss_clip": 1.06115317,
      "balance_loss_mlp": 1.02735305,
      "epoch": 0.3031323273011483,
      "flos": 23039245198080.0,
      "grad_norm": 1.8708976004905744,
      "language_loss": 0.84409654,
      "learning_rate": 3.2673182482759876e-06,
      "loss": 0.86634743,
      "num_input_tokens_seen": 54245765,
      "step": 2521,
      "time_per_iteration": 2.525531530380249
    },
    {
      "auxiliary_loss_clip": 0.01181621,
      "auxiliary_loss_mlp": 0.01029863,
      "balance_loss_clip": 1.05840027,
      "balance_loss_mlp": 1.02099955,
      "epoch": 0.3032525701917874,
      "flos": 18876695650560.0,
      "grad_norm": 2.0359406100803747,
      "language_loss": 0.66128957,
      "learning_rate": 3.266715530055659e-06,
      "loss": 0.68340433,
      "num_input_tokens_seen": 54263915,
      "step": 2522,
      "time_per_iteration": 2.496884822845459
    },
    {
      "auxiliary_loss_clip": 0.01171481,
      "auxiliary_loss_mlp": 0.01034556,
      "balance_loss_clip": 1.05373025,
      "balance_loss_mlp": 1.02518034,
      "epoch": 0.30337281308242653,
      "flos": 17782641250560.0,
      "grad_norm": 1.9916194513682184,
      "language_loss": 0.80136335,
      "learning_rate": 3.2661126196755927e-06,
      "loss": 0.82342374,
      "num_input_tokens_seen": 54283025,
      "step": 2523,
      "time_per_iteration": 2.5111122131347656
    },
    {
      "auxiliary_loss_clip": 0.01081551,
      "auxiliary_loss_mlp": 0.01002273,
      "balance_loss_clip": 1.0172832,
      "balance_loss_mlp": 1.00052094,
      "epoch": 0.3034930559730656,
      "flos": 57824298426240.0,
      "grad_norm": 0.7819536297835757,
      "language_loss": 0.55961281,
      "learning_rate": 3.265509517227248e-06,
      "loss": 0.58045107,
      "num_input_tokens_seen": 54339840,
      "step": 2524,
      "time_per_iteration": 3.0484039783477783
    },
    {
      "auxiliary_loss_clip": 0.01166663,
      "auxiliary_loss_mlp": 0.01032368,
      "balance_loss_clip": 1.05383778,
      "balance_loss_mlp": 1.02385604,
      "epoch": 0.3036132988637047,
      "flos": 14755587419520.0,
      "grad_norm": 1.6523018370074398,
      "language_loss": 0.80836821,
      "learning_rate": 3.264906222802115e-06,
      "loss": 0.83035851,
      "num_input_tokens_seen": 54357690,
      "step": 2525,
      "time_per_iteration": 2.509690999984741
    },
    {
      "auxiliary_loss_clip": 0.01198212,
      "auxiliary_loss_mlp": 0.01029357,
      "balance_loss_clip": 1.05898213,
      "balance_loss_mlp": 1.0206306,
      "epoch": 0.30373354175434375,
      "flos": 21033203460480.0,
      "grad_norm": 2.0969381309444013,
      "language_loss": 0.78439939,
      "learning_rate": 3.264302736491715e-06,
      "loss": 0.80667502,
      "num_input_tokens_seen": 54377810,
      "step": 2526,
      "time_per_iteration": 2.4733009338378906
    },
    {
      "auxiliary_loss_clip": 0.01181131,
      "auxiliary_loss_mlp": 0.01032471,
      "balance_loss_clip": 1.0604316,
      "balance_loss_mlp": 1.02381682,
      "epoch": 0.30385378464498286,
      "flos": 21143233797120.0,
      "grad_norm": 2.168278587142872,
      "language_loss": 0.87450337,
      "learning_rate": 3.263699058387594e-06,
      "loss": 0.89663947,
      "num_input_tokens_seen": 54395245,
      "step": 2527,
      "time_per_iteration": 2.5022695064544678
    },
    {
      "auxiliary_loss_clip": 0.01149483,
      "auxiliary_loss_mlp": 0.01044465,
      "balance_loss_clip": 1.05073667,
      "balance_loss_mlp": 1.03564978,
      "epoch": 0.30397402753562197,
      "flos": 20629244131200.0,
      "grad_norm": 2.1181542033417915,
      "language_loss": 0.90261078,
      "learning_rate": 3.2630951885813315e-06,
      "loss": 0.92455029,
      "num_input_tokens_seen": 54412640,
      "step": 2528,
      "time_per_iteration": 2.5604114532470703
    },
    {
      "auxiliary_loss_clip": 0.01166555,
      "auxiliary_loss_mlp": 0.01028375,
      "balance_loss_clip": 1.05124426,
      "balance_loss_mlp": 1.01988101,
      "epoch": 0.304094270426261,
      "flos": 15085678429440.0,
      "grad_norm": 2.1007850549966314,
      "language_loss": 0.78152549,
      "learning_rate": 3.262491127164533e-06,
      "loss": 0.80347478,
      "num_input_tokens_seen": 54431455,
      "step": 2529,
      "time_per_iteration": 2.523890972137451
    },
    {
      "auxiliary_loss_clip": 0.01173163,
      "auxiliary_loss_mlp": 0.0076309,
      "balance_loss_clip": 1.05515623,
      "balance_loss_mlp": 1.00034738,
      "epoch": 0.30421451331690014,
      "flos": 13845216193920.0,
      "grad_norm": 2.255856340570019,
      "language_loss": 0.79805803,
      "learning_rate": 3.2618868742288337e-06,
      "loss": 0.8174206,
      "num_input_tokens_seen": 54448380,
      "step": 2530,
      "time_per_iteration": 2.5040972232818604
    },
    {
      "auxiliary_loss_clip": 0.01182388,
      "auxiliary_loss_mlp": 0.01031417,
      "balance_loss_clip": 1.05719423,
      "balance_loss_mlp": 1.02291715,
      "epoch": 0.30433475620753925,
      "flos": 17384212615680.0,
      "grad_norm": 2.017622800504052,
      "language_loss": 0.72303164,
      "learning_rate": 3.261282429865899e-06,
      "loss": 0.74516976,
      "num_input_tokens_seen": 54466385,
      "step": 2531,
      "time_per_iteration": 2.483461380004883
    },
    {
      "auxiliary_loss_clip": 0.01175824,
      "auxiliary_loss_mlp": 0.00761999,
      "balance_loss_clip": 1.05923438,
      "balance_loss_mlp": 1.00039148,
      "epoch": 0.3044549990981783,
      "flos": 18916951818240.0,
      "grad_norm": 1.5953528267098787,
      "language_loss": 0.72071534,
      "learning_rate": 3.2606777941674225e-06,
      "loss": 0.74009359,
      "num_input_tokens_seen": 54485040,
      "step": 2532,
      "time_per_iteration": 2.5249242782592773
    },
    {
      "auxiliary_loss_clip": 0.01130786,
      "auxiliary_loss_mlp": 0.01034272,
      "balance_loss_clip": 1.05267215,
      "balance_loss_mlp": 1.02540898,
      "epoch": 0.3045752419888174,
      "flos": 21068431724160.0,
      "grad_norm": 1.909625361472053,
      "language_loss": 0.84179604,
      "learning_rate": 3.2600729672251276e-06,
      "loss": 0.86344659,
      "num_input_tokens_seen": 54502755,
      "step": 2533,
      "time_per_iteration": 3.365257740020752
    },
    {
      "auxiliary_loss_clip": 0.01198087,
      "auxiliary_loss_mlp": 0.00762996,
      "balance_loss_clip": 1.06117153,
      "balance_loss_mlp": 1.00040221,
      "epoch": 0.3046954848794565,
      "flos": 29096405516160.0,
      "grad_norm": 1.850463472491268,
      "language_loss": 0.65773833,
      "learning_rate": 3.259467949130765e-06,
      "loss": 0.67734909,
      "num_input_tokens_seen": 54524165,
      "step": 2534,
      "time_per_iteration": 2.5522944927215576
    },
    {
      "auxiliary_loss_clip": 0.01175266,
      "auxiliary_loss_mlp": 0.01034426,
      "balance_loss_clip": 1.06070638,
      "balance_loss_mlp": 1.02537215,
      "epoch": 0.3048157277700956,
      "flos": 20295346279680.0,
      "grad_norm": 2.0144911337319096,
      "language_loss": 0.82292122,
      "learning_rate": 3.2588627399761164e-06,
      "loss": 0.84501815,
      "num_input_tokens_seen": 54540160,
      "step": 2535,
      "time_per_iteration": 2.541459798812866
    },
    {
      "auxiliary_loss_clip": 0.01169891,
      "auxiliary_loss_mlp": 0.01033808,
      "balance_loss_clip": 1.05719399,
      "balance_loss_mlp": 1.02512908,
      "epoch": 0.3049359706607347,
      "flos": 22739929165440.0,
      "grad_norm": 1.599523483125588,
      "language_loss": 0.70778739,
      "learning_rate": 3.2582573398529903e-06,
      "loss": 0.72982442,
      "num_input_tokens_seen": 54557515,
      "step": 2536,
      "time_per_iteration": 2.552891492843628
    },
    {
      "auxiliary_loss_clip": 0.01152113,
      "auxiliary_loss_mlp": 0.01034635,
      "balance_loss_clip": 1.05140316,
      "balance_loss_mlp": 1.02549207,
      "epoch": 0.3050562135513738,
      "flos": 18434634969600.0,
      "grad_norm": 2.4436812299564403,
      "language_loss": 0.74142241,
      "learning_rate": 3.2576517488532265e-06,
      "loss": 0.76328993,
      "num_input_tokens_seen": 54573865,
      "step": 2537,
      "time_per_iteration": 2.526862382888794
    },
    {
      "auxiliary_loss_clip": 0.01180356,
      "auxiliary_loss_mlp": 0.01030149,
      "balance_loss_clip": 1.05485594,
      "balance_loss_mlp": 1.02172065,
      "epoch": 0.30517645644201286,
      "flos": 20370327920640.0,
      "grad_norm": 1.7330459882854072,
      "language_loss": 0.87365425,
      "learning_rate": 3.257045967068692e-06,
      "loss": 0.89575928,
      "num_input_tokens_seen": 54593120,
      "step": 2538,
      "time_per_iteration": 2.5266571044921875
    },
    {
      "auxiliary_loss_clip": 0.01198296,
      "auxiliary_loss_mlp": 0.0103558,
      "balance_loss_clip": 1.05963683,
      "balance_loss_mlp": 1.02646089,
      "epoch": 0.30529669933265197,
      "flos": 21945118970880.0,
      "grad_norm": 1.920468654371795,
      "language_loss": 0.82761389,
      "learning_rate": 3.2564399945912848e-06,
      "loss": 0.8499527,
      "num_input_tokens_seen": 54612910,
      "step": 2539,
      "time_per_iteration": 3.9914872646331787
    },
    {
      "auxiliary_loss_clip": 0.01140932,
      "auxiliary_loss_mlp": 0.01030244,
      "balance_loss_clip": 1.0504396,
      "balance_loss_mlp": 1.02242374,
      "epoch": 0.305416942223291,
      "flos": 21835411856640.0,
      "grad_norm": 2.228797552218803,
      "language_loss": 0.8238256,
      "learning_rate": 3.2558338315129287e-06,
      "loss": 0.8455373,
      "num_input_tokens_seen": 54631055,
      "step": 2540,
      "time_per_iteration": 2.631718397140503
    },
    {
      "auxiliary_loss_clip": 0.0117839,
      "auxiliary_loss_mlp": 0.01042521,
      "balance_loss_clip": 1.05852389,
      "balance_loss_mlp": 1.03335428,
      "epoch": 0.30553718511393013,
      "flos": 33911810709120.0,
      "grad_norm": 2.6107476069767293,
      "language_loss": 0.75773156,
      "learning_rate": 3.2552274779255785e-06,
      "loss": 0.77994066,
      "num_input_tokens_seen": 54651985,
      "step": 2541,
      "time_per_iteration": 3.440293073654175
    },
    {
      "auxiliary_loss_clip": 0.01183047,
      "auxiliary_loss_mlp": 0.01036176,
      "balance_loss_clip": 1.05930781,
      "balance_loss_mlp": 1.02758121,
      "epoch": 0.30565742800456924,
      "flos": 22268530051200.0,
      "grad_norm": 2.038288059235117,
      "language_loss": 0.76856625,
      "learning_rate": 3.2546209339212184e-06,
      "loss": 0.79075849,
      "num_input_tokens_seen": 54671005,
      "step": 2542,
      "time_per_iteration": 2.507427930831909
    },
    {
      "auxiliary_loss_clip": 0.0116874,
      "auxiliary_loss_mlp": 0.01033279,
      "balance_loss_clip": 1.05371165,
      "balance_loss_mlp": 1.02434397,
      "epoch": 0.3057776708952083,
      "flos": 22565044823040.0,
      "grad_norm": 1.724628098492537,
      "language_loss": 0.77561647,
      "learning_rate": 3.25401419959186e-06,
      "loss": 0.79763663,
      "num_input_tokens_seen": 54691615,
      "step": 2543,
      "time_per_iteration": 2.576265573501587
    },
    {
      "auxiliary_loss_clip": 0.0118407,
      "auxiliary_loss_mlp": 0.01038356,
      "balance_loss_clip": 1.06359279,
      "balance_loss_mlp": 1.02961802,
      "epoch": 0.3058979137858474,
      "flos": 21799213925760.0,
      "grad_norm": 1.7725735004470247,
      "language_loss": 0.76260662,
      "learning_rate": 3.253407275029545e-06,
      "loss": 0.78483087,
      "num_input_tokens_seen": 54710520,
      "step": 2544,
      "time_per_iteration": 2.529459238052368
    },
    {
      "auxiliary_loss_clip": 0.01160519,
      "auxiliary_loss_mlp": 0.01051837,
      "balance_loss_clip": 1.05822229,
      "balance_loss_mlp": 1.04193652,
      "epoch": 0.3060181566764865,
      "flos": 26979435601920.0,
      "grad_norm": 1.8135670480119648,
      "language_loss": 0.80179894,
      "learning_rate": 3.2528001603263425e-06,
      "loss": 0.82392251,
      "num_input_tokens_seen": 54732590,
      "step": 2545,
      "time_per_iteration": 2.617621421813965
    },
    {
      "auxiliary_loss_clip": 0.01181743,
      "auxiliary_loss_mlp": 0.01033662,
      "balance_loss_clip": 1.05999923,
      "balance_loss_mlp": 1.02489471,
      "epoch": 0.3061383995671256,
      "flos": 19865101173120.0,
      "grad_norm": 2.613676724602542,
      "language_loss": 0.81630123,
      "learning_rate": 3.2521928555743514e-06,
      "loss": 0.83845532,
      "num_input_tokens_seen": 54749935,
      "step": 2546,
      "time_per_iteration": 2.507024049758911
    },
    {
      "auxiliary_loss_clip": 0.01163451,
      "auxiliary_loss_mlp": 0.00763142,
      "balance_loss_clip": 1.0545032,
      "balance_loss_mlp": 1.00041592,
      "epoch": 0.3062586424577647,
      "flos": 22127509255680.0,
      "grad_norm": 2.21806823898555,
      "language_loss": 0.67127484,
      "learning_rate": 3.2515853608657e-06,
      "loss": 0.69054079,
      "num_input_tokens_seen": 54767935,
      "step": 2547,
      "time_per_iteration": 2.56483793258667
    },
    {
      "auxiliary_loss_clip": 0.01178929,
      "auxiliary_loss_mlp": 0.01035601,
      "balance_loss_clip": 1.05679226,
      "balance_loss_mlp": 1.02654147,
      "epoch": 0.3063788853484038,
      "flos": 20845497962880.0,
      "grad_norm": 2.0680961327915406,
      "language_loss": 0.75716972,
      "learning_rate": 3.250977676292545e-06,
      "loss": 0.77931499,
      "num_input_tokens_seen": 54786175,
      "step": 2548,
      "time_per_iteration": 2.5004711151123047
    },
    {
      "auxiliary_loss_clip": 0.0117219,
      "auxiliary_loss_mlp": 0.01032254,
      "balance_loss_clip": 1.05743444,
      "balance_loss_mlp": 1.02376068,
      "epoch": 0.30649912823904285,
      "flos": 16209717707520.0,
      "grad_norm": 2.1158165630208905,
      "language_loss": 0.79423428,
      "learning_rate": 3.2503698019470712e-06,
      "loss": 0.8162787,
      "num_input_tokens_seen": 54801945,
      "step": 2549,
      "time_per_iteration": 2.5124754905700684
    },
    {
      "auxiliary_loss_clip": 0.01180804,
      "auxiliary_loss_mlp": 0.01035935,
      "balance_loss_clip": 1.05454206,
      "balance_loss_mlp": 1.02642262,
      "epoch": 0.30661937112968196,
      "flos": 18617815353600.0,
      "grad_norm": 3.0040414726095026,
      "language_loss": 0.78617847,
      "learning_rate": 3.249761737921492e-06,
      "loss": 0.80834585,
      "num_input_tokens_seen": 54818475,
      "step": 2550,
      "time_per_iteration": 2.4747815132141113
    },
    {
      "auxiliary_loss_clip": 0.0116837,
      "auxiliary_loss_mlp": 0.01033765,
      "balance_loss_clip": 1.05945778,
      "balance_loss_mlp": 1.02576065,
      "epoch": 0.30673961402032107,
      "flos": 31390809638400.0,
      "grad_norm": 1.8893010958670406,
      "language_loss": 0.74109215,
      "learning_rate": 3.249153484308051e-06,
      "loss": 0.7631135,
      "num_input_tokens_seen": 54837090,
      "step": 2551,
      "time_per_iteration": 2.611985921859741
    },
    {
      "auxiliary_loss_clip": 0.0112893,
      "auxiliary_loss_mlp": 0.0103227,
      "balance_loss_clip": 1.05196702,
      "balance_loss_mlp": 1.02314484,
      "epoch": 0.3068598569109601,
      "flos": 20229809915520.0,
      "grad_norm": 2.0486994898444397,
      "language_loss": 0.77772915,
      "learning_rate": 3.2485450411990194e-06,
      "loss": 0.7993412,
      "num_input_tokens_seen": 54856445,
      "step": 2552,
      "time_per_iteration": 2.5905821323394775
    },
    {
      "auxiliary_loss_clip": 0.01198225,
      "auxiliary_loss_mlp": 0.01037468,
      "balance_loss_clip": 1.05861413,
      "balance_loss_mlp": 1.02852178,
      "epoch": 0.30698009980159924,
      "flos": 29601991399680.0,
      "grad_norm": 2.9932346671863628,
      "language_loss": 0.82199484,
      "learning_rate": 3.2479364086866983e-06,
      "loss": 0.84435177,
      "num_input_tokens_seen": 54876700,
      "step": 2553,
      "time_per_iteration": 2.5804643630981445
    },
    {
      "auxiliary_loss_clip": 0.01170804,
      "auxiliary_loss_mlp": 0.00762814,
      "balance_loss_clip": 1.05864286,
      "balance_loss_mlp": 1.00042319,
      "epoch": 0.30710034269223835,
      "flos": 23842423261440.0,
      "grad_norm": 1.8612303810835158,
      "language_loss": 0.81566066,
      "learning_rate": 3.247327586863416e-06,
      "loss": 0.83499682,
      "num_input_tokens_seen": 54897580,
      "step": 2554,
      "time_per_iteration": 2.575174570083618
    },
    {
      "auxiliary_loss_clip": 0.01161849,
      "auxiliary_loss_mlp": 0.01032091,
      "balance_loss_clip": 1.05632138,
      "balance_loss_mlp": 1.02351451,
      "epoch": 0.3072205855828774,
      "flos": 25884986152320.0,
      "grad_norm": 2.113861015884066,
      "language_loss": 0.76875246,
      "learning_rate": 3.2467185758215304e-06,
      "loss": 0.79069191,
      "num_input_tokens_seen": 54917320,
      "step": 2555,
      "time_per_iteration": 2.6148173809051514
    },
    {
      "auxiliary_loss_clip": 0.01158945,
      "auxiliary_loss_mlp": 0.00762409,
      "balance_loss_clip": 1.05702245,
      "balance_loss_mlp": 1.00041604,
      "epoch": 0.3073408284735165,
      "flos": 22236390357120.0,
      "grad_norm": 2.6073137912295334,
      "language_loss": 0.85361958,
      "learning_rate": 3.246109375653428e-06,
      "loss": 0.87283313,
      "num_input_tokens_seen": 54934085,
      "step": 2556,
      "time_per_iteration": 2.5625100135803223
    },
    {
      "auxiliary_loss_clip": 0.01198068,
      "auxiliary_loss_mlp": 0.01034281,
      "balance_loss_clip": 1.05988526,
      "balance_loss_mlp": 1.02507818,
      "epoch": 0.30746107136415557,
      "flos": 19500284689920.0,
      "grad_norm": 1.8380235341051823,
      "language_loss": 0.78277516,
      "learning_rate": 3.2454999864515243e-06,
      "loss": 0.80509865,
      "num_input_tokens_seen": 54953460,
      "step": 2557,
      "time_per_iteration": 2.463102340698242
    },
    {
      "auxiliary_loss_clip": 0.01160482,
      "auxiliary_loss_mlp": 0.00762447,
      "balance_loss_clip": 1.0546912,
      "balance_loss_mlp": 1.00040495,
      "epoch": 0.3075813142547947,
      "flos": 21724806902400.0,
      "grad_norm": 1.6634258851764445,
      "language_loss": 0.69088495,
      "learning_rate": 3.244890408308263e-06,
      "loss": 0.71011424,
      "num_input_tokens_seen": 54974165,
      "step": 2558,
      "time_per_iteration": 2.535918951034546
    },
    {
      "auxiliary_loss_clip": 0.01142064,
      "auxiliary_loss_mlp": 0.01031777,
      "balance_loss_clip": 1.05249047,
      "balance_loss_mlp": 1.02312827,
      "epoch": 0.3077015571454338,
      "flos": 24097963593600.0,
      "grad_norm": 2.267915880092834,
      "language_loss": 0.60576057,
      "learning_rate": 3.2442806413161165e-06,
      "loss": 0.62749898,
      "num_input_tokens_seen": 54993810,
      "step": 2559,
      "time_per_iteration": 3.3876137733459473
    },
    {
      "auxiliary_loss_clip": 0.01144073,
      "auxiliary_loss_mlp": 0.01035852,
      "balance_loss_clip": 1.05376577,
      "balance_loss_mlp": 1.0264883,
      "epoch": 0.30782180003607285,
      "flos": 18405476104320.0,
      "grad_norm": 2.012512699051081,
      "language_loss": 0.75576341,
      "learning_rate": 3.243670685567586e-06,
      "loss": 0.77756262,
      "num_input_tokens_seen": 55011210,
      "step": 2560,
      "time_per_iteration": 2.555206537246704
    },
    {
      "auxiliary_loss_clip": 0.01167069,
      "auxiliary_loss_mlp": 0.00761814,
      "balance_loss_clip": 1.05464947,
      "balance_loss_mlp": 1.00040722,
      "epoch": 0.30794204292671196,
      "flos": 23878549365120.0,
      "grad_norm": 2.158706153934657,
      "language_loss": 0.80485928,
      "learning_rate": 3.2430605411552012e-06,
      "loss": 0.82414818,
      "num_input_tokens_seen": 55031325,
      "step": 2561,
      "time_per_iteration": 2.547567844390869
    },
    {
      "auxiliary_loss_clip": 0.01057706,
      "auxiliary_loss_mlp": 0.01006201,
      "balance_loss_clip": 1.02114463,
      "balance_loss_mlp": 1.00437737,
      "epoch": 0.30806228581735107,
      "flos": 67927800816000.0,
      "grad_norm": 0.9006923252131376,
      "language_loss": 0.70524442,
      "learning_rate": 3.2424502081715205e-06,
      "loss": 0.72588348,
      "num_input_tokens_seen": 55094440,
      "step": 2562,
      "time_per_iteration": 3.18129301071167
    },
    {
      "auxiliary_loss_clip": 0.01169608,
      "auxiliary_loss_mlp": 0.0102933,
      "balance_loss_clip": 1.05658698,
      "balance_loss_mlp": 1.02056825,
      "epoch": 0.3081825287079901,
      "flos": 23843213360640.0,
      "grad_norm": 1.747616828451794,
      "language_loss": 0.77967119,
      "learning_rate": 3.241839686709132e-06,
      "loss": 0.80166054,
      "num_input_tokens_seen": 55115375,
      "step": 2563,
      "time_per_iteration": 2.5585925579071045
    },
    {
      "auxiliary_loss_clip": 0.01182333,
      "auxiliary_loss_mlp": 0.0103549,
      "balance_loss_clip": 1.05673349,
      "balance_loss_mlp": 1.02569675,
      "epoch": 0.30830277159862923,
      "flos": 16209969102720.0,
      "grad_norm": 2.4862505534369173,
      "language_loss": 0.81844336,
      "learning_rate": 3.2412289768606495e-06,
      "loss": 0.84062159,
      "num_input_tokens_seen": 55131945,
      "step": 2564,
      "time_per_iteration": 2.4799702167510986
    },
    {
      "auxiliary_loss_clip": 0.01187946,
      "auxiliary_loss_mlp": 0.01027215,
      "balance_loss_clip": 1.06035614,
      "balance_loss_mlp": 1.0187335,
      "epoch": 0.30842301448926834,
      "flos": 29349503723520.0,
      "grad_norm": 1.9886474931816278,
      "language_loss": 0.82627678,
      "learning_rate": 3.240618078718718e-06,
      "loss": 0.84842837,
      "num_input_tokens_seen": 55153405,
      "step": 2565,
      "time_per_iteration": 3.386650323867798
    },
    {
      "auxiliary_loss_clip": 0.01148991,
      "auxiliary_loss_mlp": 0.01029054,
      "balance_loss_clip": 1.05012143,
      "balance_loss_mlp": 1.01930296,
      "epoch": 0.3085432573799074,
      "flos": 21945190798080.0,
      "grad_norm": 2.0091404206652763,
      "language_loss": 0.7415874,
      "learning_rate": 3.240006992376011e-06,
      "loss": 0.76336789,
      "num_input_tokens_seen": 55173030,
      "step": 2566,
      "time_per_iteration": 2.591628074645996
    },
    {
      "auxiliary_loss_clip": 0.01173963,
      "auxiliary_loss_mlp": 0.01036792,
      "balance_loss_clip": 1.05764103,
      "balance_loss_mlp": 1.02825665,
      "epoch": 0.3086635002705465,
      "flos": 22054718344320.0,
      "grad_norm": 2.255674596849611,
      "language_loss": 0.76140141,
      "learning_rate": 3.2393957179252284e-06,
      "loss": 0.78350902,
      "num_input_tokens_seen": 55189565,
      "step": 2567,
      "time_per_iteration": 2.540149211883545
    },
    {
      "auxiliary_loss_clip": 0.01202387,
      "auxiliary_loss_mlp": 0.0102873,
      "balance_loss_clip": 1.06331396,
      "balance_loss_mlp": 1.01974154,
      "epoch": 0.3087837431611856,
      "flos": 32665925520000.0,
      "grad_norm": 1.8690733128055979,
      "language_loss": 0.80316991,
      "learning_rate": 3.2387842554591016e-06,
      "loss": 0.82548106,
      "num_input_tokens_seen": 55210380,
      "step": 2568,
      "time_per_iteration": 3.363516330718994
    },
    {
      "auxiliary_loss_clip": 0.01200829,
      "auxiliary_loss_mlp": 0.01038736,
      "balance_loss_clip": 1.06093061,
      "balance_loss_mlp": 1.02927351,
      "epoch": 0.3089039860518247,
      "flos": 17599245384960.0,
      "grad_norm": 2.213299967552078,
      "language_loss": 0.87851357,
      "learning_rate": 3.238172605070388e-06,
      "loss": 0.90090919,
      "num_input_tokens_seen": 55225795,
      "step": 2569,
      "time_per_iteration": 2.441365957260132
    },
    {
      "auxiliary_loss_clip": 0.01183529,
      "auxiliary_loss_mlp": 0.00763253,
      "balance_loss_clip": 1.05926013,
      "balance_loss_mlp": 1.0003866,
      "epoch": 0.3090242289424638,
      "flos": 14383839611520.0,
      "grad_norm": 2.4240794206290444,
      "language_loss": 0.78073436,
      "learning_rate": 3.2375607668518745e-06,
      "loss": 0.80020213,
      "num_input_tokens_seen": 55238830,
      "step": 2570,
      "time_per_iteration": 2.4520339965820312
    },
    {
      "auxiliary_loss_clip": 0.01162413,
      "auxiliary_loss_mlp": 0.01036923,
      "balance_loss_clip": 1.056445,
      "balance_loss_mlp": 1.02807188,
      "epoch": 0.30914447183310284,
      "flos": 16068625084800.0,
      "grad_norm": 2.070592753815051,
      "language_loss": 0.89806199,
      "learning_rate": 3.236948740896377e-06,
      "loss": 0.92005527,
      "num_input_tokens_seen": 55253630,
      "step": 2571,
      "time_per_iteration": 2.5365450382232666
    },
    {
      "auxiliary_loss_clip": 0.011845,
      "auxiliary_loss_mlp": 0.01035248,
      "balance_loss_clip": 1.05976129,
      "balance_loss_mlp": 1.02544296,
      "epoch": 0.30926471472374195,
      "flos": 32230221546240.0,
      "grad_norm": 1.6345415454869123,
      "language_loss": 0.84388793,
      "learning_rate": 3.2363365272967384e-06,
      "loss": 0.86608541,
      "num_input_tokens_seen": 55276200,
      "step": 2572,
      "time_per_iteration": 2.610645055770874
    },
    {
      "auxiliary_loss_clip": 0.01184935,
      "auxiliary_loss_mlp": 0.01036437,
      "balance_loss_clip": 1.06204081,
      "balance_loss_mlp": 1.02686501,
      "epoch": 0.30938495761438106,
      "flos": 20370722970240.0,
      "grad_norm": 1.960744900784151,
      "language_loss": 0.81505418,
      "learning_rate": 3.235724126145832e-06,
      "loss": 0.83726794,
      "num_input_tokens_seen": 55292235,
      "step": 2573,
      "time_per_iteration": 2.4930341243743896
    },
    {
      "auxiliary_loss_clip": 0.01171881,
      "auxiliary_loss_mlp": 0.01032802,
      "balance_loss_clip": 1.05516481,
      "balance_loss_mlp": 1.0237484,
      "epoch": 0.3095052005050201,
      "flos": 24061155131520.0,
      "grad_norm": 1.5050249768349608,
      "language_loss": 0.77745706,
      "learning_rate": 3.235111537536558e-06,
      "loss": 0.79950392,
      "num_input_tokens_seen": 55313050,
      "step": 2574,
      "time_per_iteration": 2.524223566055298
    },
    {
      "auxiliary_loss_clip": 0.01184952,
      "auxiliary_loss_mlp": 0.0102691,
      "balance_loss_clip": 1.05885553,
      "balance_loss_mlp": 1.01810682,
      "epoch": 0.30962544339565923,
      "flos": 23401547729280.0,
      "grad_norm": 1.9214351489062536,
      "language_loss": 0.82769883,
      "learning_rate": 3.2344987615618456e-06,
      "loss": 0.84981745,
      "num_input_tokens_seen": 55332885,
      "step": 2575,
      "time_per_iteration": 2.5329270362854004
    },
    {
      "auxiliary_loss_clip": 0.01155351,
      "auxiliary_loss_mlp": 0.01033813,
      "balance_loss_clip": 1.05764699,
      "balance_loss_mlp": 1.02538466,
      "epoch": 0.30974568628629834,
      "flos": 33799984692480.0,
      "grad_norm": 1.9556408980930045,
      "language_loss": 0.78194308,
      "learning_rate": 3.2338857983146533e-06,
      "loss": 0.8038348,
      "num_input_tokens_seen": 55354385,
      "step": 2576,
      "time_per_iteration": 2.677990674972534
    },
    {
      "auxiliary_loss_clip": 0.01163062,
      "auxiliary_loss_mlp": 0.01036515,
      "balance_loss_clip": 1.05879855,
      "balance_loss_mlp": 1.02713919,
      "epoch": 0.3098659291769374,
      "flos": 20229594433920.0,
      "grad_norm": 1.760467006113623,
      "language_loss": 0.76553208,
      "learning_rate": 3.233272647887966e-06,
      "loss": 0.7875278,
      "num_input_tokens_seen": 55373275,
      "step": 2577,
      "time_per_iteration": 2.530245542526245
    },
    {
      "auxiliary_loss_clip": 0.01199314,
      "auxiliary_loss_mlp": 0.01030704,
      "balance_loss_clip": 1.06124723,
      "balance_loss_mlp": 1.02198398,
      "epoch": 0.3099861720675765,
      "flos": 24748556682240.0,
      "grad_norm": 1.6250165243498411,
      "language_loss": 0.9014411,
      "learning_rate": 3.2326593103747985e-06,
      "loss": 0.92374122,
      "num_input_tokens_seen": 55392290,
      "step": 2578,
      "time_per_iteration": 2.4958388805389404
    },
    {
      "auxiliary_loss_clip": 0.01184279,
      "auxiliary_loss_mlp": 0.01038896,
      "balance_loss_clip": 1.06096125,
      "balance_loss_mlp": 1.03059363,
      "epoch": 0.3101064149582156,
      "flos": 11765485704960.0,
      "grad_norm": 2.186232622402892,
      "language_loss": 0.85363328,
      "learning_rate": 3.2320457858681936e-06,
      "loss": 0.87586498,
      "num_input_tokens_seen": 55410680,
      "step": 2579,
      "time_per_iteration": 2.487229108810425
    },
    {
      "auxiliary_loss_clip": 0.01167997,
      "auxiliary_loss_mlp": 0.01039804,
      "balance_loss_clip": 1.05534267,
      "balance_loss_mlp": 1.03137553,
      "epoch": 0.31022665784885467,
      "flos": 23033247626880.0,
      "grad_norm": 2.7059933582434654,
      "language_loss": 0.85236573,
      "learning_rate": 3.2314320744612228e-06,
      "loss": 0.87444377,
      "num_input_tokens_seen": 55425980,
      "step": 2580,
      "time_per_iteration": 2.5190415382385254
    },
    {
      "auxiliary_loss_clip": 0.0118183,
      "auxiliary_loss_mlp": 0.01032724,
      "balance_loss_clip": 1.05926776,
      "balance_loss_mlp": 1.02445078,
      "epoch": 0.3103469007394938,
      "flos": 16289188548480.0,
      "grad_norm": 1.5164846884592809,
      "language_loss": 0.76435959,
      "learning_rate": 3.2308181762469854e-06,
      "loss": 0.78650516,
      "num_input_tokens_seen": 55443925,
      "step": 2581,
      "time_per_iteration": 2.4898293018341064
    },
    {
      "auxiliary_loss_clip": 0.01204998,
      "auxiliary_loss_mlp": 0.01045996,
      "balance_loss_clip": 1.06319666,
      "balance_loss_mlp": 1.03603017,
      "epoch": 0.3104671436301329,
      "flos": 30515271626880.0,
      "grad_norm": 2.1156745469573432,
      "language_loss": 0.78442025,
      "learning_rate": 3.230204091318609e-06,
      "loss": 0.80693024,
      "num_input_tokens_seen": 55464465,
      "step": 2582,
      "time_per_iteration": 2.536320447921753
    },
    {
      "auxiliary_loss_clip": 0.011962,
      "auxiliary_loss_mlp": 0.00762368,
      "balance_loss_clip": 1.05881882,
      "balance_loss_mlp": 1.00037527,
      "epoch": 0.31058738652077195,
      "flos": 20047240062720.0,
      "grad_norm": 2.125422123524982,
      "language_loss": 0.84495246,
      "learning_rate": 3.2295898197692503e-06,
      "loss": 0.86453819,
      "num_input_tokens_seen": 55483425,
      "step": 2583,
      "time_per_iteration": 2.4711720943450928
    },
    {
      "auxiliary_loss_clip": 0.01196826,
      "auxiliary_loss_mlp": 0.01030056,
      "balance_loss_clip": 1.05853045,
      "balance_loss_mlp": 1.02132416,
      "epoch": 0.31070762941141106,
      "flos": 28074639237120.0,
      "grad_norm": 1.6306482694965112,
      "language_loss": 0.79127759,
      "learning_rate": 3.228975361692094e-06,
      "loss": 0.81354636,
      "num_input_tokens_seen": 55504445,
      "step": 2584,
      "time_per_iteration": 3.3604116439819336
    },
    {
      "auxiliary_loss_clip": 0.01187555,
      "auxiliary_loss_mlp": 0.0076323,
      "balance_loss_clip": 1.05737114,
      "balance_loss_mlp": 1.0003922,
      "epoch": 0.31082787230205017,
      "flos": 20521907314560.0,
      "grad_norm": 2.1715454412058417,
      "language_loss": 0.80287743,
      "learning_rate": 3.228360717180352e-06,
      "loss": 0.82238531,
      "num_input_tokens_seen": 55521970,
      "step": 2585,
      "time_per_iteration": 2.521383047103882
    },
    {
      "auxiliary_loss_clip": 0.01084792,
      "auxiliary_loss_mlp": 0.00752521,
      "balance_loss_clip": 1.02096081,
      "balance_loss_mlp": 1.00038707,
      "epoch": 0.3109481151926892,
      "flos": 62445928723200.0,
      "grad_norm": 0.8340359122380687,
      "language_loss": 0.59446681,
      "learning_rate": 3.227745886327266e-06,
      "loss": 0.61283994,
      "num_input_tokens_seen": 55580665,
      "step": 2586,
      "time_per_iteration": 3.0363762378692627
    },
    {
      "auxiliary_loss_clip": 0.01085487,
      "auxiliary_loss_mlp": 0.01014345,
      "balance_loss_clip": 1.02131474,
      "balance_loss_mlp": 1.01268756,
      "epoch": 0.31106835808332833,
      "flos": 44746744723200.0,
      "grad_norm": 0.8143845582890822,
      "language_loss": 0.55838692,
      "learning_rate": 3.227130869226105e-06,
      "loss": 0.57938522,
      "num_input_tokens_seen": 55637825,
      "step": 2587,
      "time_per_iteration": 3.05106258392334
    },
    {
      "auxiliary_loss_clip": 0.01183869,
      "auxiliary_loss_mlp": 0.01030405,
      "balance_loss_clip": 1.0571394,
      "balance_loss_mlp": 1.02188742,
      "epoch": 0.3111886009739674,
      "flos": 23403056100480.0,
      "grad_norm": 2.468970215951078,
      "language_loss": 0.82798207,
      "learning_rate": 3.226515665970167e-06,
      "loss": 0.85012478,
      "num_input_tokens_seen": 55655365,
      "step": 2588,
      "time_per_iteration": 2.5338547229766846
    },
    {
      "auxiliary_loss_clip": 0.01179685,
      "auxiliary_loss_mlp": 0.01037407,
      "balance_loss_clip": 1.05637288,
      "balance_loss_mlp": 1.02788877,
      "epoch": 0.3113088438646065,
      "flos": 17530728192000.0,
      "grad_norm": 2.3552037820621528,
      "language_loss": 0.86643624,
      "learning_rate": 3.225900276652777e-06,
      "loss": 0.8886072,
      "num_input_tokens_seen": 55672140,
      "step": 2589,
      "time_per_iteration": 2.470212697982788
    },
    {
      "auxiliary_loss_clip": 0.01176611,
      "auxiliary_loss_mlp": 0.01028521,
      "balance_loss_clip": 1.05768454,
      "balance_loss_mlp": 1.01938939,
      "epoch": 0.3114290867552456,
      "flos": 28365802882560.0,
      "grad_norm": 1.5958811305156895,
      "language_loss": 0.75541049,
      "learning_rate": 3.2252847013672906e-06,
      "loss": 0.77746189,
      "num_input_tokens_seen": 55694800,
      "step": 2590,
      "time_per_iteration": 2.618248224258423
    },
    {
      "auxiliary_loss_clip": 0.01145971,
      "auxiliary_loss_mlp": 0.010323,
      "balance_loss_clip": 1.05298853,
      "balance_loss_mlp": 1.02383578,
      "epoch": 0.31154932964588467,
      "flos": 27379157126400.0,
      "grad_norm": 1.8960331274087485,
      "language_loss": 0.75863194,
      "learning_rate": 3.224668940207089e-06,
      "loss": 0.78041464,
      "num_input_tokens_seen": 55713785,
      "step": 2591,
      "time_per_iteration": 4.065414190292358
    },
    {
      "auxiliary_loss_clip": 0.01132102,
      "auxiliary_loss_mlp": 0.01034157,
      "balance_loss_clip": 1.04990804,
      "balance_loss_mlp": 1.02453053,
      "epoch": 0.3116695725365238,
      "flos": 26541864120960.0,
      "grad_norm": 2.3134378026421927,
      "language_loss": 0.86128932,
      "learning_rate": 3.2240529932655828e-06,
      "loss": 0.88295192,
      "num_input_tokens_seen": 55733050,
      "step": 2592,
      "time_per_iteration": 2.6548502445220947
    },
    {
      "auxiliary_loss_clip": 0.01169015,
      "auxiliary_loss_mlp": 0.01027712,
      "balance_loss_clip": 1.05872762,
      "balance_loss_mlp": 1.0191766,
      "epoch": 0.3117898154271629,
      "flos": 21177600134400.0,
      "grad_norm": 2.6660959686064594,
      "language_loss": 0.88255298,
      "learning_rate": 3.223436860636211e-06,
      "loss": 0.90452015,
      "num_input_tokens_seen": 55748685,
      "step": 2593,
      "time_per_iteration": 2.5286741256713867
    },
    {
      "auxiliary_loss_clip": 0.01199401,
      "auxiliary_loss_mlp": 0.01029395,
      "balance_loss_clip": 1.06186783,
      "balance_loss_mlp": 1.02081227,
      "epoch": 0.31191005831780194,
      "flos": 27272430840960.0,
      "grad_norm": 1.548227437678018,
      "language_loss": 0.73999971,
      "learning_rate": 3.2228205424124403e-06,
      "loss": 0.76228762,
      "num_input_tokens_seen": 55771840,
      "step": 2594,
      "time_per_iteration": 3.3637282848358154
    },
    {
      "auxiliary_loss_clip": 0.01153119,
      "auxiliary_loss_mlp": 0.01033927,
      "balance_loss_clip": 1.05186605,
      "balance_loss_mlp": 1.02512956,
      "epoch": 0.31203030120844105,
      "flos": 12963501043200.0,
      "grad_norm": 2.2025592442268214,
      "language_loss": 0.74561393,
      "learning_rate": 3.222204038687765e-06,
      "loss": 0.76748437,
      "num_input_tokens_seen": 55784975,
      "step": 2595,
      "time_per_iteration": 2.65503191947937
    },
    {
      "auxiliary_loss_clip": 0.0118014,
      "auxiliary_loss_mlp": 0.01024575,
      "balance_loss_clip": 1.05810118,
      "balance_loss_mlp": 1.01636124,
      "epoch": 0.31215054409908016,
      "flos": 27562014288000.0,
      "grad_norm": 1.6037843509077017,
      "language_loss": 0.88446277,
      "learning_rate": 3.221587349555709e-06,
      "loss": 0.90650988,
      "num_input_tokens_seen": 55805235,
      "step": 2596,
      "time_per_iteration": 2.5648632049560547
    },
    {
      "auxiliary_loss_clip": 0.01171159,
      "auxiliary_loss_mlp": 0.01035573,
      "balance_loss_clip": 1.05639791,
      "balance_loss_mlp": 1.02736545,
      "epoch": 0.3122707869897192,
      "flos": 21506326427520.0,
      "grad_norm": 1.5893122690443362,
      "language_loss": 0.69436681,
      "learning_rate": 3.2209704751098236e-06,
      "loss": 0.71643412,
      "num_input_tokens_seen": 55824265,
      "step": 2597,
      "time_per_iteration": 2.5491015911102295
    },
    {
      "auxiliary_loss_clip": 0.01171617,
      "auxiliary_loss_mlp": 0.01036219,
      "balance_loss_clip": 1.05764306,
      "balance_loss_mlp": 1.02753484,
      "epoch": 0.31239102988035833,
      "flos": 15187017674880.0,
      "grad_norm": 2.0083228241920237,
      "language_loss": 0.82886815,
      "learning_rate": 3.2203534154436875e-06,
      "loss": 0.85094643,
      "num_input_tokens_seen": 55838620,
      "step": 2598,
      "time_per_iteration": 2.4859020709991455
    },
    {
      "auxiliary_loss_clip": 0.01120067,
      "auxiliary_loss_mlp": 0.01034759,
      "balance_loss_clip": 1.04914236,
      "balance_loss_mlp": 1.02587771,
      "epoch": 0.31251127277099744,
      "flos": 22053712763520.0,
      "grad_norm": 1.929881430816264,
      "language_loss": 0.75656509,
      "learning_rate": 3.2197361706509084e-06,
      "loss": 0.77811331,
      "num_input_tokens_seen": 55859375,
      "step": 2599,
      "time_per_iteration": 2.651862859725952
    },
    {
      "auxiliary_loss_clip": 0.01198037,
      "auxiliary_loss_mlp": 0.0103652,
      "balance_loss_clip": 1.05908799,
      "balance_loss_mlp": 1.02704871,
      "epoch": 0.3126315156616365,
      "flos": 15193984913280.0,
      "grad_norm": 3.0542169096592104,
      "language_loss": 0.83419394,
      "learning_rate": 3.2191187408251228e-06,
      "loss": 0.85653949,
      "num_input_tokens_seen": 55876535,
      "step": 2600,
      "time_per_iteration": 2.428502321243286
    },
    {
      "auxiliary_loss_clip": 0.01187345,
      "auxiliary_loss_mlp": 0.01037465,
      "balance_loss_clip": 1.05699444,
      "balance_loss_mlp": 1.02795172,
      "epoch": 0.3127517585522756,
      "flos": 18145338831360.0,
      "grad_norm": 1.973567597144241,
      "language_loss": 0.79035676,
      "learning_rate": 3.218501126059993e-06,
      "loss": 0.8126049,
      "num_input_tokens_seen": 55891930,
      "step": 2601,
      "time_per_iteration": 2.4654862880706787
    },
    {
      "auxiliary_loss_clip": 0.011797,
      "auxiliary_loss_mlp": 0.01039187,
      "balance_loss_clip": 1.05189085,
      "balance_loss_mlp": 1.03037763,
      "epoch": 0.31287200144291466,
      "flos": 21908633731200.0,
      "grad_norm": 2.0196101193473868,
      "language_loss": 0.81580198,
      "learning_rate": 3.2178833264492116e-06,
      "loss": 0.83799088,
      "num_input_tokens_seen": 55910635,
      "step": 2602,
      "time_per_iteration": 2.5027902126312256
    },
    {
      "auxiliary_loss_clip": 0.01189436,
      "auxiliary_loss_mlp": 0.01033247,
      "balance_loss_clip": 1.05938458,
      "balance_loss_mlp": 1.02432442,
      "epoch": 0.31299224433355377,
      "flos": 29896997800320.0,
      "grad_norm": 1.8364446185421885,
      "language_loss": 0.76316798,
      "learning_rate": 3.217265342086498e-06,
      "loss": 0.78539479,
      "num_input_tokens_seen": 55931125,
      "step": 2603,
      "time_per_iteration": 2.580620765686035
    },
    {
      "auxiliary_loss_clip": 0.01157432,
      "auxiliary_loss_mlp": 0.00763305,
      "balance_loss_clip": 1.05653739,
      "balance_loss_mlp": 1.0003705,
      "epoch": 0.3131124872241929,
      "flos": 11655886331520.0,
      "grad_norm": 2.2594832086898475,
      "language_loss": 0.73423529,
      "learning_rate": 3.216647173065599e-06,
      "loss": 0.75344265,
      "num_input_tokens_seen": 55946590,
      "step": 2604,
      "time_per_iteration": 2.5388145446777344
    },
    {
      "auxiliary_loss_clip": 0.01171201,
      "auxiliary_loss_mlp": 0.01036194,
      "balance_loss_clip": 1.06009316,
      "balance_loss_mlp": 1.02734864,
      "epoch": 0.31323273011483194,
      "flos": 49848785470080.0,
      "grad_norm": 1.7390544968430852,
      "language_loss": 0.7379992,
      "learning_rate": 3.216028819480292e-06,
      "loss": 0.76007318,
      "num_input_tokens_seen": 55967930,
      "step": 2605,
      "time_per_iteration": 2.78756046295166
    },
    {
      "auxiliary_loss_clip": 0.01154863,
      "auxiliary_loss_mlp": 0.0103104,
      "balance_loss_clip": 1.05536354,
      "balance_loss_mlp": 1.02292204,
      "epoch": 0.31335297300547105,
      "flos": 22601278667520.0,
      "grad_norm": 2.823143289570943,
      "language_loss": 0.75486314,
      "learning_rate": 3.2154102814243793e-06,
      "loss": 0.77672213,
      "num_input_tokens_seen": 55987070,
      "step": 2606,
      "time_per_iteration": 2.5432121753692627
    },
    {
      "auxiliary_loss_clip": 0.01156079,
      "auxiliary_loss_mlp": 0.01033865,
      "balance_loss_clip": 1.05355072,
      "balance_loss_mlp": 1.02533531,
      "epoch": 0.31347321589611016,
      "flos": 34710858708480.0,
      "grad_norm": 1.8815780773325272,
      "language_loss": 0.66715771,
      "learning_rate": 3.2147915589916937e-06,
      "loss": 0.68905711,
      "num_input_tokens_seen": 56008630,
      "step": 2607,
      "time_per_iteration": 2.687453031539917
    },
    {
      "auxiliary_loss_clip": 0.01159803,
      "auxiliary_loss_mlp": 0.01040475,
      "balance_loss_clip": 1.05297351,
      "balance_loss_mlp": 1.03156447,
      "epoch": 0.3135934587867492,
      "flos": 19755789108480.0,
      "grad_norm": 1.7161850040529079,
      "language_loss": 0.8293196,
      "learning_rate": 3.2141726522760938e-06,
      "loss": 0.85132241,
      "num_input_tokens_seen": 56026690,
      "step": 2608,
      "time_per_iteration": 2.5403823852539062
    },
    {
      "auxiliary_loss_clip": 0.01070148,
      "auxiliary_loss_mlp": 0.01001901,
      "balance_loss_clip": 1.01810992,
      "balance_loss_mlp": 1.00038147,
      "epoch": 0.3137137016773883,
      "flos": 65815535583360.0,
      "grad_norm": 0.7112160536712682,
      "language_loss": 0.5271278,
      "learning_rate": 3.213553561371469e-06,
      "loss": 0.54784828,
      "num_input_tokens_seen": 56090425,
      "step": 2609,
      "time_per_iteration": 3.1994080543518066
    },
    {
      "auxiliary_loss_clip": 0.01139125,
      "auxiliary_loss_mlp": 0.01029714,
      "balance_loss_clip": 1.05596209,
      "balance_loss_mlp": 1.02164948,
      "epoch": 0.31383394456802743,
      "flos": 16252739222400.0,
      "grad_norm": 2.5657608057211103,
      "language_loss": 0.95867503,
      "learning_rate": 3.212934286371733e-06,
      "loss": 0.98036343,
      "num_input_tokens_seen": 56107135,
      "step": 2610,
      "time_per_iteration": 3.3194291591644287
    },
    {
      "auxiliary_loss_clip": 0.01181141,
      "auxiliary_loss_mlp": 0.01032643,
      "balance_loss_clip": 1.05972624,
      "balance_loss_mlp": 1.02336287,
      "epoch": 0.3139541874586665,
      "flos": 38795517613440.0,
      "grad_norm": 2.027615889655219,
      "language_loss": 0.83288139,
      "learning_rate": 3.2123148273708304e-06,
      "loss": 0.85501933,
      "num_input_tokens_seen": 56127325,
      "step": 2611,
      "time_per_iteration": 2.6463425159454346
    },
    {
      "auxiliary_loss_clip": 0.01197835,
      "auxiliary_loss_mlp": 0.01030489,
      "balance_loss_clip": 1.06146288,
      "balance_loss_mlp": 1.02233505,
      "epoch": 0.3140744303493056,
      "flos": 25046328430080.0,
      "grad_norm": 2.08020493929946,
      "language_loss": 0.77134627,
      "learning_rate": 3.211695184462733e-06,
      "loss": 0.79362953,
      "num_input_tokens_seen": 56148500,
      "step": 2612,
      "time_per_iteration": 2.509791374206543
    },
    {
      "auxiliary_loss_clip": 0.01049354,
      "auxiliary_loss_mlp": 0.01001951,
      "balance_loss_clip": 1.01739073,
      "balance_loss_mlp": 1.00031805,
      "epoch": 0.3141946732399447,
      "flos": 72504254782080.0,
      "grad_norm": 0.890321072590176,
      "language_loss": 0.60453182,
      "learning_rate": 3.2110753577414383e-06,
      "loss": 0.62504488,
      "num_input_tokens_seen": 56210080,
      "step": 2613,
      "time_per_iteration": 3.141970157623291
    },
    {
      "auxiliary_loss_clip": 0.01170268,
      "auxiliary_loss_mlp": 0.01038504,
      "balance_loss_clip": 1.05535674,
      "balance_loss_mlp": 1.02889609,
      "epoch": 0.31431491613058377,
      "flos": 19239788280960.0,
      "grad_norm": 2.2365450570677967,
      "language_loss": 0.79405403,
      "learning_rate": 3.2104553473009757e-06,
      "loss": 0.81614184,
      "num_input_tokens_seen": 56228200,
      "step": 2614,
      "time_per_iteration": 2.5333120822906494
    },
    {
      "auxiliary_loss_clip": 0.01132603,
      "auxiliary_loss_mlp": 0.01028743,
      "balance_loss_clip": 1.05070233,
      "balance_loss_mlp": 1.02012992,
      "epoch": 0.3144351590212229,
      "flos": 36210596290560.0,
      "grad_norm": 1.8656159937816497,
      "language_loss": 0.67947114,
      "learning_rate": 3.209835153235399e-06,
      "loss": 0.70108461,
      "num_input_tokens_seen": 56249755,
      "step": 2615,
      "time_per_iteration": 2.694971799850464
    },
    {
      "auxiliary_loss_clip": 0.01140705,
      "auxiliary_loss_mlp": 0.01040729,
      "balance_loss_clip": 1.05084348,
      "balance_loss_mlp": 1.03248,
      "epoch": 0.314555401911862,
      "flos": 18551740285440.0,
      "grad_norm": 1.7812893564095367,
      "language_loss": 0.67835236,
      "learning_rate": 3.2092147756387916e-06,
      "loss": 0.7001667,
      "num_input_tokens_seen": 56270080,
      "step": 2616,
      "time_per_iteration": 2.591644763946533
    },
    {
      "auxiliary_loss_clip": 0.01158283,
      "auxiliary_loss_mlp": 0.01036429,
      "balance_loss_clip": 1.05169749,
      "balance_loss_mlp": 1.02752399,
      "epoch": 0.31467564480250104,
      "flos": 16362877299840.0,
      "grad_norm": 1.7067710260144815,
      "language_loss": 0.83614421,
      "learning_rate": 3.208594214605264e-06,
      "loss": 0.85809135,
      "num_input_tokens_seen": 56288625,
      "step": 2617,
      "time_per_iteration": 4.0384862422943115
    },
    {
      "auxiliary_loss_clip": 0.01154726,
      "auxiliary_loss_mlp": 0.01029515,
      "balance_loss_clip": 1.05311823,
      "balance_loss_mlp": 1.02211857,
      "epoch": 0.31479588769314015,
      "flos": 21652375127040.0,
      "grad_norm": 2.9817987484213497,
      "language_loss": 0.77218896,
      "learning_rate": 3.2079734702289553e-06,
      "loss": 0.79403138,
      "num_input_tokens_seen": 56307520,
      "step": 2618,
      "time_per_iteration": 2.5486955642700195
    },
    {
      "auxiliary_loss_clip": 0.01068005,
      "auxiliary_loss_mlp": 0.00752077,
      "balance_loss_clip": 1.018471,
      "balance_loss_mlp": 1.00029206,
      "epoch": 0.3149161305837792,
      "flos": 66051072040320.0,
      "grad_norm": 0.8032105608109557,
      "language_loss": 0.60423696,
      "learning_rate": 3.207352542604031e-06,
      "loss": 0.62243783,
      "num_input_tokens_seen": 56369855,
      "step": 2619,
      "time_per_iteration": 3.1875483989715576
    },
    {
      "auxiliary_loss_clip": 0.01137888,
      "auxiliary_loss_mlp": 0.01036628,
      "balance_loss_clip": 1.0488627,
      "balance_loss_mlp": 1.02907586,
      "epoch": 0.3150363734744183,
      "flos": 28987201192320.0,
      "grad_norm": 1.688789880168598,
      "language_loss": 0.78459257,
      "learning_rate": 3.2067314318246864e-06,
      "loss": 0.80633777,
      "num_input_tokens_seen": 56390570,
      "step": 2620,
      "time_per_iteration": 3.4361112117767334
    },
    {
      "auxiliary_loss_clip": 0.01153217,
      "auxiliary_loss_mlp": 0.01036602,
      "balance_loss_clip": 1.05605221,
      "balance_loss_mlp": 1.02753055,
      "epoch": 0.31515661636505743,
      "flos": 27636600879360.0,
      "grad_norm": 1.6761251774850259,
      "language_loss": 0.77978599,
      "learning_rate": 3.206110137985143e-06,
      "loss": 0.8016842,
      "num_input_tokens_seen": 56410775,
      "step": 2621,
      "time_per_iteration": 2.6272075176239014
    },
    {
      "auxiliary_loss_clip": 0.01141856,
      "auxiliary_loss_mlp": 0.01042495,
      "balance_loss_clip": 1.05201685,
      "balance_loss_mlp": 1.03419828,
      "epoch": 0.3152768592556965,
      "flos": 24605632465920.0,
      "grad_norm": 1.86622348775602,
      "language_loss": 0.92265069,
      "learning_rate": 3.2054886611796505e-06,
      "loss": 0.94449425,
      "num_input_tokens_seen": 56429770,
      "step": 2622,
      "time_per_iteration": 2.614384174346924
    },
    {
      "auxiliary_loss_clip": 0.0108116,
      "auxiliary_loss_mlp": 0.01003168,
      "balance_loss_clip": 1.01829171,
      "balance_loss_mlp": 1.00165367,
      "epoch": 0.3153971021463356,
      "flos": 68476908026880.0,
      "grad_norm": 0.8849685160866501,
      "language_loss": 0.63569796,
      "learning_rate": 3.204867001502487e-06,
      "loss": 0.65654123,
      "num_input_tokens_seen": 56488425,
      "step": 2623,
      "time_per_iteration": 3.0761187076568604
    },
    {
      "auxiliary_loss_clip": 0.01198412,
      "auxiliary_loss_mlp": 0.01036728,
      "balance_loss_clip": 1.06179595,
      "balance_loss_mlp": 1.02793646,
      "epoch": 0.3155173450369747,
      "flos": 25593714766080.0,
      "grad_norm": 1.7163933315117532,
      "language_loss": 0.80837202,
      "learning_rate": 3.2042451590479567e-06,
      "loss": 0.8307234,
      "num_input_tokens_seen": 56508940,
      "step": 2624,
      "time_per_iteration": 2.5493316650390625
    },
    {
      "auxiliary_loss_clip": 0.01191862,
      "auxiliary_loss_mlp": 0.01031722,
      "balance_loss_clip": 1.05890584,
      "balance_loss_mlp": 1.02378881,
      "epoch": 0.31563758792761376,
      "flos": 24309333175680.0,
      "grad_norm": 1.675538353584152,
      "language_loss": 0.87117887,
      "learning_rate": 3.203623133910394e-06,
      "loss": 0.89341462,
      "num_input_tokens_seen": 56527245,
      "step": 2625,
      "time_per_iteration": 2.4917385578155518
    },
    {
      "auxiliary_loss_clip": 0.01125117,
      "auxiliary_loss_mlp": 0.01033732,
      "balance_loss_clip": 1.04885519,
      "balance_loss_mlp": 1.02572131,
      "epoch": 0.31575783081825287,
      "flos": 31903865550720.0,
      "grad_norm": 3.464416011822075,
      "language_loss": 0.77414489,
      "learning_rate": 3.203000926184158e-06,
      "loss": 0.79573333,
      "num_input_tokens_seen": 56546170,
      "step": 2626,
      "time_per_iteration": 2.7675654888153076
    },
    {
      "auxiliary_loss_clip": 0.01193828,
      "auxiliary_loss_mlp": 0.01032746,
      "balance_loss_clip": 1.05849981,
      "balance_loss_mlp": 1.02439523,
      "epoch": 0.315878073708892,
      "flos": 30810960385920.0,
      "grad_norm": 1.64914061224499,
      "language_loss": 0.77796954,
      "learning_rate": 3.202378535963639e-06,
      "loss": 0.80023527,
      "num_input_tokens_seen": 56567085,
      "step": 2627,
      "time_per_iteration": 2.5715365409851074
    },
    {
      "auxiliary_loss_clip": 0.01155921,
      "auxiliary_loss_mlp": 0.00762502,
      "balance_loss_clip": 1.05190134,
      "balance_loss_mlp": 1.00034094,
      "epoch": 0.31599831659953104,
      "flos": 22200264253440.0,
      "grad_norm": 1.59793280589802,
      "language_loss": 0.84087592,
      "learning_rate": 3.2017559633432516e-06,
      "loss": 0.86006021,
      "num_input_tokens_seen": 56586715,
      "step": 2628,
      "time_per_iteration": 2.6015841960906982
    },
    {
      "auxiliary_loss_clip": 0.01173523,
      "auxiliary_loss_mlp": 0.01032799,
      "balance_loss_clip": 1.05579281,
      "balance_loss_mlp": 1.02412665,
      "epoch": 0.31611855949017015,
      "flos": 25593463370880.0,
      "grad_norm": 2.1497887832158384,
      "language_loss": 0.65993404,
      "learning_rate": 3.2011332084174398e-06,
      "loss": 0.6819973,
      "num_input_tokens_seen": 56607585,
      "step": 2629,
      "time_per_iteration": 2.6395835876464844
    },
    {
      "auxiliary_loss_clip": 0.01177637,
      "auxiliary_loss_mlp": 0.01031795,
      "balance_loss_clip": 1.05661488,
      "balance_loss_mlp": 1.02271152,
      "epoch": 0.31623880238080926,
      "flos": 20594087694720.0,
      "grad_norm": 1.914409029705995,
      "language_loss": 0.89154434,
      "learning_rate": 3.2005102712806756e-06,
      "loss": 0.91363865,
      "num_input_tokens_seen": 56626415,
      "step": 2630,
      "time_per_iteration": 2.5515828132629395
    },
    {
      "auxiliary_loss_clip": 0.01182769,
      "auxiliary_loss_mlp": 0.01036714,
      "balance_loss_clip": 1.05701363,
      "balance_loss_mlp": 1.02795768,
      "epoch": 0.3163590452714483,
      "flos": 12784917600000.0,
      "grad_norm": 2.5350831691850937,
      "language_loss": 0.73007977,
      "learning_rate": 3.1998871520274575e-06,
      "loss": 0.75227457,
      "num_input_tokens_seen": 56641750,
      "step": 2631,
      "time_per_iteration": 2.525724172592163
    },
    {
      "auxiliary_loss_clip": 0.01166509,
      "auxiliary_loss_mlp": 0.01033643,
      "balance_loss_clip": 1.05273175,
      "balance_loss_mlp": 1.02474999,
      "epoch": 0.3164792881620874,
      "flos": 23041292273280.0,
      "grad_norm": 1.9071740148140937,
      "language_loss": 0.85228741,
      "learning_rate": 3.199263850752312e-06,
      "loss": 0.87428904,
      "num_input_tokens_seen": 56662585,
      "step": 2632,
      "time_per_iteration": 2.612267017364502
    },
    {
      "auxiliary_loss_clip": 0.01182211,
      "auxiliary_loss_mlp": 0.01031869,
      "balance_loss_clip": 1.05890751,
      "balance_loss_mlp": 1.02358365,
      "epoch": 0.31659953105272653,
      "flos": 18296271780480.0,
      "grad_norm": 2.3301343111931327,
      "language_loss": 0.85498965,
      "learning_rate": 3.198640367549795e-06,
      "loss": 0.87713045,
      "num_input_tokens_seen": 56681480,
      "step": 2633,
      "time_per_iteration": 2.5099260807037354
    },
    {
      "auxiliary_loss_clip": 0.01180273,
      "auxiliary_loss_mlp": 0.0076222,
      "balance_loss_clip": 1.05535388,
      "balance_loss_mlp": 1.00034976,
      "epoch": 0.3167197739433656,
      "flos": 25703421880320.0,
      "grad_norm": 1.6577391102634011,
      "language_loss": 0.86008906,
      "learning_rate": 3.198016702514487e-06,
      "loss": 0.87951398,
      "num_input_tokens_seen": 56701760,
      "step": 2634,
      "time_per_iteration": 2.6008989810943604
    },
    {
      "auxiliary_loss_clip": 0.01191448,
      "auxiliary_loss_mlp": 0.0103135,
      "balance_loss_clip": 1.05753326,
      "balance_loss_mlp": 1.02389669,
      "epoch": 0.3168400168340047,
      "flos": 23546016230400.0,
      "grad_norm": 1.6940686550640216,
      "language_loss": 0.84465146,
      "learning_rate": 3.1973928557409972e-06,
      "loss": 0.86687946,
      "num_input_tokens_seen": 56719800,
      "step": 2635,
      "time_per_iteration": 2.502098321914673
    },
    {
      "auxiliary_loss_clip": 0.01190271,
      "auxiliary_loss_mlp": 0.01036714,
      "balance_loss_clip": 1.05749774,
      "balance_loss_mlp": 1.02861929,
      "epoch": 0.31696025972464376,
      "flos": 28366449327360.0,
      "grad_norm": 2.0068180523141437,
      "language_loss": 0.71295696,
      "learning_rate": 3.1967688273239636e-06,
      "loss": 0.73522681,
      "num_input_tokens_seen": 56739605,
      "step": 2636,
      "time_per_iteration": 3.3501522541046143
    },
    {
      "auxiliary_loss_clip": 0.01150769,
      "auxiliary_loss_mlp": 0.01024294,
      "balance_loss_clip": 1.05323672,
      "balance_loss_mlp": 1.01638103,
      "epoch": 0.31708050261528287,
      "flos": 16399111144320.0,
      "grad_norm": 2.638067879483366,
      "language_loss": 0.82267773,
      "learning_rate": 3.1961446173580503e-06,
      "loss": 0.84442842,
      "num_input_tokens_seen": 56756545,
      "step": 2637,
      "time_per_iteration": 2.5423877239227295
    },
    {
      "auxiliary_loss_clip": 0.01164621,
      "auxiliary_loss_mlp": 0.01034778,
      "balance_loss_clip": 1.05633152,
      "balance_loss_mlp": 1.02626014,
      "epoch": 0.317200745505922,
      "flos": 26212347728640.0,
      "grad_norm": 1.668951798371586,
      "language_loss": 0.77542055,
      "learning_rate": 3.1955202259379502e-06,
      "loss": 0.79741454,
      "num_input_tokens_seen": 56778275,
      "step": 2638,
      "time_per_iteration": 2.623281240463257
    },
    {
      "auxiliary_loss_clip": 0.01178847,
      "auxiliary_loss_mlp": 0.0103325,
      "balance_loss_clip": 1.05596209,
      "balance_loss_mlp": 1.02428567,
      "epoch": 0.31732098839656103,
      "flos": 31350876693120.0,
      "grad_norm": 1.8602806066028563,
      "language_loss": 0.8308391,
      "learning_rate": 3.194895653158381e-06,
      "loss": 0.85296011,
      "num_input_tokens_seen": 56797215,
      "step": 2639,
      "time_per_iteration": 2.5842666625976562
    },
    {
      "auxiliary_loss_clip": 0.010813,
      "auxiliary_loss_mlp": 0.01003439,
      "balance_loss_clip": 1.0184083,
      "balance_loss_mlp": 1.00191927,
      "epoch": 0.31744123128720014,
      "flos": 58989024835200.0,
      "grad_norm": 0.7765355675301934,
      "language_loss": 0.55556881,
      "learning_rate": 3.194270899114093e-06,
      "loss": 0.57641613,
      "num_input_tokens_seen": 56863010,
      "step": 2640,
      "time_per_iteration": 3.195396900177002
    },
    {
      "auxiliary_loss_clip": 0.01187455,
      "auxiliary_loss_mlp": 0.01042995,
      "balance_loss_clip": 1.05870187,
      "balance_loss_mlp": 1.03431654,
      "epoch": 0.31756147417783925,
      "flos": 17417573372160.0,
      "grad_norm": 1.8181757950168371,
      "language_loss": 0.81895441,
      "learning_rate": 3.193645963899858e-06,
      "loss": 0.84125888,
      "num_input_tokens_seen": 56880625,
      "step": 2641,
      "time_per_iteration": 2.528512477874756
    },
    {
      "auxiliary_loss_clip": 0.01159542,
      "auxiliary_loss_mlp": 0.01031409,
      "balance_loss_clip": 1.05323005,
      "balance_loss_mlp": 1.02257013,
      "epoch": 0.3176817170684783,
      "flos": 25481673267840.0,
      "grad_norm": 5.24323609099548,
      "language_loss": 0.84125865,
      "learning_rate": 3.193020847610479e-06,
      "loss": 0.86316812,
      "num_input_tokens_seen": 56900945,
      "step": 2642,
      "time_per_iteration": 2.562692403793335
    },
    {
      "auxiliary_loss_clip": 0.01157786,
      "auxiliary_loss_mlp": 0.01030856,
      "balance_loss_clip": 1.05443454,
      "balance_loss_mlp": 1.02140319,
      "epoch": 0.3178019599591174,
      "flos": 24972603765120.0,
      "grad_norm": 2.080888321157409,
      "language_loss": 0.71308017,
      "learning_rate": 3.192395550340787e-06,
      "loss": 0.73496664,
      "num_input_tokens_seen": 56918895,
      "step": 2643,
      "time_per_iteration": 3.496281862258911
    },
    {
      "auxiliary_loss_clip": 0.01178285,
      "auxiliary_loss_mlp": 0.01033699,
      "balance_loss_clip": 1.05687571,
      "balance_loss_mlp": 1.02549767,
      "epoch": 0.31792220284975653,
      "flos": 12422220019200.0,
      "grad_norm": 2.018649976935453,
      "language_loss": 0.76772875,
      "learning_rate": 3.191770072185638e-06,
      "loss": 0.78984857,
      "num_input_tokens_seen": 56935890,
      "step": 2644,
      "time_per_iteration": 2.4758667945861816
    },
    {
      "auxiliary_loss_clip": 0.01178847,
      "auxiliary_loss_mlp": 0.01032587,
      "balance_loss_clip": 1.05875802,
      "balance_loss_mlp": 1.02428722,
      "epoch": 0.3180424457403956,
      "flos": 15485759089920.0,
      "grad_norm": 2.5372633830720703,
      "language_loss": 0.73069,
      "learning_rate": 3.191144413239916e-06,
      "loss": 0.7528044,
      "num_input_tokens_seen": 56952460,
      "step": 2645,
      "time_per_iteration": 2.4814114570617676
    },
    {
      "auxiliary_loss_clip": 0.01164879,
      "auxiliary_loss_mlp": 0.01036981,
      "balance_loss_clip": 1.05476785,
      "balance_loss_mlp": 1.02783799,
      "epoch": 0.3181626886310347,
      "flos": 26174964648960.0,
      "grad_norm": 1.9477532923758072,
      "language_loss": 0.8827486,
      "learning_rate": 3.190518573598534e-06,
      "loss": 0.90476716,
      "num_input_tokens_seen": 56969065,
      "step": 2646,
      "time_per_iteration": 3.4279568195343018
    },
    {
      "auxiliary_loss_clip": 0.01158273,
      "auxiliary_loss_mlp": 0.01034064,
      "balance_loss_clip": 1.05383265,
      "balance_loss_mlp": 1.02538562,
      "epoch": 0.3182829315216738,
      "flos": 25483109811840.0,
      "grad_norm": 1.5171096439817755,
      "language_loss": 0.7737484,
      "learning_rate": 3.1898925533564308e-06,
      "loss": 0.79567182,
      "num_input_tokens_seen": 56990535,
      "step": 2647,
      "time_per_iteration": 2.6136019229888916
    },
    {
      "auxiliary_loss_clip": 0.01137716,
      "auxiliary_loss_mlp": 0.01029777,
      "balance_loss_clip": 1.04845619,
      "balance_loss_mlp": 1.02066374,
      "epoch": 0.31840317441231286,
      "flos": 18113701927680.0,
      "grad_norm": 1.8613188658150552,
      "language_loss": 0.63824475,
      "learning_rate": 3.1892663526085733e-06,
      "loss": 0.65991968,
      "num_input_tokens_seen": 57008910,
      "step": 2648,
      "time_per_iteration": 2.559540033340454
    },
    {
      "auxiliary_loss_clip": 0.01081809,
      "auxiliary_loss_mlp": 0.01003627,
      "balance_loss_clip": 1.01885808,
      "balance_loss_mlp": 1.00200617,
      "epoch": 0.31852341730295197,
      "flos": 64741948957440.0,
      "grad_norm": 0.769466184298523,
      "language_loss": 0.56930357,
      "learning_rate": 3.188639971449956e-06,
      "loss": 0.59015793,
      "num_input_tokens_seen": 57074960,
      "step": 2649,
      "time_per_iteration": 3.057826519012451
    },
    {
      "auxiliary_loss_clip": 0.0119535,
      "auxiliary_loss_mlp": 0.01031263,
      "balance_loss_clip": 1.05960917,
      "balance_loss_mlp": 1.02302003,
      "epoch": 0.318643660193591,
      "flos": 20668135582080.0,
      "grad_norm": 1.7832300950106799,
      "language_loss": 0.72373605,
      "learning_rate": 3.1880134099756e-06,
      "loss": 0.7460022,
      "num_input_tokens_seen": 57094595,
      "step": 2650,
      "time_per_iteration": 2.473712921142578
    },
    {
      "auxiliary_loss_clip": 0.01174526,
      "auxiliary_loss_mlp": 0.01031098,
      "balance_loss_clip": 1.05319715,
      "balance_loss_mlp": 1.02305746,
      "epoch": 0.31876390308423014,
      "flos": 26943345411840.0,
      "grad_norm": 1.9483728768602098,
      "language_loss": 0.69597852,
      "learning_rate": 3.1873866682805535e-06,
      "loss": 0.7180348,
      "num_input_tokens_seen": 57115290,
      "step": 2651,
      "time_per_iteration": 2.541693925857544
    },
    {
      "auxiliary_loss_clip": 0.01166766,
      "auxiliary_loss_mlp": 0.01030791,
      "balance_loss_clip": 1.05519795,
      "balance_loss_mlp": 1.02270222,
      "epoch": 0.31888414597486925,
      "flos": 18041916597120.0,
      "grad_norm": 1.8911633812860198,
      "language_loss": 0.88600028,
      "learning_rate": 3.186759746459894e-06,
      "loss": 0.90797579,
      "num_input_tokens_seen": 57134400,
      "step": 2652,
      "time_per_iteration": 2.5064198970794678
    },
    {
      "auxiliary_loss_clip": 0.01163276,
      "auxiliary_loss_mlp": 0.01032238,
      "balance_loss_clip": 1.05516267,
      "balance_loss_mlp": 1.02396452,
      "epoch": 0.3190043888655083,
      "flos": 25149319701120.0,
      "grad_norm": 1.8117284301261898,
      "language_loss": 0.79823989,
      "learning_rate": 3.1861326446087246e-06,
      "loss": 0.82019496,
      "num_input_tokens_seen": 57153140,
      "step": 2653,
      "time_per_iteration": 2.5550949573516846
    },
    {
      "auxiliary_loss_clip": 0.01182722,
      "auxiliary_loss_mlp": 0.01029753,
      "balance_loss_clip": 1.0578866,
      "balance_loss_mlp": 1.02137852,
      "epoch": 0.3191246317561474,
      "flos": 22053892331520.0,
      "grad_norm": 1.8996230243304466,
      "language_loss": 0.71730888,
      "learning_rate": 3.1855053628221763e-06,
      "loss": 0.73943359,
      "num_input_tokens_seen": 57172395,
      "step": 2654,
      "time_per_iteration": 2.5159761905670166
    },
    {
      "auxiliary_loss_clip": 0.0114049,
      "auxiliary_loss_mlp": 0.01035162,
      "balance_loss_clip": 1.04897881,
      "balance_loss_mlp": 1.02589965,
      "epoch": 0.3192448746467865,
      "flos": 14901815687040.0,
      "grad_norm": 2.379114194613131,
      "language_loss": 0.89773118,
      "learning_rate": 3.184877901195407e-06,
      "loss": 0.91948771,
      "num_input_tokens_seen": 57189090,
      "step": 2655,
      "time_per_iteration": 2.5219740867614746
    },
    {
      "auxiliary_loss_clip": 0.01065623,
      "auxiliary_loss_mlp": 0.01003226,
      "balance_loss_clip": 1.02396584,
      "balance_loss_mlp": 1.00133097,
      "epoch": 0.3193651175374256,
      "flos": 67234832657280.0,
      "grad_norm": 0.7948835507769505,
      "language_loss": 0.62837976,
      "learning_rate": 3.184250259823602e-06,
      "loss": 0.64906824,
      "num_input_tokens_seen": 57251620,
      "step": 2656,
      "time_per_iteration": 3.2015395164489746
    },
    {
      "auxiliary_loss_clip": 0.01153462,
      "auxiliary_loss_mlp": 0.01031338,
      "balance_loss_clip": 1.05340707,
      "balance_loss_mlp": 1.02280271,
      "epoch": 0.3194853604280647,
      "flos": 12233077977600.0,
      "grad_norm": 1.9561590052542814,
      "language_loss": 0.81286621,
      "learning_rate": 3.183622438801974e-06,
      "loss": 0.83471429,
      "num_input_tokens_seen": 57266910,
      "step": 2657,
      "time_per_iteration": 2.5572798252105713
    },
    {
      "auxiliary_loss_clip": 0.01197131,
      "auxiliary_loss_mlp": 0.01033712,
      "balance_loss_clip": 1.06114006,
      "balance_loss_mlp": 1.02530766,
      "epoch": 0.3196056033187038,
      "flos": 14939917038720.0,
      "grad_norm": 1.8431228324286453,
      "language_loss": 0.75488532,
      "learning_rate": 3.1829944382257637e-06,
      "loss": 0.77719378,
      "num_input_tokens_seen": 57285040,
      "step": 2658,
      "time_per_iteration": 2.4816553592681885
    },
    {
      "auxiliary_loss_clip": 0.01179223,
      "auxiliary_loss_mlp": 0.01029998,
      "balance_loss_clip": 1.0590024,
      "balance_loss_mlp": 1.02207112,
      "epoch": 0.31972584620934286,
      "flos": 23768878164480.0,
      "grad_norm": 2.1946276165519625,
      "language_loss": 0.81406397,
      "learning_rate": 3.1823662581902373e-06,
      "loss": 0.83615619,
      "num_input_tokens_seen": 57302725,
      "step": 2659,
      "time_per_iteration": 2.540379762649536
    },
    {
      "auxiliary_loss_clip": 0.01135013,
      "auxiliary_loss_mlp": 0.01035547,
      "balance_loss_clip": 1.04629803,
      "balance_loss_mlp": 1.02719676,
      "epoch": 0.31984608909998197,
      "flos": 21251540280960.0,
      "grad_norm": 2.005244393547803,
      "language_loss": 0.74909663,
      "learning_rate": 3.1817378987906896e-06,
      "loss": 0.7708022,
      "num_input_tokens_seen": 57322230,
      "step": 2660,
      "time_per_iteration": 2.582280397415161
    },
    {
      "auxiliary_loss_clip": 0.01133156,
      "auxiliary_loss_mlp": 0.01040223,
      "balance_loss_clip": 1.05480337,
      "balance_loss_mlp": 1.03211069,
      "epoch": 0.3199663319906211,
      "flos": 18296235866880.0,
      "grad_norm": 1.939811434699013,
      "language_loss": 0.80028826,
      "learning_rate": 3.181109360122442e-06,
      "loss": 0.82202208,
      "num_input_tokens_seen": 57339820,
      "step": 2661,
      "time_per_iteration": 2.578160524368286
    },
    {
      "auxiliary_loss_clip": 0.01146673,
      "auxiliary_loss_mlp": 0.01033084,
      "balance_loss_clip": 1.05174303,
      "balance_loss_mlp": 1.02455509,
      "epoch": 0.32008657488126013,
      "flos": 18733627779840.0,
      "grad_norm": 2.8415841755471622,
      "language_loss": 0.78306174,
      "learning_rate": 3.1804806422808445e-06,
      "loss": 0.80485934,
      "num_input_tokens_seen": 57356955,
      "step": 2662,
      "time_per_iteration": 2.5615053176879883
    },
    {
      "auxiliary_loss_clip": 0.01155654,
      "auxiliary_loss_mlp": 0.01034959,
      "balance_loss_clip": 1.05432642,
      "balance_loss_mlp": 1.02626288,
      "epoch": 0.32020681777189924,
      "flos": 20595344670720.0,
      "grad_norm": 1.6790661683635506,
      "language_loss": 0.72832751,
      "learning_rate": 3.1798517453612714e-06,
      "loss": 0.75023359,
      "num_input_tokens_seen": 57376760,
      "step": 2663,
      "time_per_iteration": 3.279682159423828
    },
    {
      "auxiliary_loss_clip": 0.01176888,
      "auxiliary_loss_mlp": 0.01036639,
      "balance_loss_clip": 1.05929756,
      "balance_loss_mlp": 1.02835989,
      "epoch": 0.32032706066253835,
      "flos": 35261692750080.0,
      "grad_norm": 2.1262285027409646,
      "language_loss": 0.74783993,
      "learning_rate": 3.1792226694591265e-06,
      "loss": 0.76997524,
      "num_input_tokens_seen": 57398145,
      "step": 2664,
      "time_per_iteration": 2.6269819736480713
    },
    {
      "auxiliary_loss_clip": 0.01149373,
      "auxiliary_loss_mlp": 0.01039119,
      "balance_loss_clip": 1.05503643,
      "balance_loss_mlp": 1.03132248,
      "epoch": 0.3204473035531774,
      "flos": 15304230731520.0,
      "grad_norm": 1.807630391158078,
      "language_loss": 0.80303264,
      "learning_rate": 3.178593414669841e-06,
      "loss": 0.82491755,
      "num_input_tokens_seen": 57416730,
      "step": 2665,
      "time_per_iteration": 2.590731143951416
    },
    {
      "auxiliary_loss_clip": 0.01180263,
      "auxiliary_loss_mlp": 0.01036542,
      "balance_loss_clip": 1.05708146,
      "balance_loss_mlp": 1.02688622,
      "epoch": 0.3205675464438165,
      "flos": 24462564595200.0,
      "grad_norm": 2.315232937448908,
      "language_loss": 0.7082088,
      "learning_rate": 3.1779639810888707e-06,
      "loss": 0.73037684,
      "num_input_tokens_seen": 57436325,
      "step": 2666,
      "time_per_iteration": 2.523946762084961
    },
    {
      "auxiliary_loss_clip": 0.01176751,
      "auxiliary_loss_mlp": 0.01029251,
      "balance_loss_clip": 1.05775309,
      "balance_loss_mlp": 1.02138281,
      "epoch": 0.3206877893344556,
      "flos": 22456235548800.0,
      "grad_norm": 1.8092318757780659,
      "language_loss": 0.75824231,
      "learning_rate": 3.1773343688117013e-06,
      "loss": 0.78030235,
      "num_input_tokens_seen": 57457235,
      "step": 2667,
      "time_per_iteration": 2.5497312545776367
    },
    {
      "auxiliary_loss_clip": 0.01166177,
      "auxiliary_loss_mlp": 0.00761484,
      "balance_loss_clip": 1.05307436,
      "balance_loss_mlp": 1.00031805,
      "epoch": 0.3208080322250947,
      "flos": 20412236113920.0,
      "grad_norm": 2.274709374364616,
      "language_loss": 0.84455216,
      "learning_rate": 3.1767045779338445e-06,
      "loss": 0.86382878,
      "num_input_tokens_seen": 57474895,
      "step": 2668,
      "time_per_iteration": 2.5442593097686768
    },
    {
      "auxiliary_loss_clip": 0.01176835,
      "auxiliary_loss_mlp": 0.01033701,
      "balance_loss_clip": 1.05360711,
      "balance_loss_mlp": 1.02522564,
      "epoch": 0.3209282751157338,
      "flos": 21762118154880.0,
      "grad_norm": 2.3339706861538345,
      "language_loss": 0.91385859,
      "learning_rate": 3.176074608550839e-06,
      "loss": 0.93596399,
      "num_input_tokens_seen": 57490715,
      "step": 2669,
      "time_per_iteration": 4.0234880447387695
    },
    {
      "auxiliary_loss_clip": 0.01125642,
      "auxiliary_loss_mlp": 0.01038041,
      "balance_loss_clip": 1.05138493,
      "balance_loss_mlp": 1.02958322,
      "epoch": 0.32104851800637285,
      "flos": 22055041566720.0,
      "grad_norm": 2.301230044446547,
      "language_loss": 0.82308203,
      "learning_rate": 3.17544446075825e-06,
      "loss": 0.84471881,
      "num_input_tokens_seen": 57509880,
      "step": 2670,
      "time_per_iteration": 2.6307272911071777
    },
    {
      "auxiliary_loss_clip": 0.01172142,
      "auxiliary_loss_mlp": 0.01032435,
      "balance_loss_clip": 1.0573349,
      "balance_loss_mlp": 1.02517533,
      "epoch": 0.32116876089701196,
      "flos": 37012301896320.0,
      "grad_norm": 1.6008794202942818,
      "language_loss": 0.71016449,
      "learning_rate": 3.174814134651671e-06,
      "loss": 0.73221034,
      "num_input_tokens_seen": 57532430,
      "step": 2671,
      "time_per_iteration": 2.6674296855926514
    },
    {
      "auxiliary_loss_clip": 0.01189318,
      "auxiliary_loss_mlp": 0.01034882,
      "balance_loss_clip": 1.05798197,
      "balance_loss_mlp": 1.02722287,
      "epoch": 0.3212890037876511,
      "flos": 21979233912960.0,
      "grad_norm": 1.73480445853841,
      "language_loss": 0.8050698,
      "learning_rate": 3.1741836303267215e-06,
      "loss": 0.82731175,
      "num_input_tokens_seen": 57551965,
      "step": 2672,
      "time_per_iteration": 3.3182523250579834
    },
    {
      "auxiliary_loss_clip": 0.01189596,
      "auxiliary_loss_mlp": 0.01031391,
      "balance_loss_clip": 1.05821872,
      "balance_loss_mlp": 1.02254534,
      "epoch": 0.32140924667829013,
      "flos": 10342345875840.0,
      "grad_norm": 1.9143589968022052,
      "language_loss": 0.75182116,
      "learning_rate": 3.1735529478790496e-06,
      "loss": 0.77403104,
      "num_input_tokens_seen": 57569955,
      "step": 2673,
      "time_per_iteration": 2.4377593994140625
    },
    {
      "auxiliary_loss_clip": 0.0117883,
      "auxiliary_loss_mlp": 0.01031195,
      "balance_loss_clip": 1.05787694,
      "balance_loss_mlp": 1.02237916,
      "epoch": 0.32152948956892924,
      "flos": 50798910072960.0,
      "grad_norm": 1.8540781847067014,
      "language_loss": 0.79164964,
      "learning_rate": 3.172922087404328e-06,
      "loss": 0.81374991,
      "num_input_tokens_seen": 57592215,
      "step": 2674,
      "time_per_iteration": 2.7556631565093994
    },
    {
      "auxiliary_loss_clip": 0.01079644,
      "auxiliary_loss_mlp": 0.01006971,
      "balance_loss_clip": 1.01718402,
      "balance_loss_mlp": 1.00530195,
      "epoch": 0.32164973245956835,
      "flos": 63863250549120.0,
      "grad_norm": 0.774204213820679,
      "language_loss": 0.55248016,
      "learning_rate": 3.1722910489982586e-06,
      "loss": 0.57334632,
      "num_input_tokens_seen": 57652575,
      "step": 2675,
      "time_per_iteration": 3.1508233547210693
    },
    {
      "auxiliary_loss_clip": 0.01157738,
      "auxiliary_loss_mlp": 0.010371,
      "balance_loss_clip": 1.05244756,
      "balance_loss_mlp": 1.02843308,
      "epoch": 0.3217699753502074,
      "flos": 23513948363520.0,
      "grad_norm": 1.5461703968423186,
      "language_loss": 0.80198348,
      "learning_rate": 3.1716598327565694e-06,
      "loss": 0.82393181,
      "num_input_tokens_seen": 57672215,
      "step": 2676,
      "time_per_iteration": 2.550035238265991
    },
    {
      "auxiliary_loss_clip": 0.01192439,
      "auxiliary_loss_mlp": 0.01027035,
      "balance_loss_clip": 1.05886412,
      "balance_loss_mlp": 1.01994836,
      "epoch": 0.3218902182408465,
      "flos": 19062533640960.0,
      "grad_norm": 1.4767632616792772,
      "language_loss": 0.84132409,
      "learning_rate": 3.171028438775015e-06,
      "loss": 0.86351877,
      "num_input_tokens_seen": 57691410,
      "step": 2677,
      "time_per_iteration": 2.463961601257324
    },
    {
      "auxiliary_loss_clip": 0.01193673,
      "auxiliary_loss_mlp": 0.01029074,
      "balance_loss_clip": 1.05921602,
      "balance_loss_mlp": 1.02109885,
      "epoch": 0.3220104611314856,
      "flos": 20375571306240.0,
      "grad_norm": 2.16058998437503,
      "language_loss": 0.84063214,
      "learning_rate": 3.170396867149377e-06,
      "loss": 0.86285961,
      "num_input_tokens_seen": 57709415,
      "step": 2678,
      "time_per_iteration": 2.4640145301818848
    },
    {
      "auxiliary_loss_clip": 0.01127343,
      "auxiliary_loss_mlp": 0.01037295,
      "balance_loss_clip": 1.05160975,
      "balance_loss_mlp": 1.02830625,
      "epoch": 0.3221307040221247,
      "flos": 20117014231680.0,
      "grad_norm": 1.6902532311491472,
      "language_loss": 0.86485612,
      "learning_rate": 3.1697651179754653e-06,
      "loss": 0.8865025,
      "num_input_tokens_seen": 57728075,
      "step": 2679,
      "time_per_iteration": 2.5881197452545166
    },
    {
      "auxiliary_loss_clip": 0.01151689,
      "auxiliary_loss_mlp": 0.01032881,
      "balance_loss_clip": 1.05696213,
      "balance_loss_mlp": 1.02455473,
      "epoch": 0.3222509469127638,
      "flos": 23987789602560.0,
      "grad_norm": 1.770196753166316,
      "language_loss": 0.73259848,
      "learning_rate": 3.1691331913491153e-06,
      "loss": 0.75444412,
      "num_input_tokens_seen": 57750645,
      "step": 2680,
      "time_per_iteration": 2.6227028369903564
    },
    {
      "auxiliary_loss_clip": 0.01189915,
      "auxiliary_loss_mlp": 0.01032205,
      "balance_loss_clip": 1.05653465,
      "balance_loss_mlp": 1.02427745,
      "epoch": 0.32237118980340285,
      "flos": 17675735397120.0,
      "grad_norm": 2.0405114950899867,
      "language_loss": 0.84846628,
      "learning_rate": 3.1685010873661898e-06,
      "loss": 0.87068748,
      "num_input_tokens_seen": 57769820,
      "step": 2681,
      "time_per_iteration": 2.4417991638183594
    },
    {
      "auxiliary_loss_clip": 0.01174669,
      "auxiliary_loss_mlp": 0.01032684,
      "balance_loss_clip": 1.05594146,
      "balance_loss_mlp": 1.02377272,
      "epoch": 0.32249143269404196,
      "flos": 23147982645120.0,
      "grad_norm": 2.0358809136965608,
      "language_loss": 0.79536301,
      "learning_rate": 3.167868806122578e-06,
      "loss": 0.81743658,
      "num_input_tokens_seen": 57788870,
      "step": 2682,
      "time_per_iteration": 2.509596347808838
    },
    {
      "auxiliary_loss_clip": 0.01168401,
      "auxiliary_loss_mlp": 0.01035691,
      "balance_loss_clip": 1.05699515,
      "balance_loss_mlp": 1.027156,
      "epoch": 0.32261167558468107,
      "flos": 24422308427520.0,
      "grad_norm": 2.06066875729796,
      "language_loss": 0.66108555,
      "learning_rate": 3.1672363477141968e-06,
      "loss": 0.68312651,
      "num_input_tokens_seen": 57808165,
      "step": 2683,
      "time_per_iteration": 2.59277606010437
    },
    {
      "auxiliary_loss_clip": 0.01167181,
      "auxiliary_loss_mlp": 0.01030846,
      "balance_loss_clip": 1.05432749,
      "balance_loss_mlp": 1.0221796,
      "epoch": 0.3227319184753201,
      "flos": 30367175852160.0,
      "grad_norm": 1.8488892103490198,
      "language_loss": 0.85220861,
      "learning_rate": 3.1666037122369903e-06,
      "loss": 0.87418884,
      "num_input_tokens_seen": 57828825,
      "step": 2684,
      "time_per_iteration": 2.6028285026550293
    },
    {
      "auxiliary_loss_clip": 0.01173479,
      "auxiliary_loss_mlp": 0.01030209,
      "balance_loss_clip": 1.05186534,
      "balance_loss_mlp": 1.02182901,
      "epoch": 0.32285216136595923,
      "flos": 16946174257920.0,
      "grad_norm": 1.8799019904958336,
      "language_loss": 0.86331713,
      "learning_rate": 3.165970899786928e-06,
      "loss": 0.88535404,
      "num_input_tokens_seen": 57846740,
      "step": 2685,
      "time_per_iteration": 2.467625379562378
    },
    {
      "auxiliary_loss_clip": 0.01148893,
      "auxiliary_loss_mlp": 0.01028768,
      "balance_loss_clip": 1.05100632,
      "balance_loss_mlp": 1.02082872,
      "epoch": 0.32297240425659834,
      "flos": 21981532383360.0,
      "grad_norm": 1.6796931237795973,
      "language_loss": 0.75592911,
      "learning_rate": 3.1653379104600067e-06,
      "loss": 0.77770573,
      "num_input_tokens_seen": 57866885,
      "step": 2686,
      "time_per_iteration": 2.5971837043762207
    },
    {
      "auxiliary_loss_clip": 0.0117496,
      "auxiliary_loss_mlp": 0.0103335,
      "balance_loss_clip": 1.05391312,
      "balance_loss_mlp": 1.02533627,
      "epoch": 0.3230926471472374,
      "flos": 22748045639040.0,
      "grad_norm": 1.4919727754951255,
      "language_loss": 0.69383216,
      "learning_rate": 3.164704744352251e-06,
      "loss": 0.7159152,
      "num_input_tokens_seen": 57887690,
      "step": 2687,
      "time_per_iteration": 2.5362489223480225
    },
    {
      "auxiliary_loss_clip": 0.01174164,
      "auxiliary_loss_mlp": 0.01036351,
      "balance_loss_clip": 1.05433273,
      "balance_loss_mlp": 1.02859366,
      "epoch": 0.3232128900378765,
      "flos": 16942977947520.0,
      "grad_norm": 1.802104443626859,
      "language_loss": 0.8097564,
      "learning_rate": 3.164071401559713e-06,
      "loss": 0.8318615,
      "num_input_tokens_seen": 57905090,
      "step": 2688,
      "time_per_iteration": 2.4820687770843506
    },
    {
      "auxiliary_loss_clip": 0.01164094,
      "auxiliary_loss_mlp": 0.0103561,
      "balance_loss_clip": 1.05433321,
      "balance_loss_mlp": 1.02716422,
      "epoch": 0.3233331329285156,
      "flos": 24023736138240.0,
      "grad_norm": 1.6977396890435756,
      "language_loss": 0.71448612,
      "learning_rate": 3.1634378821784674e-06,
      "loss": 0.7364831,
      "num_input_tokens_seen": 57925305,
      "step": 2689,
      "time_per_iteration": 3.2998459339141846
    },
    {
      "auxiliary_loss_clip": 0.01147915,
      "auxiliary_loss_mlp": 0.01027756,
      "balance_loss_clip": 1.05422831,
      "balance_loss_mlp": 1.01974535,
      "epoch": 0.3234533758191547,
      "flos": 18113845582080.0,
      "grad_norm": 2.2282647007153726,
      "language_loss": 0.73976028,
      "learning_rate": 3.1628041863046208e-06,
      "loss": 0.76151705,
      "num_input_tokens_seen": 57942720,
      "step": 2690,
      "time_per_iteration": 2.523242950439453
    },
    {
      "auxiliary_loss_clip": 0.01192856,
      "auxiliary_loss_mlp": 0.01031491,
      "balance_loss_clip": 1.05513275,
      "balance_loss_mlp": 1.02295589,
      "epoch": 0.3235736187097938,
      "flos": 16946138344320.0,
      "grad_norm": 2.0322075622204294,
      "language_loss": 0.91231847,
      "learning_rate": 3.162170314034304e-06,
      "loss": 0.93456185,
      "num_input_tokens_seen": 57960135,
      "step": 2691,
      "time_per_iteration": 2.4767720699310303
    },
    {
      "auxiliary_loss_clip": 0.01192774,
      "auxiliary_loss_mlp": 0.01034648,
      "balance_loss_clip": 1.05705285,
      "balance_loss_mlp": 1.02523649,
      "epoch": 0.3236938616004329,
      "flos": 22127150119680.0,
      "grad_norm": 1.5068799721655897,
      "language_loss": 0.80879593,
      "learning_rate": 3.1615362654636738e-06,
      "loss": 0.83107018,
      "num_input_tokens_seen": 57980875,
      "step": 2692,
      "time_per_iteration": 2.4932491779327393
    },
    {
      "auxiliary_loss_clip": 0.01143731,
      "auxiliary_loss_mlp": 0.01028186,
      "balance_loss_clip": 1.05539083,
      "balance_loss_mlp": 1.02082467,
      "epoch": 0.32381410449107195,
      "flos": 17164618819200.0,
      "grad_norm": 1.7703835091331566,
      "language_loss": 0.86959296,
      "learning_rate": 3.1609020406889163e-06,
      "loss": 0.89131212,
      "num_input_tokens_seen": 57998310,
      "step": 2693,
      "time_per_iteration": 2.5520403385162354
    },
    {
      "auxiliary_loss_clip": 0.01166215,
      "auxiliary_loss_mlp": 0.01036976,
      "balance_loss_clip": 1.05484378,
      "balance_loss_mlp": 1.02888799,
      "epoch": 0.32393434738171106,
      "flos": 16578125550720.0,
      "grad_norm": 1.6874284054941056,
      "language_loss": 0.8482362,
      "learning_rate": 3.1602676398062416e-06,
      "loss": 0.87026811,
      "num_input_tokens_seen": 58017220,
      "step": 2694,
      "time_per_iteration": 2.5032873153686523
    },
    {
      "auxiliary_loss_clip": 0.01176668,
      "auxiliary_loss_mlp": 0.01032511,
      "balance_loss_clip": 1.05738974,
      "balance_loss_mlp": 1.02423167,
      "epoch": 0.3240545902723502,
      "flos": 25483612602240.0,
      "grad_norm": 2.0368208302838315,
      "language_loss": 0.6113776,
      "learning_rate": 3.1596330629118886e-06,
      "loss": 0.6334694,
      "num_input_tokens_seen": 58037190,
      "step": 2695,
      "time_per_iteration": 3.3222272396087646
    },
    {
      "auxiliary_loss_clip": 0.01128769,
      "auxiliary_loss_mlp": 0.01029448,
      "balance_loss_clip": 1.05058968,
      "balance_loss_mlp": 1.02148533,
      "epoch": 0.32417483316298923,
      "flos": 35845851634560.0,
      "grad_norm": 2.0148952043331514,
      "language_loss": 0.7324667,
      "learning_rate": 3.1589983101021223e-06,
      "loss": 0.75404894,
      "num_input_tokens_seen": 58055820,
      "step": 2696,
      "time_per_iteration": 3.4463837146759033
    },
    {
      "auxiliary_loss_clip": 0.01161817,
      "auxiliary_loss_mlp": 0.01030865,
      "balance_loss_clip": 1.05300069,
      "balance_loss_mlp": 1.0230484,
      "epoch": 0.32429507605362834,
      "flos": 30080501406720.0,
      "grad_norm": 1.902221735412763,
      "language_loss": 0.84459406,
      "learning_rate": 3.1583633814732337e-06,
      "loss": 0.86652088,
      "num_input_tokens_seen": 58075340,
      "step": 2697,
      "time_per_iteration": 2.595353126525879
    },
    {
      "auxiliary_loss_clip": 0.01188433,
      "auxiliary_loss_mlp": 0.01033703,
      "balance_loss_clip": 1.05471683,
      "balance_loss_mlp": 1.02483368,
      "epoch": 0.3244153189442674,
      "flos": 18223265387520.0,
      "grad_norm": 3.13325621450345,
      "language_loss": 0.72047395,
      "learning_rate": 3.157728277121541e-06,
      "loss": 0.74269533,
      "num_input_tokens_seen": 58093515,
      "step": 2698,
      "time_per_iteration": 3.2836217880249023
    },
    {
      "auxiliary_loss_clip": 0.01189792,
      "auxiliary_loss_mlp": 0.01028426,
      "balance_loss_clip": 1.0548569,
      "balance_loss_mlp": 1.02047193,
      "epoch": 0.3245355618349065,
      "flos": 17710317216000.0,
      "grad_norm": 2.7706940251215526,
      "language_loss": 0.78712898,
      "learning_rate": 3.1570929971433897e-06,
      "loss": 0.80931121,
      "num_input_tokens_seen": 58109300,
      "step": 2699,
      "time_per_iteration": 2.425074338912964
    },
    {
      "auxiliary_loss_clip": 0.01178251,
      "auxiliary_loss_mlp": 0.01033414,
      "balance_loss_clip": 1.05926263,
      "balance_loss_mlp": 1.02528119,
      "epoch": 0.3246558047255456,
      "flos": 23440798316160.0,
      "grad_norm": 2.675091527204574,
      "language_loss": 0.83813697,
      "learning_rate": 3.1564575416351504e-06,
      "loss": 0.86025363,
      "num_input_tokens_seen": 58128000,
      "step": 2700,
      "time_per_iteration": 2.5393710136413574
    },
    {
      "auxiliary_loss_clip": 0.01192558,
      "auxiliary_loss_mlp": 0.0102911,
      "balance_loss_clip": 1.05829406,
      "balance_loss_mlp": 1.02050948,
      "epoch": 0.32477604761618467,
      "flos": 21760861178880.0,
      "grad_norm": 2.126879159121157,
      "language_loss": 0.74361873,
      "learning_rate": 3.155821910693221e-06,
      "loss": 0.7658354,
      "num_input_tokens_seen": 58147415,
      "step": 2701,
      "time_per_iteration": 2.4670562744140625
    },
    {
      "auxiliary_loss_clip": 0.01160577,
      "auxiliary_loss_mlp": 0.01031745,
      "balance_loss_clip": 1.05232048,
      "balance_loss_mlp": 1.02300727,
      "epoch": 0.3248962905068238,
      "flos": 19828328624640.0,
      "grad_norm": 1.5987712268787602,
      "language_loss": 0.85843623,
      "learning_rate": 3.1551861044140275e-06,
      "loss": 0.88035947,
      "num_input_tokens_seen": 58167050,
      "step": 2702,
      "time_per_iteration": 2.583242416381836
    },
    {
      "auxiliary_loss_clip": 0.01128653,
      "auxiliary_loss_mlp": 0.01030682,
      "balance_loss_clip": 1.05159807,
      "balance_loss_mlp": 1.02282584,
      "epoch": 0.3250165333974629,
      "flos": 23948215793280.0,
      "grad_norm": 1.7565730512338849,
      "language_loss": 0.77677548,
      "learning_rate": 3.15455012289402e-06,
      "loss": 0.79836887,
      "num_input_tokens_seen": 58186695,
      "step": 2703,
      "time_per_iteration": 2.6574344635009766
    },
    {
      "auxiliary_loss_clip": 0.01180142,
      "auxiliary_loss_mlp": 0.01027634,
      "balance_loss_clip": 1.05871403,
      "balance_loss_mlp": 1.01933062,
      "epoch": 0.32513677628810195,
      "flos": 23989333887360.0,
      "grad_norm": 2.9584641997873877,
      "language_loss": 0.84100384,
      "learning_rate": 3.153913966229677e-06,
      "loss": 0.86308157,
      "num_input_tokens_seen": 58205815,
      "step": 2704,
      "time_per_iteration": 2.572122097015381
    },
    {
      "auxiliary_loss_clip": 0.01077912,
      "auxiliary_loss_mlp": 0.01002846,
      "balance_loss_clip": 1.02524996,
      "balance_loss_mlp": 1.00115287,
      "epoch": 0.32525701917874106,
      "flos": 70655790009600.0,
      "grad_norm": 0.6462304245529202,
      "language_loss": 0.50295472,
      "learning_rate": 3.1532776345175027e-06,
      "loss": 0.52376229,
      "num_input_tokens_seen": 58270960,
      "step": 2705,
      "time_per_iteration": 3.101923704147339
    },
    {
      "auxiliary_loss_clip": 0.01188721,
      "auxiliary_loss_mlp": 0.01027797,
      "balance_loss_clip": 1.05635953,
      "balance_loss_mlp": 1.02018559,
      "epoch": 0.32537726206938017,
      "flos": 19682639061120.0,
      "grad_norm": 1.8580538008003864,
      "language_loss": 0.78358245,
      "learning_rate": 3.1526411278540285e-06,
      "loss": 0.80574763,
      "num_input_tokens_seen": 58289390,
      "step": 2706,
      "time_per_iteration": 2.4639458656311035
    },
    {
      "auxiliary_loss_clip": 0.01167913,
      "auxiliary_loss_mlp": 0.0103819,
      "balance_loss_clip": 1.05293274,
      "balance_loss_mlp": 1.02975011,
      "epoch": 0.3254975049600192,
      "flos": 28760999293440.0,
      "grad_norm": 2.387863359396133,
      "language_loss": 0.81136453,
      "learning_rate": 3.1520044463358116e-06,
      "loss": 0.83342552,
      "num_input_tokens_seen": 58306120,
      "step": 2707,
      "time_per_iteration": 2.6199350357055664
    },
    {
      "auxiliary_loss_clip": 0.01171478,
      "auxiliary_loss_mlp": 0.01034212,
      "balance_loss_clip": 1.05456603,
      "balance_loss_mlp": 1.02677011,
      "epoch": 0.32561774785065833,
      "flos": 18877378008960.0,
      "grad_norm": 1.5908859156961546,
      "language_loss": 0.80231822,
      "learning_rate": 3.151367590059436e-06,
      "loss": 0.82437509,
      "num_input_tokens_seen": 58324545,
      "step": 2708,
      "time_per_iteration": 2.5507938861846924
    },
    {
      "auxiliary_loss_clip": 0.0119257,
      "auxiliary_loss_mlp": 0.00761626,
      "balance_loss_clip": 1.05902362,
      "balance_loss_mlp": 1.00036347,
      "epoch": 0.32573799074129745,
      "flos": 23112107936640.0,
      "grad_norm": 4.49294439836967,
      "language_loss": 0.86844987,
      "learning_rate": 3.1507305591215117e-06,
      "loss": 0.88799185,
      "num_input_tokens_seen": 58342455,
      "step": 2709,
      "time_per_iteration": 2.4843313694000244
    },
    {
      "auxiliary_loss_clip": 0.01074781,
      "auxiliary_loss_mlp": 0.01002841,
      "balance_loss_clip": 1.02304006,
      "balance_loss_mlp": 1.00133348,
      "epoch": 0.3258582336319365,
      "flos": 71237650423680.0,
      "grad_norm": 0.6713513865112859,
      "language_loss": 0.55770934,
      "learning_rate": 3.150093353618677e-06,
      "loss": 0.57848561,
      "num_input_tokens_seen": 58407185,
      "step": 2710,
      "time_per_iteration": 3.1566555500030518
    },
    {
      "auxiliary_loss_clip": 0.01178871,
      "auxiliary_loss_mlp": 0.01032146,
      "balance_loss_clip": 1.05354643,
      "balance_loss_mlp": 1.02409947,
      "epoch": 0.3259784765225756,
      "flos": 22456020067200.0,
      "grad_norm": 2.3493225824300845,
      "language_loss": 0.88633859,
      "learning_rate": 3.149455973647596e-06,
      "loss": 0.9084487,
      "num_input_tokens_seen": 58425245,
      "step": 2711,
      "time_per_iteration": 2.506105899810791
    },
    {
      "auxiliary_loss_clip": 0.01137927,
      "auxiliary_loss_mlp": 0.01037493,
      "balance_loss_clip": 1.04619646,
      "balance_loss_mlp": 1.02949452,
      "epoch": 0.32609871941321467,
      "flos": 20484811543680.0,
      "grad_norm": 1.7900351868850153,
      "language_loss": 0.77288747,
      "learning_rate": 3.1488184193049563e-06,
      "loss": 0.79464173,
      "num_input_tokens_seen": 58444780,
      "step": 2712,
      "time_per_iteration": 2.574157953262329
    },
    {
      "auxiliary_loss_clip": 0.01188959,
      "auxiliary_loss_mlp": 0.01033811,
      "balance_loss_clip": 1.05755162,
      "balance_loss_mlp": 1.02609205,
      "epoch": 0.3262189623038538,
      "flos": 22416805393920.0,
      "grad_norm": 1.5553554168469481,
      "language_loss": 0.720245,
      "learning_rate": 3.1481806906874767e-06,
      "loss": 0.74247271,
      "num_input_tokens_seen": 58466090,
      "step": 2713,
      "time_per_iteration": 2.5082011222839355
    },
    {
      "auxiliary_loss_clip": 0.01190309,
      "auxiliary_loss_mlp": 0.01029179,
      "balance_loss_clip": 1.05787659,
      "balance_loss_mlp": 1.02173471,
      "epoch": 0.3263392051944929,
      "flos": 20923496346240.0,
      "grad_norm": 1.6018333398601972,
      "language_loss": 0.87659931,
      "learning_rate": 3.147542787891899e-06,
      "loss": 0.89879417,
      "num_input_tokens_seen": 58485435,
      "step": 2714,
      "time_per_iteration": 2.4971280097961426
    },
    {
      "auxiliary_loss_clip": 0.01162566,
      "auxiliary_loss_mlp": 0.01030657,
      "balance_loss_clip": 1.05491936,
      "balance_loss_mlp": 1.02239585,
      "epoch": 0.32645944808513194,
      "flos": 24025172682240.0,
      "grad_norm": 1.9100232475315968,
      "language_loss": 0.75222838,
      "learning_rate": 3.1469047110149926e-06,
      "loss": 0.77416062,
      "num_input_tokens_seen": 58504175,
      "step": 2715,
      "time_per_iteration": 3.346435546875
    },
    {
      "auxiliary_loss_clip": 0.01131481,
      "auxiliary_loss_mlp": 0.01032543,
      "balance_loss_clip": 1.05448437,
      "balance_loss_mlp": 1.02425742,
      "epoch": 0.32657969097577105,
      "flos": 21032413361280.0,
      "grad_norm": 1.8524256847397533,
      "language_loss": 0.85097885,
      "learning_rate": 3.146266460153554e-06,
      "loss": 0.87261903,
      "num_input_tokens_seen": 58523885,
      "step": 2716,
      "time_per_iteration": 2.604234457015991
    },
    {
      "auxiliary_loss_clip": 0.01159569,
      "auxiliary_loss_mlp": 0.00762018,
      "balance_loss_clip": 1.05214119,
      "balance_loss_mlp": 1.00036168,
      "epoch": 0.32669993386641016,
      "flos": 22710267509760.0,
      "grad_norm": 1.5782940324467045,
      "language_loss": 0.79963642,
      "learning_rate": 3.145628035404404e-06,
      "loss": 0.81885231,
      "num_input_tokens_seen": 58543085,
      "step": 2717,
      "time_per_iteration": 2.5571608543395996
    },
    {
      "auxiliary_loss_clip": 0.0107023,
      "auxiliary_loss_mlp": 0.01003885,
      "balance_loss_clip": 1.01939845,
      "balance_loss_mlp": 1.00236464,
      "epoch": 0.3268201767570492,
      "flos": 72105718406400.0,
      "grad_norm": 0.9006603205711438,
      "language_loss": 0.57495886,
      "learning_rate": 3.1449894368643922e-06,
      "loss": 0.59570003,
      "num_input_tokens_seen": 58605400,
      "step": 2718,
      "time_per_iteration": 3.164618730545044
    },
    {
      "auxiliary_loss_clip": 0.01152454,
      "auxiliary_loss_mlp": 0.01040249,
      "balance_loss_clip": 1.05986309,
      "balance_loss_mlp": 1.03298354,
      "epoch": 0.32694041964768833,
      "flos": 24535175938560.0,
      "grad_norm": 1.4621272017518916,
      "language_loss": 0.71248221,
      "learning_rate": 3.1443506646303934e-06,
      "loss": 0.73440927,
      "num_input_tokens_seen": 58626700,
      "step": 2719,
      "time_per_iteration": 2.6308202743530273
    },
    {
      "auxiliary_loss_clip": 0.0117605,
      "auxiliary_loss_mlp": 0.01035333,
      "balance_loss_clip": 1.0531739,
      "balance_loss_mlp": 1.02765,
      "epoch": 0.32706066253832744,
      "flos": 33183003755520.0,
      "grad_norm": 7.353733216894235,
      "language_loss": 0.66784394,
      "learning_rate": 3.1437117187993086e-06,
      "loss": 0.6899578,
      "num_input_tokens_seen": 58649020,
      "step": 2720,
      "time_per_iteration": 2.607372283935547
    },
    {
      "auxiliary_loss_clip": 0.01139685,
      "auxiliary_loss_mlp": 0.01030025,
      "balance_loss_clip": 1.04875445,
      "balance_loss_mlp": 1.02206743,
      "epoch": 0.3271809054289665,
      "flos": 24061622008320.0,
      "grad_norm": 1.545054659690862,
      "language_loss": 0.79804105,
      "learning_rate": 3.143072599468065e-06,
      "loss": 0.81973815,
      "num_input_tokens_seen": 58668845,
      "step": 2721,
      "time_per_iteration": 4.12703275680542
    },
    {
      "auxiliary_loss_clip": 0.01162847,
      "auxiliary_loss_mlp": 0.01027879,
      "balance_loss_clip": 1.05552804,
      "balance_loss_mlp": 1.01939714,
      "epoch": 0.3273011483196056,
      "flos": 38253769712640.0,
      "grad_norm": 1.5793637624877568,
      "language_loss": 0.75895953,
      "learning_rate": 3.1424333067336174e-06,
      "loss": 0.7808668,
      "num_input_tokens_seen": 58691610,
      "step": 2722,
      "time_per_iteration": 2.686497926712036
    },
    {
      "auxiliary_loss_clip": 0.01179696,
      "auxiliary_loss_mlp": 0.01034878,
      "balance_loss_clip": 1.05328429,
      "balance_loss_mlp": 1.02637863,
      "epoch": 0.3274213912102447,
      "flos": 29054389582080.0,
      "grad_norm": 1.7647835417419477,
      "language_loss": 0.7811048,
      "learning_rate": 3.141793840692945e-06,
      "loss": 0.80325061,
      "num_input_tokens_seen": 58712360,
      "step": 2723,
      "time_per_iteration": 2.572366237640381
    },
    {
      "auxiliary_loss_clip": 0.01151558,
      "auxiliary_loss_mlp": 0.01032514,
      "balance_loss_clip": 1.05204642,
      "balance_loss_mlp": 1.02399063,
      "epoch": 0.32754163410088377,
      "flos": 29133249891840.0,
      "grad_norm": 1.9916535411542797,
      "language_loss": 0.61171758,
      "learning_rate": 3.1411542014430553e-06,
      "loss": 0.63355827,
      "num_input_tokens_seen": 58733440,
      "step": 2724,
      "time_per_iteration": 3.474208354949951
    },
    {
      "auxiliary_loss_clip": 0.0114448,
      "auxiliary_loss_mlp": 0.01039276,
      "balance_loss_clip": 1.04724705,
      "balance_loss_mlp": 1.03109264,
      "epoch": 0.3276618769915229,
      "flos": 20631075724800.0,
      "grad_norm": 1.6531987056918394,
      "language_loss": 0.8197487,
      "learning_rate": 3.1405143890809804e-06,
      "loss": 0.84158623,
      "num_input_tokens_seen": 58752735,
      "step": 2725,
      "time_per_iteration": 2.5931005477905273
    },
    {
      "auxiliary_loss_clip": 0.01158684,
      "auxiliary_loss_mlp": 0.01032809,
      "balance_loss_clip": 1.05203819,
      "balance_loss_mlp": 1.02494144,
      "epoch": 0.327782119882162,
      "flos": 18657425076480.0,
      "grad_norm": 1.7158991914535384,
      "language_loss": 0.69822943,
      "learning_rate": 3.1398744037037796e-06,
      "loss": 0.72014439,
      "num_input_tokens_seen": 58772070,
      "step": 2726,
      "time_per_iteration": 2.561450242996216
    },
    {
      "auxiliary_loss_clip": 0.01164304,
      "auxiliary_loss_mlp": 0.0103036,
      "balance_loss_clip": 1.05624008,
      "balance_loss_mlp": 1.02220619,
      "epoch": 0.32790236277280105,
      "flos": 21795802133760.0,
      "grad_norm": 3.6370374515952517,
      "language_loss": 0.8450945,
      "learning_rate": 3.139234245408538e-06,
      "loss": 0.86704111,
      "num_input_tokens_seen": 58790950,
      "step": 2727,
      "time_per_iteration": 2.564940929412842
    },
    {
      "auxiliary_loss_clip": 0.01150741,
      "auxiliary_loss_mlp": 0.00762283,
      "balance_loss_clip": 1.05496395,
      "balance_loss_mlp": 1.00031722,
      "epoch": 0.32802260566344016,
      "flos": 23331414424320.0,
      "grad_norm": 1.3269052366302796,
      "language_loss": 0.76134205,
      "learning_rate": 3.1385939142923666e-06,
      "loss": 0.78047234,
      "num_input_tokens_seen": 58813340,
      "step": 2728,
      "time_per_iteration": 2.6398210525512695
    },
    {
      "auxiliary_loss_clip": 0.0116201,
      "auxiliary_loss_mlp": 0.01033609,
      "balance_loss_clip": 1.05210423,
      "balance_loss_mlp": 1.02469182,
      "epoch": 0.3281428485540792,
      "flos": 24206988349440.0,
      "grad_norm": 1.9369738584059697,
      "language_loss": 0.78389728,
      "learning_rate": 3.137953410452405e-06,
      "loss": 0.80585349,
      "num_input_tokens_seen": 58833610,
      "step": 2729,
      "time_per_iteration": 2.6131255626678467
    },
    {
      "auxiliary_loss_clip": 0.01156224,
      "auxiliary_loss_mlp": 0.01032607,
      "balance_loss_clip": 1.04875994,
      "balance_loss_mlp": 1.02414906,
      "epoch": 0.3282630914447183,
      "flos": 34128962380800.0,
      "grad_norm": 2.531757288333698,
      "language_loss": 0.74834931,
      "learning_rate": 3.1373127339858146e-06,
      "loss": 0.77023762,
      "num_input_tokens_seen": 58856210,
      "step": 2730,
      "time_per_iteration": 2.6980152130126953
    },
    {
      "auxiliary_loss_clip": 0.01141139,
      "auxiliary_loss_mlp": 0.01030016,
      "balance_loss_clip": 1.04927754,
      "balance_loss_mlp": 1.02257156,
      "epoch": 0.32838333433535744,
      "flos": 27600726170880.0,
      "grad_norm": 1.7453414053558873,
      "language_loss": 0.7479049,
      "learning_rate": 3.136671884989787e-06,
      "loss": 0.76961637,
      "num_input_tokens_seen": 58876120,
      "step": 2731,
      "time_per_iteration": 2.655062675476074
    },
    {
      "auxiliary_loss_clip": 0.01122058,
      "auxiliary_loss_mlp": 0.01036019,
      "balance_loss_clip": 1.04897451,
      "balance_loss_mlp": 1.02740002,
      "epoch": 0.3285035772259965,
      "flos": 12349500935040.0,
      "grad_norm": 2.649306392081033,
      "language_loss": 0.86878198,
      "learning_rate": 3.1360308635615383e-06,
      "loss": 0.89036274,
      "num_input_tokens_seen": 58894660,
      "step": 2732,
      "time_per_iteration": 2.6797726154327393
    },
    {
      "auxiliary_loss_clip": 0.01169531,
      "auxiliary_loss_mlp": 0.01028369,
      "balance_loss_clip": 1.05489635,
      "balance_loss_mlp": 1.01908255,
      "epoch": 0.3286238201166356,
      "flos": 24316084932480.0,
      "grad_norm": 1.8635316579138768,
      "language_loss": 0.78682673,
      "learning_rate": 3.135389669798311e-06,
      "loss": 0.8088057,
      "num_input_tokens_seen": 58912720,
      "step": 2733,
      "time_per_iteration": 2.6584837436676025
    },
    {
      "auxiliary_loss_clip": 0.01174309,
      "auxiliary_loss_mlp": 0.00761707,
      "balance_loss_clip": 1.05413806,
      "balance_loss_mlp": 1.000278,
      "epoch": 0.3287440630072747,
      "flos": 21392812471680.0,
      "grad_norm": 2.1597863399826673,
      "language_loss": 0.80061519,
      "learning_rate": 3.134748303797373e-06,
      "loss": 0.81997532,
      "num_input_tokens_seen": 58930090,
      "step": 2734,
      "time_per_iteration": 2.5492138862609863
    },
    {
      "auxiliary_loss_clip": 0.01131924,
      "auxiliary_loss_mlp": 0.01033658,
      "balance_loss_clip": 1.04943109,
      "balance_loss_mlp": 1.02431774,
      "epoch": 0.32886430589791377,
      "flos": 23732536579200.0,
      "grad_norm": 1.7768059144153656,
      "language_loss": 0.8112843,
      "learning_rate": 3.1341067656560203e-06,
      "loss": 0.8329401,
      "num_input_tokens_seen": 58947935,
      "step": 2735,
      "time_per_iteration": 2.666851282119751
    },
    {
      "auxiliary_loss_clip": 0.01168043,
      "auxiliary_loss_mlp": 0.01035629,
      "balance_loss_clip": 1.05197322,
      "balance_loss_mlp": 1.02760649,
      "epoch": 0.3289845487885529,
      "flos": 22418708814720.0,
      "grad_norm": 2.2640664974989195,
      "language_loss": 0.86291277,
      "learning_rate": 3.133465055471572e-06,
      "loss": 0.88494945,
      "num_input_tokens_seen": 58967720,
      "step": 2736,
      "time_per_iteration": 2.5866215229034424
    },
    {
      "auxiliary_loss_clip": 0.01142905,
      "auxiliary_loss_mlp": 0.01032043,
      "balance_loss_clip": 1.05273509,
      "balance_loss_mlp": 1.02386487,
      "epoch": 0.329104791679192,
      "flos": 19682603147520.0,
      "grad_norm": 2.210843071015765,
      "language_loss": 0.66612774,
      "learning_rate": 3.1328231733413767e-06,
      "loss": 0.68787718,
      "num_input_tokens_seen": 58984360,
      "step": 2737,
      "time_per_iteration": 2.6438491344451904
    },
    {
      "auxiliary_loss_clip": 0.01172249,
      "auxiliary_loss_mlp": 0.01027752,
      "balance_loss_clip": 1.05476451,
      "balance_loss_mlp": 1.0195682,
      "epoch": 0.32922503456983104,
      "flos": 15997234803840.0,
      "grad_norm": 2.206875015761754,
      "language_loss": 0.90782261,
      "learning_rate": 3.1321811193628067e-06,
      "loss": 0.92982256,
      "num_input_tokens_seen": 59002505,
      "step": 2738,
      "time_per_iteration": 2.533637046813965
    },
    {
      "auxiliary_loss_clip": 0.01175697,
      "auxiliary_loss_mlp": 0.00762798,
      "balance_loss_clip": 1.05777073,
      "balance_loss_mlp": 1.00032806,
      "epoch": 0.32934527746047015,
      "flos": 26834069260800.0,
      "grad_norm": 1.9606160401906019,
      "language_loss": 0.70246625,
      "learning_rate": 3.131538893633261e-06,
      "loss": 0.72185123,
      "num_input_tokens_seen": 59022065,
      "step": 2739,
      "time_per_iteration": 2.6229028701782227
    },
    {
      "auxiliary_loss_clip": 0.01192844,
      "auxiliary_loss_mlp": 0.01033595,
      "balance_loss_clip": 1.05827308,
      "balance_loss_mlp": 1.02590632,
      "epoch": 0.32946552035110926,
      "flos": 23403774372480.0,
      "grad_norm": 2.1682453427527872,
      "language_loss": 0.78371221,
      "learning_rate": 3.130896496250165e-06,
      "loss": 0.80597663,
      "num_input_tokens_seen": 59041890,
      "step": 2740,
      "time_per_iteration": 2.5483930110931396
    },
    {
      "auxiliary_loss_clip": 0.01189709,
      "auxiliary_loss_mlp": 0.01030926,
      "balance_loss_clip": 1.05564237,
      "balance_loss_mlp": 1.02246869,
      "epoch": 0.3295857632417483,
      "flos": 14172470029440.0,
      "grad_norm": 2.0467908361598077,
      "language_loss": 0.86837316,
      "learning_rate": 3.1302539273109693e-06,
      "loss": 0.89057952,
      "num_input_tokens_seen": 59058715,
      "step": 2741,
      "time_per_iteration": 3.2041032314300537
    },
    {
      "auxiliary_loss_clip": 0.01157412,
      "auxiliary_loss_mlp": 0.0103531,
      "balance_loss_clip": 1.05510879,
      "balance_loss_mlp": 1.02701867,
      "epoch": 0.32970600613238743,
      "flos": 22196708807040.0,
      "grad_norm": 1.6788229921084543,
      "language_loss": 0.80675042,
      "learning_rate": 3.1296111869131513e-06,
      "loss": 0.82867765,
      "num_input_tokens_seen": 59076140,
      "step": 2742,
      "time_per_iteration": 2.6136317253112793
    },
    {
      "auxiliary_loss_clip": 0.0118984,
      "auxiliary_loss_mlp": 0.01028374,
      "balance_loss_clip": 1.05642319,
      "balance_loss_mlp": 1.02050614,
      "epoch": 0.32982624902302654,
      "flos": 22053784590720.0,
      "grad_norm": 1.7123549393077075,
      "language_loss": 0.85732198,
      "learning_rate": 3.1289682751542153e-06,
      "loss": 0.87950408,
      "num_input_tokens_seen": 59095700,
      "step": 2743,
      "time_per_iteration": 2.5157485008239746
    },
    {
      "auxiliary_loss_clip": 0.01174111,
      "auxiliary_loss_mlp": 0.01039938,
      "balance_loss_clip": 1.0534749,
      "balance_loss_mlp": 1.03169477,
      "epoch": 0.3299464919136656,
      "flos": 18661626967680.0,
      "grad_norm": 2.003460012113146,
      "language_loss": 0.71528089,
      "learning_rate": 3.1283251921316883e-06,
      "loss": 0.73742139,
      "num_input_tokens_seen": 59113445,
      "step": 2744,
      "time_per_iteration": 2.5308997631073
    },
    {
      "auxiliary_loss_clip": 0.0113307,
      "auxiliary_loss_mlp": 0.01029896,
      "balance_loss_clip": 1.05424511,
      "balance_loss_mlp": 1.02179599,
      "epoch": 0.3300667348043047,
      "flos": 13407357404160.0,
      "grad_norm": 1.8534293225606333,
      "language_loss": 0.8071481,
      "learning_rate": 3.1276819379431277e-06,
      "loss": 0.82877779,
      "num_input_tokens_seen": 59131535,
      "step": 2745,
      "time_per_iteration": 2.6399288177490234
    },
    {
      "auxiliary_loss_clip": 0.01169652,
      "auxiliary_loss_mlp": 0.00761677,
      "balance_loss_clip": 1.05673838,
      "balance_loss_mlp": 1.00027359,
      "epoch": 0.33018697769494376,
      "flos": 15742556398080.0,
      "grad_norm": 2.474496866471333,
      "language_loss": 0.7509402,
      "learning_rate": 3.1270385126861134e-06,
      "loss": 0.77025348,
      "num_input_tokens_seen": 59149520,
      "step": 2746,
      "time_per_iteration": 2.574673652648926
    },
    {
      "auxiliary_loss_clip": 0.01191952,
      "auxiliary_loss_mlp": 0.01035888,
      "balance_loss_clip": 1.05723381,
      "balance_loss_mlp": 1.02790701,
      "epoch": 0.3303072205855829,
      "flos": 18258601392000.0,
      "grad_norm": 1.7894008463564421,
      "language_loss": 0.81990361,
      "learning_rate": 3.1263949164582533e-06,
      "loss": 0.84218198,
      "num_input_tokens_seen": 59169170,
      "step": 2747,
      "time_per_iteration": 3.983107328414917
    },
    {
      "auxiliary_loss_clip": 0.01190331,
      "auxiliary_loss_mlp": 0.01029727,
      "balance_loss_clip": 1.05483842,
      "balance_loss_mlp": 1.0211972,
      "epoch": 0.330427463476222,
      "flos": 17749424148480.0,
      "grad_norm": 1.9480111981355575,
      "language_loss": 0.77727699,
      "learning_rate": 3.1257511493571797e-06,
      "loss": 0.79947758,
      "num_input_tokens_seen": 59187675,
      "step": 2748,
      "time_per_iteration": 2.5156633853912354
    },
    {
      "auxiliary_loss_clip": 0.01150365,
      "auxiliary_loss_mlp": 0.01033398,
      "balance_loss_clip": 1.05326605,
      "balance_loss_mlp": 1.02541685,
      "epoch": 0.33054770636686104,
      "flos": 27162580072320.0,
      "grad_norm": 1.6932416676096262,
      "language_loss": 0.7828669,
      "learning_rate": 3.125107211480552e-06,
      "loss": 0.80470455,
      "num_input_tokens_seen": 59207610,
      "step": 2749,
      "time_per_iteration": 3.5037143230438232
    },
    {
      "auxiliary_loss_clip": 0.01117597,
      "auxiliary_loss_mlp": 0.01032454,
      "balance_loss_clip": 1.04909766,
      "balance_loss_mlp": 1.02458572,
      "epoch": 0.33066794925750015,
      "flos": 20117193799680.0,
      "grad_norm": 1.676027627611513,
      "language_loss": 0.79965723,
      "learning_rate": 3.124463102926054e-06,
      "loss": 0.82115769,
      "num_input_tokens_seen": 59226945,
      "step": 2750,
      "time_per_iteration": 2.6969716548919678
    },
    {
      "auxiliary_loss_clip": 0.01072794,
      "auxiliary_loss_mlp": 0.0100307,
      "balance_loss_clip": 1.02583826,
      "balance_loss_mlp": 1.00165689,
      "epoch": 0.33078819214813926,
      "flos": 70642609718400.0,
      "grad_norm": 0.7627786915019239,
      "language_loss": 0.61694014,
      "learning_rate": 3.1238188237913984e-06,
      "loss": 0.63769877,
      "num_input_tokens_seen": 59291485,
      "step": 2751,
      "time_per_iteration": 3.222346305847168
    },
    {
      "auxiliary_loss_clip": 0.01200296,
      "auxiliary_loss_mlp": 0.01035768,
      "balance_loss_clip": 1.06190431,
      "balance_loss_mlp": 1.02710748,
      "epoch": 0.3309084350387783,
      "flos": 21141940907520.0,
      "grad_norm": 1.9444929296606603,
      "language_loss": 0.76296949,
      "learning_rate": 3.1231743741743202e-06,
      "loss": 0.78533006,
      "num_input_tokens_seen": 59310990,
      "step": 2752,
      "time_per_iteration": 2.534473180770874
    },
    {
      "auxiliary_loss_clip": 0.01171874,
      "auxiliary_loss_mlp": 0.01032142,
      "balance_loss_clip": 1.05313396,
      "balance_loss_mlp": 1.02370238,
      "epoch": 0.3310286779294174,
      "flos": 14209350318720.0,
      "grad_norm": 2.1359025044125475,
      "language_loss": 0.83574426,
      "learning_rate": 3.122529754172582e-06,
      "loss": 0.85778439,
      "num_input_tokens_seen": 59327875,
      "step": 2753,
      "time_per_iteration": 2.526928186416626
    },
    {
      "auxiliary_loss_clip": 0.0117828,
      "auxiliary_loss_mlp": 0.01031137,
      "balance_loss_clip": 1.05664873,
      "balance_loss_mlp": 1.02276897,
      "epoch": 0.33114892082005654,
      "flos": 20778130005120.0,
      "grad_norm": 1.7762631572684908,
      "language_loss": 0.72551751,
      "learning_rate": 3.1218849638839736e-06,
      "loss": 0.74761164,
      "num_input_tokens_seen": 59347135,
      "step": 2754,
      "time_per_iteration": 2.5743556022644043
    },
    {
      "auxiliary_loss_clip": 0.01135986,
      "auxiliary_loss_mlp": 0.01031967,
      "balance_loss_clip": 1.04770517,
      "balance_loss_mlp": 1.02297211,
      "epoch": 0.3312691637106956,
      "flos": 17090750499840.0,
      "grad_norm": 1.808933091809179,
      "language_loss": 0.78529561,
      "learning_rate": 3.121240003406307e-06,
      "loss": 0.80697513,
      "num_input_tokens_seen": 59365985,
      "step": 2755,
      "time_per_iteration": 2.608844518661499
    },
    {
      "auxiliary_loss_clip": 0.01153643,
      "auxiliary_loss_mlp": 0.01030758,
      "balance_loss_clip": 1.055722,
      "balance_loss_mlp": 1.02233613,
      "epoch": 0.3313894066013347,
      "flos": 29456230008960.0,
      "grad_norm": 1.8143974951055173,
      "language_loss": 0.72303003,
      "learning_rate": 3.120594872837425e-06,
      "loss": 0.74487412,
      "num_input_tokens_seen": 59384655,
      "step": 2756,
      "time_per_iteration": 2.6767666339874268
    },
    {
      "auxiliary_loss_clip": 0.01067696,
      "auxiliary_loss_mlp": 0.00752422,
      "balance_loss_clip": 1.01838791,
      "balance_loss_mlp": 1.000458,
      "epoch": 0.3315096494919738,
      "flos": 61419242280960.0,
      "grad_norm": 0.8308167306280487,
      "language_loss": 0.62396312,
      "learning_rate": 3.1199495722751906e-06,
      "loss": 0.64216429,
      "num_input_tokens_seen": 59444185,
      "step": 2757,
      "time_per_iteration": 3.151993989944458
    },
    {
      "auxiliary_loss_clip": 0.01137065,
      "auxiliary_loss_mlp": 0.01033263,
      "balance_loss_clip": 1.04964137,
      "balance_loss_mlp": 1.02507937,
      "epoch": 0.33162989238261287,
      "flos": 21653057485440.0,
      "grad_norm": 2.7570209205688605,
      "language_loss": 0.83463514,
      "learning_rate": 3.1193041018174972e-06,
      "loss": 0.85633844,
      "num_input_tokens_seen": 59464900,
      "step": 2758,
      "time_per_iteration": 2.654470443725586
    },
    {
      "auxiliary_loss_clip": 0.01181138,
      "auxiliary_loss_mlp": 0.01038067,
      "balance_loss_clip": 1.05776823,
      "balance_loss_mlp": 1.029037,
      "epoch": 0.331750135273252,
      "flos": 22674787850880.0,
      "grad_norm": 1.9562257207840053,
      "language_loss": 0.94763362,
      "learning_rate": 3.118658461562261e-06,
      "loss": 0.96982563,
      "num_input_tokens_seen": 59481000,
      "step": 2759,
      "time_per_iteration": 2.557013750076294
    },
    {
      "auxiliary_loss_clip": 0.01163437,
      "auxiliary_loss_mlp": 0.01033806,
      "balance_loss_clip": 1.05728507,
      "balance_loss_mlp": 1.02503228,
      "epoch": 0.33187037816389103,
      "flos": 22746896403840.0,
      "grad_norm": 1.3694378427038796,
      "language_loss": 0.84908628,
      "learning_rate": 3.118012651607426e-06,
      "loss": 0.8710587,
      "num_input_tokens_seen": 59502605,
      "step": 2760,
      "time_per_iteration": 2.610518217086792
    },
    {
      "auxiliary_loss_clip": 0.01189626,
      "auxiliary_loss_mlp": 0.01029668,
      "balance_loss_clip": 1.05629873,
      "balance_loss_mlp": 1.02120996,
      "epoch": 0.33199062105453014,
      "flos": 19203769918080.0,
      "grad_norm": 2.3366222055349217,
      "language_loss": 0.83406103,
      "learning_rate": 3.1173666720509603e-06,
      "loss": 0.85625398,
      "num_input_tokens_seen": 59519540,
      "step": 2761,
      "time_per_iteration": 2.464991569519043
    },
    {
      "auxiliary_loss_clip": 0.01167727,
      "auxiliary_loss_mlp": 0.01030567,
      "balance_loss_clip": 1.05404973,
      "balance_loss_mlp": 1.02206743,
      "epoch": 0.33211086394516925,
      "flos": 31577006764800.0,
      "grad_norm": 1.704019387385593,
      "language_loss": 0.68158388,
      "learning_rate": 3.116720522990859e-06,
      "loss": 0.70356685,
      "num_input_tokens_seen": 59540415,
      "step": 2762,
      "time_per_iteration": 2.639462471008301
    },
    {
      "auxiliary_loss_clip": 0.01120063,
      "auxiliary_loss_mlp": 0.01034166,
      "balance_loss_clip": 1.05033457,
      "balance_loss_mlp": 1.02556539,
      "epoch": 0.3322311068358083,
      "flos": 17932496791680.0,
      "grad_norm": 1.732106871289698,
      "language_loss": 0.62254059,
      "learning_rate": 3.116074204525142e-06,
      "loss": 0.6440829,
      "num_input_tokens_seen": 59558590,
      "step": 2763,
      "time_per_iteration": 2.645892858505249
    },
    {
      "auxiliary_loss_clip": 0.01167822,
      "auxiliary_loss_mlp": 0.01032664,
      "balance_loss_clip": 1.05402088,
      "balance_loss_mlp": 1.02388406,
      "epoch": 0.3323513497264474,
      "flos": 32269831269120.0,
      "grad_norm": 1.4403111684202343,
      "language_loss": 0.83469957,
      "learning_rate": 3.1154277167518553e-06,
      "loss": 0.85670441,
      "num_input_tokens_seen": 59580205,
      "step": 2764,
      "time_per_iteration": 2.632929563522339
    },
    {
      "auxiliary_loss_clip": 0.01051608,
      "auxiliary_loss_mlp": 0.01008725,
      "balance_loss_clip": 1.01588202,
      "balance_loss_mlp": 1.00716352,
      "epoch": 0.33247159261708653,
      "flos": 52668674588160.0,
      "grad_norm": 0.7806237321066729,
      "language_loss": 0.59484404,
      "learning_rate": 3.114781059769072e-06,
      "loss": 0.61544734,
      "num_input_tokens_seen": 59631530,
      "step": 2765,
      "time_per_iteration": 2.9860823154449463
    },
    {
      "auxiliary_loss_clip": 0.01162701,
      "auxiliary_loss_mlp": 0.01030087,
      "balance_loss_clip": 1.05218077,
      "balance_loss_mlp": 1.0213666,
      "epoch": 0.3325918355077256,
      "flos": 27125232906240.0,
      "grad_norm": 3.886624193414405,
      "language_loss": 0.67224443,
      "learning_rate": 3.1141342336748874e-06,
      "loss": 0.69417232,
      "num_input_tokens_seen": 59651090,
      "step": 2766,
      "time_per_iteration": 2.6145172119140625
    },
    {
      "auxiliary_loss_clip": 0.01177613,
      "auxiliary_loss_mlp": 0.01033033,
      "balance_loss_clip": 1.05881524,
      "balance_loss_mlp": 1.02517748,
      "epoch": 0.3327120783983647,
      "flos": 23664414435840.0,
      "grad_norm": 1.4338403622526639,
      "language_loss": 0.82168186,
      "learning_rate": 3.1134872385674253e-06,
      "loss": 0.84378827,
      "num_input_tokens_seen": 59675245,
      "step": 2767,
      "time_per_iteration": 3.3863301277160645
    },
    {
      "auxiliary_loss_clip": 0.01163542,
      "auxiliary_loss_mlp": 0.01028764,
      "balance_loss_clip": 1.0504458,
      "balance_loss_mlp": 1.02086878,
      "epoch": 0.3328323212890038,
      "flos": 19171378828800.0,
      "grad_norm": 1.801009305326311,
      "language_loss": 0.85479379,
      "learning_rate": 3.1128400745448353e-06,
      "loss": 0.87671679,
      "num_input_tokens_seen": 59694625,
      "step": 2768,
      "time_per_iteration": 2.5556745529174805
    },
    {
      "auxiliary_loss_clip": 0.0117963,
      "auxiliary_loss_mlp": 0.01033306,
      "balance_loss_clip": 1.05656183,
      "balance_loss_mlp": 1.02527726,
      "epoch": 0.33295256417964286,
      "flos": 37706347463040.0,
      "grad_norm": 2.618110291672558,
      "language_loss": 0.629264,
      "learning_rate": 3.11219274170529e-06,
      "loss": 0.65139341,
      "num_input_tokens_seen": 59716435,
      "step": 2769,
      "time_per_iteration": 2.648175001144409
    },
    {
      "auxiliary_loss_clip": 0.0115509,
      "auxiliary_loss_mlp": 0.01033546,
      "balance_loss_clip": 1.05046296,
      "balance_loss_mlp": 1.02582097,
      "epoch": 0.333072807070282,
      "flos": 26505989412480.0,
      "grad_norm": 1.875694110402851,
      "language_loss": 0.81791508,
      "learning_rate": 3.1115452401469903e-06,
      "loss": 0.83980143,
      "num_input_tokens_seen": 59736835,
      "step": 2770,
      "time_per_iteration": 2.6086814403533936
    },
    {
      "auxiliary_loss_clip": 0.01122061,
      "auxiliary_loss_mlp": 0.01031263,
      "balance_loss_clip": 1.04608107,
      "balance_loss_mlp": 1.02384245,
      "epoch": 0.3331930499609211,
      "flos": 21430913823360.0,
      "grad_norm": 2.0117282403765464,
      "language_loss": 0.86488992,
      "learning_rate": 3.1108975699681613e-06,
      "loss": 0.88642317,
      "num_input_tokens_seen": 59754230,
      "step": 2771,
      "time_per_iteration": 2.6085333824157715
    },
    {
      "auxiliary_loss_clip": 0.01147027,
      "auxiliary_loss_mlp": 0.01035928,
      "balance_loss_clip": 1.05264485,
      "balance_loss_mlp": 1.02839684,
      "epoch": 0.33331329285156014,
      "flos": 20659947281280.0,
      "grad_norm": 2.010846690736538,
      "language_loss": 0.71842688,
      "learning_rate": 3.1102497312670542e-06,
      "loss": 0.74025643,
      "num_input_tokens_seen": 59772235,
      "step": 2772,
      "time_per_iteration": 3.3850274085998535
    },
    {
      "auxiliary_loss_clip": 0.01151412,
      "auxiliary_loss_mlp": 0.0102869,
      "balance_loss_clip": 1.05319405,
      "balance_loss_mlp": 1.02075684,
      "epoch": 0.33343353574219925,
      "flos": 28001596930560.0,
      "grad_norm": 2.016638957595795,
      "language_loss": 0.80477053,
      "learning_rate": 3.109601724141946e-06,
      "loss": 0.82657152,
      "num_input_tokens_seen": 59791230,
      "step": 2773,
      "time_per_iteration": 3.3427720069885254
    },
    {
      "auxiliary_loss_clip": 0.01156323,
      "auxiliary_loss_mlp": 0.01028335,
      "balance_loss_clip": 1.05165076,
      "balance_loss_mlp": 1.02002597,
      "epoch": 0.33355377863283836,
      "flos": 23764963582080.0,
      "grad_norm": 1.6426050152662954,
      "language_loss": 0.67951334,
      "learning_rate": 3.108953548691138e-06,
      "loss": 0.70135987,
      "num_input_tokens_seen": 59811315,
      "step": 2774,
      "time_per_iteration": 2.5613765716552734
    },
    {
      "auxiliary_loss_clip": 0.0119178,
      "auxiliary_loss_mlp": 0.0102882,
      "balance_loss_clip": 1.05712712,
      "balance_loss_mlp": 1.02036762,
      "epoch": 0.3336740215234774,
      "flos": 37779677078400.0,
      "grad_norm": 2.1292685985883097,
      "language_loss": 0.72697675,
      "learning_rate": 3.108305205012959e-06,
      "loss": 0.7491827,
      "num_input_tokens_seen": 59832010,
      "step": 2775,
      "time_per_iteration": 3.4604666233062744
    },
    {
      "auxiliary_loss_clip": 0.0116306,
      "auxiliary_loss_mlp": 0.01028951,
      "balance_loss_clip": 1.05439198,
      "balance_loss_mlp": 1.02104211,
      "epoch": 0.3337942644141165,
      "flos": 25519056347520.0,
      "grad_norm": 2.293062378267243,
      "language_loss": 0.87311637,
      "learning_rate": 3.107656693205761e-06,
      "loss": 0.89503646,
      "num_input_tokens_seen": 59851450,
      "step": 2776,
      "time_per_iteration": 2.616868019104004
    },
    {
      "auxiliary_loss_clip": 0.01196814,
      "auxiliary_loss_mlp": 0.01033485,
      "balance_loss_clip": 1.05877006,
      "balance_loss_mlp": 1.02416921,
      "epoch": 0.3339145073047556,
      "flos": 25989844930560.0,
      "grad_norm": 4.014427190952157,
      "language_loss": 0.70470929,
      "learning_rate": 3.107008013367924e-06,
      "loss": 0.72701228,
      "num_input_tokens_seen": 59870245,
      "step": 2777,
      "time_per_iteration": 2.5142111778259277
    },
    {
      "auxiliary_loss_clip": 0.01146123,
      "auxiliary_loss_mlp": 0.01028223,
      "balance_loss_clip": 1.0510112,
      "balance_loss_mlp": 1.02070725,
      "epoch": 0.3340347501953947,
      "flos": 19062569554560.0,
      "grad_norm": 1.8741735741192642,
      "language_loss": 0.86698985,
      "learning_rate": 3.1063591655978507e-06,
      "loss": 0.88873327,
      "num_input_tokens_seen": 59886195,
      "step": 2778,
      "time_per_iteration": 2.581897497177124
    },
    {
      "auxiliary_loss_clip": 0.01120684,
      "auxiliary_loss_mlp": 0.01027746,
      "balance_loss_clip": 1.04538083,
      "balance_loss_mlp": 1.01946676,
      "epoch": 0.3341549930860338,
      "flos": 18109715518080.0,
      "grad_norm": 1.8599079760513766,
      "language_loss": 0.79472232,
      "learning_rate": 3.105710149993972e-06,
      "loss": 0.81620663,
      "num_input_tokens_seen": 59905525,
      "step": 2779,
      "time_per_iteration": 2.6028857231140137
    },
    {
      "auxiliary_loss_clip": 0.01191275,
      "auxiliary_loss_mlp": 0.01036671,
      "balance_loss_clip": 1.05697846,
      "balance_loss_mlp": 1.0284512,
      "epoch": 0.33427523597667286,
      "flos": 22674967418880.0,
      "grad_norm": 2.075112232561446,
      "language_loss": 0.8538534,
      "learning_rate": 3.1050609666547427e-06,
      "loss": 0.87613285,
      "num_input_tokens_seen": 59925085,
      "step": 2780,
      "time_per_iteration": 2.5111191272735596
    },
    {
      "auxiliary_loss_clip": 0.01154694,
      "auxiliary_loss_mlp": 0.01038985,
      "balance_loss_clip": 1.0551194,
      "balance_loss_mlp": 1.03074145,
      "epoch": 0.33439547886731197,
      "flos": 22638338524800.0,
      "grad_norm": 1.9617442772195952,
      "language_loss": 0.77440172,
      "learning_rate": 3.104411615678644e-06,
      "loss": 0.7963385,
      "num_input_tokens_seen": 59943935,
      "step": 2781,
      "time_per_iteration": 2.5984508991241455
    },
    {
      "auxiliary_loss_clip": 0.0115671,
      "auxiliary_loss_mlp": 0.01041678,
      "balance_loss_clip": 1.05235517,
      "balance_loss_mlp": 1.03285027,
      "epoch": 0.3345157217579511,
      "flos": 24096383395200.0,
      "grad_norm": 7.865890284207227,
      "language_loss": 0.73277342,
      "learning_rate": 3.1037620971641803e-06,
      "loss": 0.75475729,
      "num_input_tokens_seen": 59963725,
      "step": 2782,
      "time_per_iteration": 2.585015296936035
    },
    {
      "auxiliary_loss_clip": 0.0119025,
      "auxiliary_loss_mlp": 0.01037698,
      "balance_loss_clip": 1.05544424,
      "balance_loss_mlp": 1.0283227,
      "epoch": 0.33463596464859013,
      "flos": 18989491334400.0,
      "grad_norm": 2.984575702655751,
      "language_loss": 0.6515348,
      "learning_rate": 3.1031124112098844e-06,
      "loss": 0.6738143,
      "num_input_tokens_seen": 59981935,
      "step": 2783,
      "time_per_iteration": 2.4900197982788086
    },
    {
      "auxiliary_loss_clip": 0.0116661,
      "auxiliary_loss_mlp": 0.01033239,
      "balance_loss_clip": 1.05442941,
      "balance_loss_mlp": 1.02484632,
      "epoch": 0.33475620753922924,
      "flos": 20375607219840.0,
      "grad_norm": 1.8656066285759203,
      "language_loss": 0.72581875,
      "learning_rate": 3.1024625579143127e-06,
      "loss": 0.74781722,
      "num_input_tokens_seen": 59999455,
      "step": 2784,
      "time_per_iteration": 2.560081720352173
    },
    {
      "auxiliary_loss_clip": 0.01191043,
      "auxiliary_loss_mlp": 0.01030768,
      "balance_loss_clip": 1.05780494,
      "balance_loss_mlp": 1.02262604,
      "epoch": 0.33487645042986836,
      "flos": 18182578256640.0,
      "grad_norm": 1.9874618109270874,
      "language_loss": 0.732072,
      "learning_rate": 3.101812537376048e-06,
      "loss": 0.7542901,
      "num_input_tokens_seen": 60018475,
      "step": 2785,
      "time_per_iteration": 2.4848573207855225
    },
    {
      "auxiliary_loss_clip": 0.01152986,
      "auxiliary_loss_mlp": 0.00761565,
      "balance_loss_clip": 1.0525943,
      "balance_loss_mlp": 1.00025129,
      "epoch": 0.3349966933205074,
      "flos": 25848824135040.0,
      "grad_norm": 2.022043152550842,
      "language_loss": 0.83934689,
      "learning_rate": 3.1011623496936973e-06,
      "loss": 0.85849237,
      "num_input_tokens_seen": 60036770,
      "step": 2786,
      "time_per_iteration": 2.5965471267700195
    },
    {
      "auxiliary_loss_clip": 0.01188979,
      "auxiliary_loss_mlp": 0.01033154,
      "balance_loss_clip": 1.05653226,
      "balance_loss_mlp": 1.02538157,
      "epoch": 0.3351169362111465,
      "flos": 28111447699200.0,
      "grad_norm": 2.697073426846691,
      "language_loss": 0.69428551,
      "learning_rate": 3.100511994965893e-06,
      "loss": 0.71650684,
      "num_input_tokens_seen": 60056725,
      "step": 2787,
      "time_per_iteration": 2.531032085418701
    },
    {
      "auxiliary_loss_clip": 0.01166323,
      "auxiliary_loss_mlp": 0.01039735,
      "balance_loss_clip": 1.05203164,
      "balance_loss_mlp": 1.03142571,
      "epoch": 0.33523717910178563,
      "flos": 22673315393280.0,
      "grad_norm": 1.6062580179427184,
      "language_loss": 0.8428483,
      "learning_rate": 3.0998614732912947e-06,
      "loss": 0.86490887,
      "num_input_tokens_seen": 60076100,
      "step": 2788,
      "time_per_iteration": 2.546074151992798
    },
    {
      "auxiliary_loss_clip": 0.01174853,
      "auxiliary_loss_mlp": 0.01032863,
      "balance_loss_clip": 1.05619121,
      "balance_loss_mlp": 1.02389288,
      "epoch": 0.3353574219924247,
      "flos": 15669801400320.0,
      "grad_norm": 2.227338128291741,
      "language_loss": 0.68426156,
      "learning_rate": 3.0992107847685855e-06,
      "loss": 0.70633876,
      "num_input_tokens_seen": 60093815,
      "step": 2789,
      "time_per_iteration": 2.472705125808716
    },
    {
      "auxiliary_loss_clip": 0.01164388,
      "auxiliary_loss_mlp": 0.01038141,
      "balance_loss_clip": 1.0572772,
      "balance_loss_mlp": 1.02976644,
      "epoch": 0.3354776648830638,
      "flos": 24790644443520.0,
      "grad_norm": 1.7158237031351875,
      "language_loss": 0.79667324,
      "learning_rate": 3.0985599294964736e-06,
      "loss": 0.81869847,
      "num_input_tokens_seen": 60113370,
      "step": 2790,
      "time_per_iteration": 2.579221248626709
    },
    {
      "auxiliary_loss_clip": 0.01157669,
      "auxiliary_loss_mlp": 0.01037215,
      "balance_loss_clip": 1.05213571,
      "balance_loss_mlp": 1.02877474,
      "epoch": 0.33559790777370285,
      "flos": 28694852398080.0,
      "grad_norm": 1.8772414419336398,
      "language_loss": 0.6990332,
      "learning_rate": 3.097908907573695e-06,
      "loss": 0.72098207,
      "num_input_tokens_seen": 60131350,
      "step": 2791,
      "time_per_iteration": 2.62852144241333
    },
    {
      "auxiliary_loss_clip": 0.01121277,
      "auxiliary_loss_mlp": 0.01035459,
      "balance_loss_clip": 1.05153251,
      "balance_loss_mlp": 1.02746582,
      "epoch": 0.33571815066434196,
      "flos": 22235779825920.0,
      "grad_norm": 1.9235489123786909,
      "language_loss": 0.89623404,
      "learning_rate": 3.0972577190990067e-06,
      "loss": 0.91780138,
      "num_input_tokens_seen": 60149830,
      "step": 2792,
      "time_per_iteration": 2.6461944580078125
    },
    {
      "auxiliary_loss_clip": 0.01151419,
      "auxiliary_loss_mlp": 0.01030589,
      "balance_loss_clip": 1.05165362,
      "balance_loss_mlp": 1.02253628,
      "epoch": 0.3358383935549811,
      "flos": 23842279607040.0,
      "grad_norm": 1.8340833430337022,
      "language_loss": 0.79663914,
      "learning_rate": 3.096606364171196e-06,
      "loss": 0.81845927,
      "num_input_tokens_seen": 60169620,
      "step": 2793,
      "time_per_iteration": 3.3135764598846436
    },
    {
      "auxiliary_loss_clip": 0.01132791,
      "auxiliary_loss_mlp": 0.0103006,
      "balance_loss_clip": 1.04867601,
      "balance_loss_mlp": 1.02207875,
      "epoch": 0.33595863644562013,
      "flos": 22267308988800.0,
      "grad_norm": 1.7378180038215276,
      "language_loss": 0.84903878,
      "learning_rate": 3.0959548428890703e-06,
      "loss": 0.87066722,
      "num_input_tokens_seen": 60188490,
      "step": 2794,
      "time_per_iteration": 2.5777885913848877
    },
    {
      "auxiliary_loss_clip": 0.01175729,
      "auxiliary_loss_mlp": 0.01032814,
      "balance_loss_clip": 1.05611241,
      "balance_loss_mlp": 1.02458274,
      "epoch": 0.33607887933625924,
      "flos": 20119779578880.0,
      "grad_norm": 1.8117735820156302,
      "language_loss": 0.83922189,
      "learning_rate": 3.095303155351468e-06,
      "loss": 0.86130732,
      "num_input_tokens_seen": 60208695,
      "step": 2795,
      "time_per_iteration": 2.5712294578552246
    },
    {
      "auxiliary_loss_clip": 0.01125677,
      "auxiliary_loss_mlp": 0.0103854,
      "balance_loss_clip": 1.05122256,
      "balance_loss_mlp": 1.02967715,
      "epoch": 0.33619912222689835,
      "flos": 19318109886720.0,
      "grad_norm": 2.1665663620046787,
      "language_loss": 0.79349971,
      "learning_rate": 3.0946513016572464e-06,
      "loss": 0.81514192,
      "num_input_tokens_seen": 60227600,
      "step": 2796,
      "time_per_iteration": 2.638458728790283
    },
    {
      "auxiliary_loss_clip": 0.01176085,
      "auxiliary_loss_mlp": 0.01033815,
      "balance_loss_clip": 1.0537467,
      "balance_loss_mlp": 1.02448738,
      "epoch": 0.3363193651175374,
      "flos": 16800664262400.0,
      "grad_norm": 1.8667049658915744,
      "language_loss": 0.77005142,
      "learning_rate": 3.0939992819052938e-06,
      "loss": 0.79215038,
      "num_input_tokens_seen": 60245110,
      "step": 2797,
      "time_per_iteration": 2.48612380027771
    },
    {
      "auxiliary_loss_clip": 0.0116314,
      "auxiliary_loss_mlp": 0.01032059,
      "balance_loss_clip": 1.05588043,
      "balance_loss_mlp": 1.02307034,
      "epoch": 0.3364396080081765,
      "flos": 23550289948800.0,
      "grad_norm": 1.8498968668099007,
      "language_loss": 0.80902696,
      "learning_rate": 3.0933470961945193e-06,
      "loss": 0.83097899,
      "num_input_tokens_seen": 60263405,
      "step": 2798,
      "time_per_iteration": 3.356311798095703
    },
    {
      "auxiliary_loss_clip": 0.01158257,
      "auxiliary_loss_mlp": 0.01038165,
      "balance_loss_clip": 1.05419207,
      "balance_loss_mlp": 1.02962947,
      "epoch": 0.3365598508988156,
      "flos": 28037902602240.0,
      "grad_norm": 1.6774446358835056,
      "language_loss": 0.68206668,
      "learning_rate": 3.0926947446238597e-06,
      "loss": 0.70403087,
      "num_input_tokens_seen": 60282975,
      "step": 2799,
      "time_per_iteration": 3.313974380493164
    },
    {
      "auxiliary_loss_clip": 0.01178751,
      "auxiliary_loss_mlp": 0.01027048,
      "balance_loss_clip": 1.05204892,
      "balance_loss_mlp": 1.0184114,
      "epoch": 0.3366800937894547,
      "flos": 16982767238400.0,
      "grad_norm": 2.3878338157915295,
      "language_loss": 0.82329828,
      "learning_rate": 3.092042227292276e-06,
      "loss": 0.84535629,
      "num_input_tokens_seen": 60299810,
      "step": 2800,
      "time_per_iteration": 2.5129904747009277
    },
    {
      "auxiliary_loss_clip": 0.01189863,
      "auxiliary_loss_mlp": 0.01036857,
      "balance_loss_clip": 1.05771589,
      "balance_loss_mlp": 1.02843499,
      "epoch": 0.3368003366800938,
      "flos": 23915321913600.0,
      "grad_norm": 1.5429127966283056,
      "language_loss": 0.87740183,
      "learning_rate": 3.0913895442987557e-06,
      "loss": 0.89966899,
      "num_input_tokens_seen": 60320775,
      "step": 2801,
      "time_per_iteration": 3.3553528785705566
    },
    {
      "auxiliary_loss_clip": 0.01146949,
      "auxiliary_loss_mlp": 0.00761997,
      "balance_loss_clip": 1.05314827,
      "balance_loss_mlp": 1.0002563,
      "epoch": 0.3369205795707329,
      "flos": 24791219061120.0,
      "grad_norm": 1.5694243333978366,
      "language_loss": 0.85833454,
      "learning_rate": 3.090736695742308e-06,
      "loss": 0.877424,
      "num_input_tokens_seen": 60341905,
      "step": 2802,
      "time_per_iteration": 2.6145522594451904
    },
    {
      "auxiliary_loss_clip": 0.01129569,
      "auxiliary_loss_mlp": 0.0103306,
      "balance_loss_clip": 1.05161941,
      "balance_loss_mlp": 1.02509046,
      "epoch": 0.33704082246137196,
      "flos": 17931096161280.0,
      "grad_norm": 2.2847277011876828,
      "language_loss": 0.52509338,
      "learning_rate": 3.0900836817219713e-06,
      "loss": 0.54671967,
      "num_input_tokens_seen": 60358335,
      "step": 2803,
      "time_per_iteration": 2.6026904582977295
    },
    {
      "auxiliary_loss_clip": 0.01190892,
      "auxiliary_loss_mlp": 0.01032457,
      "balance_loss_clip": 1.05671525,
      "balance_loss_mlp": 1.02435076,
      "epoch": 0.33716106535201107,
      "flos": 21286517149440.0,
      "grad_norm": 1.5846024561861811,
      "language_loss": 0.83552849,
      "learning_rate": 3.089430502336807e-06,
      "loss": 0.85776198,
      "num_input_tokens_seen": 60378305,
      "step": 2804,
      "time_per_iteration": 2.495375156402588
    },
    {
      "auxiliary_loss_clip": 0.01181283,
      "auxiliary_loss_mlp": 0.01036679,
      "balance_loss_clip": 1.05780101,
      "balance_loss_mlp": 1.02791071,
      "epoch": 0.3372813082426502,
      "flos": 18402962152320.0,
      "grad_norm": 3.8361396244985366,
      "language_loss": 0.90315086,
      "learning_rate": 3.088777157685902e-06,
      "loss": 0.92533046,
      "num_input_tokens_seen": 60393895,
      "step": 2805,
      "time_per_iteration": 2.485994338989258
    },
    {
      "auxiliary_loss_clip": 0.01160191,
      "auxiliary_loss_mlp": 0.01027439,
      "balance_loss_clip": 1.05477929,
      "balance_loss_mlp": 1.01983976,
      "epoch": 0.33740155113328923,
      "flos": 17201391367680.0,
      "grad_norm": 2.0103006718874736,
      "language_loss": 0.85996425,
      "learning_rate": 3.088123647868367e-06,
      "loss": 0.88184059,
      "num_input_tokens_seen": 60410445,
      "step": 2806,
      "time_per_iteration": 2.5338659286499023
    },
    {
      "auxiliary_loss_clip": 0.01181153,
      "auxiliary_loss_mlp": 0.01030724,
      "balance_loss_clip": 1.05569839,
      "balance_loss_mlp": 1.02209377,
      "epoch": 0.33752179402392835,
      "flos": 29058950609280.0,
      "grad_norm": 1.8241599806688602,
      "language_loss": 0.81265974,
      "learning_rate": 3.0874699729833405e-06,
      "loss": 0.83477843,
      "num_input_tokens_seen": 60431815,
      "step": 2807,
      "time_per_iteration": 2.5975427627563477
    },
    {
      "auxiliary_loss_clip": 0.01157675,
      "auxiliary_loss_mlp": 0.01027428,
      "balance_loss_clip": 1.0527966,
      "balance_loss_mlp": 1.01934564,
      "epoch": 0.3376420369145674,
      "flos": 25080730680960.0,
      "grad_norm": 1.568958079945656,
      "language_loss": 0.79569912,
      "learning_rate": 3.086816133129983e-06,
      "loss": 0.81755012,
      "num_input_tokens_seen": 60452075,
      "step": 2808,
      "time_per_iteration": 2.605307102203369
    },
    {
      "auxiliary_loss_clip": 0.01194462,
      "auxiliary_loss_mlp": 0.01035035,
      "balance_loss_clip": 1.05949771,
      "balance_loss_mlp": 1.02713132,
      "epoch": 0.3377622798052065,
      "flos": 27490624007040.0,
      "grad_norm": 1.923238008300245,
      "language_loss": 0.76340878,
      "learning_rate": 3.0861621284074826e-06,
      "loss": 0.78570378,
      "num_input_tokens_seen": 60472600,
      "step": 2809,
      "time_per_iteration": 2.5336546897888184
    },
    {
      "auxiliary_loss_clip": 0.01170634,
      "auxiliary_loss_mlp": 0.01038935,
      "balance_loss_clip": 1.05683911,
      "balance_loss_mlp": 1.03134727,
      "epoch": 0.3378825226958456,
      "flos": 21975211589760.0,
      "grad_norm": 1.8511064054458861,
      "language_loss": 0.7292335,
      "learning_rate": 3.085507958915051e-06,
      "loss": 0.75132918,
      "num_input_tokens_seen": 60491030,
      "step": 2810,
      "time_per_iteration": 2.5961081981658936
    },
    {
      "auxiliary_loss_clip": 0.01163064,
      "auxiliary_loss_mlp": 0.0103852,
      "balance_loss_clip": 1.05676854,
      "balance_loss_mlp": 1.02937102,
      "epoch": 0.3380027655864847,
      "flos": 42523189200000.0,
      "grad_norm": 1.9966346595225377,
      "language_loss": 0.71038234,
      "learning_rate": 3.084853624751925e-06,
      "loss": 0.73239821,
      "num_input_tokens_seen": 60512615,
      "step": 2811,
      "time_per_iteration": 2.7424263954162598
    },
    {
      "auxiliary_loss_clip": 0.01151478,
      "auxiliary_loss_mlp": 0.01027376,
      "balance_loss_clip": 1.05503249,
      "balance_loss_mlp": 1.01865566,
      "epoch": 0.3381230084771238,
      "flos": 26725080418560.0,
      "grad_norm": 1.6389833589842036,
      "language_loss": 0.8572036,
      "learning_rate": 3.0841991260173668e-06,
      "loss": 0.8789922,
      "num_input_tokens_seen": 60532520,
      "step": 2812,
      "time_per_iteration": 2.636681318283081
    },
    {
      "auxiliary_loss_clip": 0.01193057,
      "auxiliary_loss_mlp": 0.01032131,
      "balance_loss_clip": 1.05889618,
      "balance_loss_mlp": 1.02369142,
      "epoch": 0.3382432513677629,
      "flos": 22710375250560.0,
      "grad_norm": 1.8027076801080775,
      "language_loss": 0.80156368,
      "learning_rate": 3.0835444628106634e-06,
      "loss": 0.82381558,
      "num_input_tokens_seen": 60551500,
      "step": 2813,
      "time_per_iteration": 2.4759881496429443
    },
    {
      "auxiliary_loss_clip": 0.01192352,
      "auxiliary_loss_mlp": 0.00761946,
      "balance_loss_clip": 1.05862045,
      "balance_loss_mlp": 1.00026762,
      "epoch": 0.33836349425840195,
      "flos": 22122409524480.0,
      "grad_norm": 1.872294907964951,
      "language_loss": 0.8309536,
      "learning_rate": 3.082889635231126e-06,
      "loss": 0.85049653,
      "num_input_tokens_seen": 60570160,
      "step": 2814,
      "time_per_iteration": 2.5127387046813965
    },
    {
      "auxiliary_loss_clip": 0.01164526,
      "auxiliary_loss_mlp": 0.01041515,
      "balance_loss_clip": 1.05288625,
      "balance_loss_mlp": 1.03273559,
      "epoch": 0.33848373714904106,
      "flos": 27308090067840.0,
      "grad_norm": 2.187105608582856,
      "language_loss": 0.76437581,
      "learning_rate": 3.0822346433780925e-06,
      "loss": 0.78643626,
      "num_input_tokens_seen": 60590885,
      "step": 2815,
      "time_per_iteration": 2.597376585006714
    },
    {
      "auxiliary_loss_clip": 0.01174435,
      "auxiliary_loss_mlp": 0.01034725,
      "balance_loss_clip": 1.05118477,
      "balance_loss_mlp": 1.02583754,
      "epoch": 0.3386039800396802,
      "flos": 25848716394240.0,
      "grad_norm": 2.727904837231005,
      "language_loss": 0.87216246,
      "learning_rate": 3.0815794873509237e-06,
      "loss": 0.89425397,
      "num_input_tokens_seen": 60609170,
      "step": 2816,
      "time_per_iteration": 2.5557913780212402
    },
    {
      "auxiliary_loss_clip": 0.01191807,
      "auxiliary_loss_mlp": 0.01031581,
      "balance_loss_clip": 1.05713618,
      "balance_loss_mlp": 1.02314091,
      "epoch": 0.33872422293031923,
      "flos": 18880646146560.0,
      "grad_norm": 1.9449016605541014,
      "language_loss": 0.72560668,
      "learning_rate": 3.0809241672490066e-06,
      "loss": 0.74784064,
      "num_input_tokens_seen": 60627340,
      "step": 2817,
      "time_per_iteration": 2.4725780487060547
    },
    {
      "auxiliary_loss_clip": 0.01164464,
      "auxiliary_loss_mlp": 0.01032304,
      "balance_loss_clip": 1.05596066,
      "balance_loss_mlp": 1.02420366,
      "epoch": 0.33884446582095834,
      "flos": 23146977064320.0,
      "grad_norm": 1.7563921586113573,
      "language_loss": 0.85033894,
      "learning_rate": 3.080268683171753e-06,
      "loss": 0.87230659,
      "num_input_tokens_seen": 60647630,
      "step": 2818,
      "time_per_iteration": 2.5601589679718018
    },
    {
      "auxiliary_loss_clip": 0.01176963,
      "auxiliary_loss_mlp": 0.01030315,
      "balance_loss_clip": 1.05374336,
      "balance_loss_mlp": 1.02178597,
      "epoch": 0.33896470871159745,
      "flos": 15997342544640.0,
      "grad_norm": 2.011754210651904,
      "language_loss": 0.89363098,
      "learning_rate": 3.0796130352185985e-06,
      "loss": 0.91570377,
      "num_input_tokens_seen": 60664485,
      "step": 2819,
      "time_per_iteration": 3.278252601623535
    },
    {
      "auxiliary_loss_clip": 0.011513,
      "auxiliary_loss_mlp": 0.00762806,
      "balance_loss_clip": 1.05013561,
      "balance_loss_mlp": 1.00028777,
      "epoch": 0.3390849516022365,
      "flos": 34495754112000.0,
      "grad_norm": 1.7191188425091566,
      "language_loss": 0.66406536,
      "learning_rate": 3.0789572234890057e-06,
      "loss": 0.68320644,
      "num_input_tokens_seen": 60686125,
      "step": 2820,
      "time_per_iteration": 2.6657886505126953
    },
    {
      "auxiliary_loss_clip": 0.01161132,
      "auxiliary_loss_mlp": 0.01033412,
      "balance_loss_clip": 1.05485559,
      "balance_loss_mlp": 1.02496564,
      "epoch": 0.3392051944928756,
      "flos": 16180307447040.0,
      "grad_norm": 1.5860837649000734,
      "language_loss": 0.77403164,
      "learning_rate": 3.0783012480824596e-06,
      "loss": 0.79597712,
      "num_input_tokens_seen": 60705270,
      "step": 2821,
      "time_per_iteration": 2.512835741043091
    },
    {
      "auxiliary_loss_clip": 0.01191022,
      "auxiliary_loss_mlp": 0.01040228,
      "balance_loss_clip": 1.05701792,
      "balance_loss_mlp": 1.0314095,
      "epoch": 0.33932543738351467,
      "flos": 17086656349440.0,
      "grad_norm": 2.4135441818121617,
      "language_loss": 0.7426461,
      "learning_rate": 3.077645109098471e-06,
      "loss": 0.76495862,
      "num_input_tokens_seen": 60721540,
      "step": 2822,
      "time_per_iteration": 2.4528794288635254
    },
    {
      "auxiliary_loss_clip": 0.01136201,
      "auxiliary_loss_mlp": 0.01030761,
      "balance_loss_clip": 1.05002451,
      "balance_loss_mlp": 1.02241611,
      "epoch": 0.3394456802741538,
      "flos": 22126970551680.0,
      "grad_norm": 1.6545927537226381,
      "language_loss": 0.72225451,
      "learning_rate": 3.076988806636577e-06,
      "loss": 0.74392414,
      "num_input_tokens_seen": 60739300,
      "step": 2823,
      "time_per_iteration": 2.564561605453491
    },
    {
      "auxiliary_loss_clip": 0.01166071,
      "auxiliary_loss_mlp": 0.00762682,
      "balance_loss_clip": 1.05588567,
      "balance_loss_mlp": 1.00028861,
      "epoch": 0.3395659231647929,
      "flos": 25226887121280.0,
      "grad_norm": 1.8257072180149934,
      "language_loss": 0.88669872,
      "learning_rate": 3.0763323407963377e-06,
      "loss": 0.90598625,
      "num_input_tokens_seen": 60758910,
      "step": 2824,
      "time_per_iteration": 3.3646271228790283
    },
    {
      "auxiliary_loss_clip": 0.01175345,
      "auxiliary_loss_mlp": 0.01030599,
      "balance_loss_clip": 1.053478,
      "balance_loss_mlp": 1.02217054,
      "epoch": 0.33968616605543195,
      "flos": 29096477343360.0,
      "grad_norm": 2.198216914701041,
      "language_loss": 0.80240387,
      "learning_rate": 3.075675711677337e-06,
      "loss": 0.82446325,
      "num_input_tokens_seen": 60779005,
      "step": 2825,
      "time_per_iteration": 3.29198956489563
    },
    {
      "auxiliary_loss_clip": 0.01157058,
      "auxiliary_loss_mlp": 0.01036488,
      "balance_loss_clip": 1.05518126,
      "balance_loss_mlp": 1.02806616,
      "epoch": 0.33980640894607106,
      "flos": 21433966479360.0,
      "grad_norm": 1.8007682715217816,
      "language_loss": 0.78273833,
      "learning_rate": 3.0750189193791865e-06,
      "loss": 0.80467379,
      "num_input_tokens_seen": 60798590,
      "step": 2826,
      "time_per_iteration": 2.53039288520813
    },
    {
      "auxiliary_loss_clip": 0.01176278,
      "auxiliary_loss_mlp": 0.01030003,
      "balance_loss_clip": 1.05625188,
      "balance_loss_mlp": 1.02203047,
      "epoch": 0.33992665183671017,
      "flos": 32490035596800.0,
      "grad_norm": 1.7157291951064704,
      "language_loss": 0.70254266,
      "learning_rate": 3.0743619640015203e-06,
      "loss": 0.72460544,
      "num_input_tokens_seen": 60818840,
      "step": 2827,
      "time_per_iteration": 3.4182803630828857
    },
    {
      "auxiliary_loss_clip": 0.01165959,
      "auxiliary_loss_mlp": 0.01037944,
      "balance_loss_clip": 1.05197906,
      "balance_loss_mlp": 1.02927744,
      "epoch": 0.3400468947273492,
      "flos": 17055414495360.0,
      "grad_norm": 1.9955731156893937,
      "language_loss": 0.92947441,
      "learning_rate": 3.073704845643999e-06,
      "loss": 0.95151347,
      "num_input_tokens_seen": 60835965,
      "step": 2828,
      "time_per_iteration": 2.52634596824646
    },
    {
      "auxiliary_loss_clip": 0.01176676,
      "auxiliary_loss_mlp": 0.01040316,
      "balance_loss_clip": 1.05217743,
      "balance_loss_mlp": 1.03164339,
      "epoch": 0.34016713761798834,
      "flos": 16872988296960.0,
      "grad_norm": 3.3029907220170065,
      "language_loss": 0.77702093,
      "learning_rate": 3.0730475644063063e-06,
      "loss": 0.79919082,
      "num_input_tokens_seen": 60851065,
      "step": 2829,
      "time_per_iteration": 2.4972875118255615
    },
    {
      "auxiliary_loss_clip": 0.01152056,
      "auxiliary_loss_mlp": 0.00761674,
      "balance_loss_clip": 1.0493679,
      "balance_loss_mlp": 1.00028718,
      "epoch": 0.34028738050862745,
      "flos": 21907161273600.0,
      "grad_norm": 1.86100310974661,
      "language_loss": 0.6520046,
      "learning_rate": 3.072390120388151e-06,
      "loss": 0.67114186,
      "num_input_tokens_seen": 60869390,
      "step": 2830,
      "time_per_iteration": 2.5432450771331787
    },
    {
      "auxiliary_loss_clip": 0.01177649,
      "auxiliary_loss_mlp": 0.01036666,
      "balance_loss_clip": 1.05566072,
      "balance_loss_mlp": 1.02739763,
      "epoch": 0.3404076233992665,
      "flos": 22746034477440.0,
      "grad_norm": 2.0940198208574152,
      "language_loss": 0.71006203,
      "learning_rate": 3.071732513689267e-06,
      "loss": 0.73220527,
      "num_input_tokens_seen": 60887925,
      "step": 2831,
      "time_per_iteration": 2.5226480960845947
    },
    {
      "auxiliary_loss_clip": 0.01177813,
      "auxiliary_loss_mlp": 0.01034603,
      "balance_loss_clip": 1.05777347,
      "balance_loss_mlp": 1.02562642,
      "epoch": 0.3405278662899056,
      "flos": 17052361839360.0,
      "grad_norm": 2.3294458236542357,
      "language_loss": 0.67156649,
      "learning_rate": 3.0710747444094134e-06,
      "loss": 0.69369066,
      "num_input_tokens_seen": 60905955,
      "step": 2832,
      "time_per_iteration": 2.498551607131958
    },
    {
      "auxiliary_loss_clip": 0.01166697,
      "auxiliary_loss_mlp": 0.01030942,
      "balance_loss_clip": 1.05585432,
      "balance_loss_mlp": 1.02193558,
      "epoch": 0.3406481091805447,
      "flos": 42813131783040.0,
      "grad_norm": 2.0546305736164494,
      "language_loss": 0.64696074,
      "learning_rate": 3.070416812648372e-06,
      "loss": 0.66893715,
      "num_input_tokens_seen": 60929405,
      "step": 2833,
      "time_per_iteration": 2.7838943004608154
    },
    {
      "auxiliary_loss_clip": 0.01144971,
      "auxiliary_loss_mlp": 0.01030355,
      "balance_loss_clip": 1.0482626,
      "balance_loss_mlp": 1.02218914,
      "epoch": 0.3407683520711838,
      "flos": 26761457917440.0,
      "grad_norm": 2.2514700970999577,
      "language_loss": 0.64593649,
      "learning_rate": 3.069758718505951e-06,
      "loss": 0.66768974,
      "num_input_tokens_seen": 60951145,
      "step": 2834,
      "time_per_iteration": 2.6428771018981934
    },
    {
      "auxiliary_loss_clip": 0.01190717,
      "auxiliary_loss_mlp": 0.01038,
      "balance_loss_clip": 1.05737376,
      "balance_loss_mlp": 1.02921391,
      "epoch": 0.3408885949618229,
      "flos": 28767643309440.0,
      "grad_norm": 1.602862020007956,
      "language_loss": 0.80054921,
      "learning_rate": 3.0691004620819836e-06,
      "loss": 0.8228364,
      "num_input_tokens_seen": 60971275,
      "step": 2835,
      "time_per_iteration": 2.530599355697632
    },
    {
      "auxiliary_loss_clip": 0.0104268,
      "auxiliary_loss_mlp": 0.01003975,
      "balance_loss_clip": 1.02423918,
      "balance_loss_mlp": 1.00251508,
      "epoch": 0.341008837852462,
      "flos": 63576252881280.0,
      "grad_norm": 0.802397151761232,
      "language_loss": 0.60220003,
      "learning_rate": 3.0684420434763254e-06,
      "loss": 0.6226666,
      "num_input_tokens_seen": 61037460,
      "step": 2836,
      "time_per_iteration": 3.2236483097076416
    },
    {
      "auxiliary_loss_clip": 0.01138947,
      "auxiliary_loss_mlp": 0.01028595,
      "balance_loss_clip": 1.05140924,
      "balance_loss_mlp": 1.02095413,
      "epoch": 0.34112908074310105,
      "flos": 20812173120000.0,
      "grad_norm": 1.6965154559731834,
      "language_loss": 0.76980627,
      "learning_rate": 3.06778346278886e-06,
      "loss": 0.79148173,
      "num_input_tokens_seen": 61056295,
      "step": 2837,
      "time_per_iteration": 2.5683131217956543
    },
    {
      "auxiliary_loss_clip": 0.01192904,
      "auxiliary_loss_mlp": 0.01028243,
      "balance_loss_clip": 1.05972683,
      "balance_loss_mlp": 1.01948118,
      "epoch": 0.34124932363374016,
      "flos": 24976446520320.0,
      "grad_norm": 1.6446923067510189,
      "language_loss": 0.78823817,
      "learning_rate": 3.0671247201194906e-06,
      "loss": 0.8104496,
      "num_input_tokens_seen": 61078430,
      "step": 2838,
      "time_per_iteration": 2.5430715084075928
    },
    {
      "auxiliary_loss_clip": 0.01151785,
      "auxiliary_loss_mlp": 0.01033189,
      "balance_loss_clip": 1.05216396,
      "balance_loss_mlp": 1.02413535,
      "epoch": 0.3413695665243792,
      "flos": 28402970480640.0,
      "grad_norm": 2.009135435927844,
      "language_loss": 0.75689852,
      "learning_rate": 3.066465815568151e-06,
      "loss": 0.77874833,
      "num_input_tokens_seen": 61099260,
      "step": 2839,
      "time_per_iteration": 2.6277284622192383
    },
    {
      "auxiliary_loss_clip": 0.01178034,
      "auxiliary_loss_mlp": 0.01032827,
      "balance_loss_clip": 1.05331826,
      "balance_loss_mlp": 1.02374387,
      "epoch": 0.34148980941501833,
      "flos": 25302012416640.0,
      "grad_norm": 1.700372809016299,
      "language_loss": 0.68840528,
      "learning_rate": 3.0658067492347947e-06,
      "loss": 0.71051395,
      "num_input_tokens_seen": 61121900,
      "step": 2840,
      "time_per_iteration": 2.582031726837158
    },
    {
      "auxiliary_loss_clip": 0.011007,
      "auxiliary_loss_mlp": 0.01030934,
      "balance_loss_clip": 1.04895031,
      "balance_loss_mlp": 1.02226198,
      "epoch": 0.34161005230565744,
      "flos": 17530081747200.0,
      "grad_norm": 1.8531225065152177,
      "language_loss": 0.6694631,
      "learning_rate": 3.065147521219402e-06,
      "loss": 0.69077945,
      "num_input_tokens_seen": 61141155,
      "step": 2841,
      "time_per_iteration": 2.6534531116485596
    },
    {
      "auxiliary_loss_clip": 0.01153665,
      "auxiliary_loss_mlp": 0.01030281,
      "balance_loss_clip": 1.05413032,
      "balance_loss_mlp": 1.02260971,
      "epoch": 0.3417302951962965,
      "flos": 43650101566080.0,
      "grad_norm": 1.426722176702241,
      "language_loss": 0.74023664,
      "learning_rate": 3.064488131621977e-06,
      "loss": 0.76207614,
      "num_input_tokens_seen": 61164480,
      "step": 2842,
      "time_per_iteration": 2.7668306827545166
    },
    {
      "auxiliary_loss_clip": 0.01172887,
      "auxiliary_loss_mlp": 0.01037833,
      "balance_loss_clip": 1.05551434,
      "balance_loss_mlp": 1.02982235,
      "epoch": 0.3418505380869356,
      "flos": 30882207012480.0,
      "grad_norm": 2.0514195590819932,
      "language_loss": 0.73701549,
      "learning_rate": 3.063828580542549e-06,
      "loss": 0.75912273,
      "num_input_tokens_seen": 61185675,
      "step": 2843,
      "time_per_iteration": 2.587906837463379
    },
    {
      "auxiliary_loss_clip": 0.01165168,
      "auxiliary_loss_mlp": 0.01032619,
      "balance_loss_clip": 1.05492759,
      "balance_loss_mlp": 1.02457857,
      "epoch": 0.3419707809775747,
      "flos": 19463871277440.0,
      "grad_norm": 1.893643085613405,
      "language_loss": 0.73403263,
      "learning_rate": 3.0631688680811706e-06,
      "loss": 0.75601059,
      "num_input_tokens_seen": 61205300,
      "step": 2844,
      "time_per_iteration": 3.3214094638824463
    },
    {
      "auxiliary_loss_clip": 0.0119249,
      "auxiliary_loss_mlp": 0.01030194,
      "balance_loss_clip": 1.05659175,
      "balance_loss_mlp": 1.02208149,
      "epoch": 0.3420910238682138,
      "flos": 28727818104960.0,
      "grad_norm": 1.8121295838037192,
      "language_loss": 0.75232315,
      "learning_rate": 3.062508994337921e-06,
      "loss": 0.77455002,
      "num_input_tokens_seen": 61224905,
      "step": 2845,
      "time_per_iteration": 2.5559680461883545
    },
    {
      "auxiliary_loss_clip": 0.01175876,
      "auxiliary_loss_mlp": 0.01034293,
      "balance_loss_clip": 1.05421233,
      "balance_loss_mlp": 1.02475059,
      "epoch": 0.3422112667588529,
      "flos": 21397265758080.0,
      "grad_norm": 2.0255203308878897,
      "language_loss": 0.79191744,
      "learning_rate": 3.0618489594129013e-06,
      "loss": 0.81401908,
      "num_input_tokens_seen": 61243045,
      "step": 2846,
      "time_per_iteration": 2.5383517742156982
    },
    {
      "auxiliary_loss_clip": 0.01152829,
      "auxiliary_loss_mlp": 0.01039213,
      "balance_loss_clip": 1.05480444,
      "balance_loss_mlp": 1.03037977,
      "epoch": 0.342331509649492,
      "flos": 13881450038400.0,
      "grad_norm": 1.985554383450499,
      "language_loss": 0.7089622,
      "learning_rate": 3.061188763406239e-06,
      "loss": 0.73088264,
      "num_input_tokens_seen": 61259190,
      "step": 2847,
      "time_per_iteration": 2.5748870372772217
    },
    {
      "auxiliary_loss_clip": 0.01160092,
      "auxiliary_loss_mlp": 0.01034729,
      "balance_loss_clip": 1.05452132,
      "balance_loss_mlp": 1.02525771,
      "epoch": 0.34245175254013105,
      "flos": 28621450955520.0,
      "grad_norm": 2.286083562027519,
      "language_loss": 0.82508421,
      "learning_rate": 3.060528406418085e-06,
      "loss": 0.84703243,
      "num_input_tokens_seen": 61279040,
      "step": 2848,
      "time_per_iteration": 2.622223377227783
    },
    {
      "auxiliary_loss_clip": 0.01155057,
      "auxiliary_loss_mlp": 0.01031394,
      "balance_loss_clip": 1.05324817,
      "balance_loss_mlp": 1.02378893,
      "epoch": 0.34257199543077016,
      "flos": 34127058960000.0,
      "grad_norm": 1.9795640356133584,
      "language_loss": 0.61918843,
      "learning_rate": 3.0598678885486145e-06,
      "loss": 0.6410529,
      "num_input_tokens_seen": 61301580,
      "step": 2849,
      "time_per_iteration": 2.708325147628784
    },
    {
      "auxiliary_loss_clip": 0.0114752,
      "auxiliary_loss_mlp": 0.00761769,
      "balance_loss_clip": 1.04964995,
      "balance_loss_mlp": 1.00029707,
      "epoch": 0.34269223832140927,
      "flos": 19974018188160.0,
      "grad_norm": 1.6692644985382394,
      "language_loss": 0.74580455,
      "learning_rate": 3.0592072098980282e-06,
      "loss": 0.76489741,
      "num_input_tokens_seen": 61321240,
      "step": 2850,
      "time_per_iteration": 3.3515524864196777
    },
    {
      "auxiliary_loss_clip": 0.01155259,
      "auxiliary_loss_mlp": 0.01036409,
      "balance_loss_clip": 1.05211473,
      "balance_loss_mlp": 1.02810073,
      "epoch": 0.3428124812120483,
      "flos": 27235658292480.0,
      "grad_norm": 1.7357771225804999,
      "language_loss": 0.72894919,
      "learning_rate": 3.0585463705665514e-06,
      "loss": 0.75086594,
      "num_input_tokens_seen": 61341615,
      "step": 2851,
      "time_per_iteration": 3.339571952819824
    },
    {
      "auxiliary_loss_clip": 0.01146664,
      "auxiliary_loss_mlp": 0.0103402,
      "balance_loss_clip": 1.0509665,
      "balance_loss_mlp": 1.02541327,
      "epoch": 0.34293272410268744,
      "flos": 24570871079040.0,
      "grad_norm": 2.3144846115702187,
      "language_loss": 0.70466191,
      "learning_rate": 3.0578853706544304e-06,
      "loss": 0.72646868,
      "num_input_tokens_seen": 61359005,
      "step": 2852,
      "time_per_iteration": 2.6213293075561523
    },
    {
      "auxiliary_loss_clip": 0.01151094,
      "auxiliary_loss_mlp": 0.00762408,
      "balance_loss_clip": 1.05276978,
      "balance_loss_mlp": 1.00028503,
      "epoch": 0.34305296699332655,
      "flos": 21506865131520.0,
      "grad_norm": 2.079107564018096,
      "language_loss": 0.65206051,
      "learning_rate": 3.0572242102619404e-06,
      "loss": 0.67119551,
      "num_input_tokens_seen": 61376160,
      "step": 2853,
      "time_per_iteration": 3.4474191665649414
    },
    {
      "auxiliary_loss_clip": 0.01164893,
      "auxiliary_loss_mlp": 0.01037592,
      "balance_loss_clip": 1.05714607,
      "balance_loss_mlp": 1.02923536,
      "epoch": 0.3431732098839656,
      "flos": 24056665931520.0,
      "grad_norm": 3.8136508447317925,
      "language_loss": 0.80387664,
      "learning_rate": 3.0565628894893784e-06,
      "loss": 0.82590151,
      "num_input_tokens_seen": 61396795,
      "step": 2854,
      "time_per_iteration": 2.615370273590088
    },
    {
      "auxiliary_loss_clip": 0.01169414,
      "auxiliary_loss_mlp": 0.01039949,
      "balance_loss_clip": 1.05549955,
      "balance_loss_mlp": 1.03111589,
      "epoch": 0.3432934527746047,
      "flos": 16800879744000.0,
      "grad_norm": 2.157918941321947,
      "language_loss": 0.74816394,
      "learning_rate": 3.0559014084370655e-06,
      "loss": 0.77025759,
      "num_input_tokens_seen": 61415320,
      "step": 2855,
      "time_per_iteration": 2.5045769214630127
    },
    {
      "auxiliary_loss_clip": 0.0116914,
      "auxiliary_loss_mlp": 0.0103999,
      "balance_loss_clip": 1.05488348,
      "balance_loss_mlp": 1.03076303,
      "epoch": 0.34341369566524377,
      "flos": 23439720908160.0,
      "grad_norm": 1.6457957070063196,
      "language_loss": 0.78553891,
      "learning_rate": 3.055239767205349e-06,
      "loss": 0.80763018,
      "num_input_tokens_seen": 61437070,
      "step": 2856,
      "time_per_iteration": 2.608302354812622
    },
    {
      "auxiliary_loss_clip": 0.01180788,
      "auxiliary_loss_mlp": 0.01034974,
      "balance_loss_clip": 1.06112814,
      "balance_loss_mlp": 1.02634954,
      "epoch": 0.3435339385558829,
      "flos": 17267466435840.0,
      "grad_norm": 1.9175001253014992,
      "language_loss": 0.7821793,
      "learning_rate": 3.054577965894599e-06,
      "loss": 0.80433691,
      "num_input_tokens_seen": 61453215,
      "step": 2857,
      "time_per_iteration": 2.4958202838897705
    },
    {
      "auxiliary_loss_clip": 0.01172673,
      "auxiliary_loss_mlp": 0.01040618,
      "balance_loss_clip": 1.05834937,
      "balance_loss_mlp": 1.0313344,
      "epoch": 0.343654181446522,
      "flos": 22199366413440.0,
      "grad_norm": 2.607507147438953,
      "language_loss": 0.70414674,
      "learning_rate": 3.0539160046052094e-06,
      "loss": 0.72627962,
      "num_input_tokens_seen": 61472915,
      "step": 2858,
      "time_per_iteration": 2.592711925506592
    },
    {
      "auxiliary_loss_clip": 0.01154559,
      "auxiliary_loss_mlp": 0.0103763,
      "balance_loss_clip": 1.05229402,
      "balance_loss_mlp": 1.02805746,
      "epoch": 0.34377442433716104,
      "flos": 19901801894400.0,
      "grad_norm": 2.1551212831913644,
      "language_loss": 0.7042076,
      "learning_rate": 3.0532538834376003e-06,
      "loss": 0.72612953,
      "num_input_tokens_seen": 61492475,
      "step": 2859,
      "time_per_iteration": 2.543410301208496
    },
    {
      "auxiliary_loss_clip": 0.01184308,
      "auxiliary_loss_mlp": 0.01032663,
      "balance_loss_clip": 1.05665779,
      "balance_loss_mlp": 1.02385008,
      "epoch": 0.34389466722780015,
      "flos": 22197678474240.0,
      "grad_norm": 2.0530482106261903,
      "language_loss": 0.78173244,
      "learning_rate": 3.0525916024922143e-06,
      "loss": 0.80390209,
      "num_input_tokens_seen": 61511660,
      "step": 2860,
      "time_per_iteration": 2.5327742099761963
    },
    {
      "auxiliary_loss_clip": 0.01160181,
      "auxiliary_loss_mlp": 0.01031739,
      "balance_loss_clip": 1.0522635,
      "balance_loss_mlp": 1.02348423,
      "epoch": 0.34401491011843927,
      "flos": 18624567110400.0,
      "grad_norm": 2.74416779527112,
      "language_loss": 0.84634113,
      "learning_rate": 3.0519291618695193e-06,
      "loss": 0.86826032,
      "num_input_tokens_seen": 61529060,
      "step": 2861,
      "time_per_iteration": 2.512873411178589
    },
    {
      "auxiliary_loss_clip": 0.01135972,
      "auxiliary_loss_mlp": 0.01033796,
      "balance_loss_clip": 1.04670668,
      "balance_loss_mlp": 1.02531433,
      "epoch": 0.3441351530090783,
      "flos": 17858197509120.0,
      "grad_norm": 1.605653584109235,
      "language_loss": 0.75582016,
      "learning_rate": 3.0512665616700065e-06,
      "loss": 0.7775178,
      "num_input_tokens_seen": 61548125,
      "step": 2862,
      "time_per_iteration": 2.5660040378570557
    },
    {
      "auxiliary_loss_clip": 0.01124829,
      "auxiliary_loss_mlp": 0.01040939,
      "balance_loss_clip": 1.04654288,
      "balance_loss_mlp": 1.03178382,
      "epoch": 0.34425539589971743,
      "flos": 23112754381440.0,
      "grad_norm": 2.208463504641682,
      "language_loss": 0.88893104,
      "learning_rate": 3.0506038019941933e-06,
      "loss": 0.91058868,
      "num_input_tokens_seen": 61568135,
      "step": 2863,
      "time_per_iteration": 2.6123194694519043
    },
    {
      "auxiliary_loss_clip": 0.01152503,
      "auxiliary_loss_mlp": 0.01044163,
      "balance_loss_clip": 1.05508792,
      "balance_loss_mlp": 1.03495371,
      "epoch": 0.34437563879035654,
      "flos": 21907699977600.0,
      "grad_norm": 2.4922091281140397,
      "language_loss": 0.67449784,
      "learning_rate": 3.049940882942617e-06,
      "loss": 0.69646448,
      "num_input_tokens_seen": 61586920,
      "step": 2864,
      "time_per_iteration": 2.6509993076324463
    },
    {
      "auxiliary_loss_clip": 0.01191668,
      "auxiliary_loss_mlp": 0.01032557,
      "balance_loss_clip": 1.05774689,
      "balance_loss_mlp": 1.02364647,
      "epoch": 0.3444958816809956,
      "flos": 23076915586560.0,
      "grad_norm": 2.2966588028618347,
      "language_loss": 0.80542159,
      "learning_rate": 3.0492778046158448e-06,
      "loss": 0.82766384,
      "num_input_tokens_seen": 61608340,
      "step": 2865,
      "time_per_iteration": 2.5026814937591553
    },
    {
      "auxiliary_loss_clip": 0.0117909,
      "auxiliary_loss_mlp": 0.01036214,
      "balance_loss_clip": 1.05948448,
      "balance_loss_mlp": 1.02755952,
      "epoch": 0.3446161245716347,
      "flos": 21908633731200.0,
      "grad_norm": 2.7434743514803266,
      "language_loss": 0.7633732,
      "learning_rate": 3.0486145671144633e-06,
      "loss": 0.78552628,
      "num_input_tokens_seen": 61628130,
      "step": 2866,
      "time_per_iteration": 2.5402798652648926
    },
    {
      "auxiliary_loss_clip": 0.01100667,
      "auxiliary_loss_mlp": 0.01038936,
      "balance_loss_clip": 1.04558408,
      "balance_loss_mlp": 1.0299474,
      "epoch": 0.3447363674622738,
      "flos": 25112834461440.0,
      "grad_norm": 2.285007974990173,
      "language_loss": 0.77324098,
      "learning_rate": 3.047951170539086e-06,
      "loss": 0.79463702,
      "num_input_tokens_seen": 61647755,
      "step": 2867,
      "time_per_iteration": 2.6735188961029053
    },
    {
      "auxiliary_loss_clip": 0.01146246,
      "auxiliary_loss_mlp": 0.01032734,
      "balance_loss_clip": 1.05466199,
      "balance_loss_mlp": 1.02493763,
      "epoch": 0.3448566103529129,
      "flos": 11984684451840.0,
      "grad_norm": 1.8631617225080825,
      "language_loss": 0.84134984,
      "learning_rate": 3.047287614990349e-06,
      "loss": 0.86313963,
      "num_input_tokens_seen": 61665675,
      "step": 2868,
      "time_per_iteration": 2.598815441131592
    },
    {
      "auxiliary_loss_clip": 0.0115339,
      "auxiliary_loss_mlp": 0.01039917,
      "balance_loss_clip": 1.05296254,
      "balance_loss_mlp": 1.03088093,
      "epoch": 0.344976853243552,
      "flos": 40187882465280.0,
      "grad_norm": 2.942548438883092,
      "language_loss": 0.61826527,
      "learning_rate": 3.046623900568914e-06,
      "loss": 0.64019835,
      "num_input_tokens_seen": 61688240,
      "step": 2869,
      "time_per_iteration": 2.720792531967163
    },
    {
      "auxiliary_loss_clip": 0.01157242,
      "auxiliary_loss_mlp": 0.0104119,
      "balance_loss_clip": 1.05133724,
      "balance_loss_mlp": 1.03134918,
      "epoch": 0.34509709613419104,
      "flos": 28723652127360.0,
      "grad_norm": 3.3139639472960547,
      "language_loss": 0.70045179,
      "learning_rate": 3.045960027375465e-06,
      "loss": 0.72243613,
      "num_input_tokens_seen": 61706075,
      "step": 2870,
      "time_per_iteration": 2.6303727626800537
    },
    {
      "auxiliary_loss_clip": 0.0118224,
      "auxiliary_loss_mlp": 0.01035027,
      "balance_loss_clip": 1.05481529,
      "balance_loss_mlp": 1.02666473,
      "epoch": 0.34521733902483015,
      "flos": 29967597982080.0,
      "grad_norm": 3.8583288313236626,
      "language_loss": 0.82627445,
      "learning_rate": 3.045295995510711e-06,
      "loss": 0.84844708,
      "num_input_tokens_seen": 61723045,
      "step": 2871,
      "time_per_iteration": 3.4023141860961914
    },
    {
      "auxiliary_loss_clip": 0.01159703,
      "auxiliary_loss_mlp": 0.01026248,
      "balance_loss_clip": 1.05464697,
      "balance_loss_mlp": 1.01860034,
      "epoch": 0.34533758191546926,
      "flos": 27923059843200.0,
      "grad_norm": 1.817959585658311,
      "language_loss": 0.73653615,
      "learning_rate": 3.0446318050753865e-06,
      "loss": 0.75839567,
      "num_input_tokens_seen": 61743525,
      "step": 2872,
      "time_per_iteration": 2.61146879196167
    },
    {
      "auxiliary_loss_clip": 0.01169738,
      "auxiliary_loss_mlp": 0.01033053,
      "balance_loss_clip": 1.05406547,
      "balance_loss_mlp": 1.02536976,
      "epoch": 0.3454578248061083,
      "flos": 27125879351040.0,
      "grad_norm": 2.2719004137655014,
      "language_loss": 0.77598071,
      "learning_rate": 3.0439674561702474e-06,
      "loss": 0.79800856,
      "num_input_tokens_seen": 61763025,
      "step": 2873,
      "time_per_iteration": 2.573303699493408
    },
    {
      "auxiliary_loss_clip": 0.01177314,
      "auxiliary_loss_mlp": 0.01036719,
      "balance_loss_clip": 1.05665421,
      "balance_loss_mlp": 1.02827883,
      "epoch": 0.3455780676967474,
      "flos": 19024899166080.0,
      "grad_norm": 2.222576806120435,
      "language_loss": 0.87929481,
      "learning_rate": 3.043302948896076e-06,
      "loss": 0.90143514,
      "num_input_tokens_seen": 61781630,
      "step": 2874,
      "time_per_iteration": 2.496525287628174
    },
    {
      "auxiliary_loss_clip": 0.01124002,
      "auxiliary_loss_mlp": 0.01037269,
      "balance_loss_clip": 1.050933,
      "balance_loss_mlp": 1.02807212,
      "epoch": 0.34569831058738654,
      "flos": 34496005507200.0,
      "grad_norm": 2.0787179683340393,
      "language_loss": 0.6029523,
      "learning_rate": 3.0426382833536756e-06,
      "loss": 0.62456501,
      "num_input_tokens_seen": 61804985,
      "step": 2875,
      "time_per_iteration": 2.721073627471924
    },
    {
      "auxiliary_loss_clip": 0.01141009,
      "auxiliary_loss_mlp": 0.01032114,
      "balance_loss_clip": 1.04827654,
      "balance_loss_mlp": 1.02369213,
      "epoch": 0.3458185534780256,
      "flos": 31138681098240.0,
      "grad_norm": 1.9925954619707191,
      "language_loss": 0.77910405,
      "learning_rate": 3.041973459643877e-06,
      "loss": 0.80083525,
      "num_input_tokens_seen": 61824440,
      "step": 2876,
      "time_per_iteration": 3.4137444496154785
    },
    {
      "auxiliary_loss_clip": 0.01125096,
      "auxiliary_loss_mlp": 0.01028455,
      "balance_loss_clip": 1.04619622,
      "balance_loss_mlp": 1.02003241,
      "epoch": 0.3459387963686647,
      "flos": 32452508862720.0,
      "grad_norm": 2.1972526276650415,
      "language_loss": 0.67016023,
      "learning_rate": 3.0413084778675334e-06,
      "loss": 0.69169575,
      "num_input_tokens_seen": 61845690,
      "step": 2877,
      "time_per_iteration": 3.3869800567626953
    },
    {
      "auxiliary_loss_clip": 0.01152927,
      "auxiliary_loss_mlp": 0.00761456,
      "balance_loss_clip": 1.04979467,
      "balance_loss_mlp": 1.00029004,
      "epoch": 0.3460590392593038,
      "flos": 24675658030080.0,
      "grad_norm": 3.85020938508407,
      "language_loss": 0.83856714,
      "learning_rate": 3.0406433381255214e-06,
      "loss": 0.85771102,
      "num_input_tokens_seen": 61863725,
      "step": 2878,
      "time_per_iteration": 2.6087992191314697
    },
    {
      "auxiliary_loss_clip": 0.01176354,
      "auxiliary_loss_mlp": 0.01033122,
      "balance_loss_clip": 1.05801678,
      "balance_loss_mlp": 1.02454448,
      "epoch": 0.34617928214994287,
      "flos": 18807316531200.0,
      "grad_norm": 2.2854980551959168,
      "language_loss": 0.82412827,
      "learning_rate": 3.0399780405187425e-06,
      "loss": 0.846223,
      "num_input_tokens_seen": 61882720,
      "step": 2879,
      "time_per_iteration": 3.3461132049560547
    },
    {
      "auxiliary_loss_clip": 0.01170945,
      "auxiliary_loss_mlp": 0.01036051,
      "balance_loss_clip": 1.05232525,
      "balance_loss_mlp": 1.0279572,
      "epoch": 0.346299525040582,
      "flos": 24857653265280.0,
      "grad_norm": 1.9681873582175926,
      "language_loss": 0.78590351,
      "learning_rate": 3.0393125851481216e-06,
      "loss": 0.8079735,
      "num_input_tokens_seen": 61902595,
      "step": 2880,
      "time_per_iteration": 2.5765316486358643
    },
    {
      "auxiliary_loss_clip": 0.01143471,
      "auxiliary_loss_mlp": 0.01029632,
      "balance_loss_clip": 1.05329752,
      "balance_loss_mlp": 1.02148092,
      "epoch": 0.3464197679312211,
      "flos": 16434914025600.0,
      "grad_norm": 2.3105710559914803,
      "language_loss": 0.86268967,
      "learning_rate": 3.038646972114608e-06,
      "loss": 0.88442069,
      "num_input_tokens_seen": 61918920,
      "step": 2881,
      "time_per_iteration": 2.541077136993408
    },
    {
      "auxiliary_loss_clip": 0.01149061,
      "auxiliary_loss_mlp": 0.01039922,
      "balance_loss_clip": 1.05784321,
      "balance_loss_mlp": 1.03132725,
      "epoch": 0.34654001082186014,
      "flos": 22382474970240.0,
      "grad_norm": 2.492189286843929,
      "language_loss": 0.67583495,
      "learning_rate": 3.037981201519174e-06,
      "loss": 0.69772476,
      "num_input_tokens_seen": 61939520,
      "step": 2882,
      "time_per_iteration": 2.646679401397705
    },
    {
      "auxiliary_loss_clip": 0.01177203,
      "auxiliary_loss_mlp": 0.01028431,
      "balance_loss_clip": 1.05959392,
      "balance_loss_mlp": 1.0204916,
      "epoch": 0.34666025371249926,
      "flos": 19573901614080.0,
      "grad_norm": 1.99103174915717,
      "language_loss": 0.71375644,
      "learning_rate": 3.0373152734628175e-06,
      "loss": 0.73581278,
      "num_input_tokens_seen": 61957800,
      "step": 2883,
      "time_per_iteration": 2.5149483680725098
    },
    {
      "auxiliary_loss_clip": 0.01169244,
      "auxiliary_loss_mlp": 0.01033179,
      "balance_loss_clip": 1.05266118,
      "balance_loss_mlp": 1.02457762,
      "epoch": 0.34678049660313837,
      "flos": 15267637751040.0,
      "grad_norm": 2.0146061392046097,
      "language_loss": 0.76009762,
      "learning_rate": 3.0366491880465584e-06,
      "loss": 0.7821219,
      "num_input_tokens_seen": 61975820,
      "step": 2884,
      "time_per_iteration": 2.53934645652771
    },
    {
      "auxiliary_loss_clip": 0.01193781,
      "auxiliary_loss_mlp": 0.01045227,
      "balance_loss_clip": 1.05823076,
      "balance_loss_mlp": 1.03692985,
      "epoch": 0.3469007394937774,
      "flos": 21181550630400.0,
      "grad_norm": 1.4890128602858559,
      "language_loss": 0.81952977,
      "learning_rate": 3.035982945371443e-06,
      "loss": 0.8419199,
      "num_input_tokens_seen": 61997515,
      "step": 2885,
      "time_per_iteration": 2.516177177429199
    },
    {
      "auxiliary_loss_clip": 0.01172027,
      "auxiliary_loss_mlp": 0.01036625,
      "balance_loss_clip": 1.05747581,
      "balance_loss_mlp": 1.0285604,
      "epoch": 0.34702098238441653,
      "flos": 22375471818240.0,
      "grad_norm": 2.117832509133662,
      "language_loss": 0.85273528,
      "learning_rate": 3.035316545538537e-06,
      "loss": 0.8748219,
      "num_input_tokens_seen": 62016310,
      "step": 2886,
      "time_per_iteration": 2.5861568450927734
    },
    {
      "auxiliary_loss_clip": 0.0116115,
      "auxiliary_loss_mlp": 0.0103564,
      "balance_loss_clip": 1.05753851,
      "balance_loss_mlp": 1.02778959,
      "epoch": 0.3471412252750556,
      "flos": 22929430343040.0,
      "grad_norm": 3.5052503795249366,
      "language_loss": 0.79116368,
      "learning_rate": 3.034649988648935e-06,
      "loss": 0.81313157,
      "num_input_tokens_seen": 62036075,
      "step": 2887,
      "time_per_iteration": 2.5679469108581543
    },
    {
      "auxiliary_loss_clip": 0.01163963,
      "auxiliary_loss_mlp": 0.01036505,
      "balance_loss_clip": 1.05419993,
      "balance_loss_mlp": 1.02849448,
      "epoch": 0.3472614681656947,
      "flos": 21324259365120.0,
      "grad_norm": 1.637903499431211,
      "language_loss": 0.80743718,
      "learning_rate": 3.033983274803752e-06,
      "loss": 0.82944185,
      "num_input_tokens_seen": 62055865,
      "step": 2888,
      "time_per_iteration": 2.5939276218414307
    },
    {
      "auxiliary_loss_clip": 0.01158088,
      "auxiliary_loss_mlp": 0.01037794,
      "balance_loss_clip": 1.05203962,
      "balance_loss_mlp": 1.02955365,
      "epoch": 0.3473817110563338,
      "flos": 23475739271040.0,
      "grad_norm": 2.182814968388955,
      "language_loss": 0.72399664,
      "learning_rate": 3.0333164041041283e-06,
      "loss": 0.74595547,
      "num_input_tokens_seen": 62072180,
      "step": 2889,
      "time_per_iteration": 2.58298397064209
    },
    {
      "auxiliary_loss_clip": 0.0112093,
      "auxiliary_loss_mlp": 0.01028831,
      "balance_loss_clip": 1.04796851,
      "balance_loss_mlp": 1.02080846,
      "epoch": 0.34750195394697286,
      "flos": 22346025644160.0,
      "grad_norm": 1.733505955159925,
      "language_loss": 0.72082144,
      "learning_rate": 3.032649376651228e-06,
      "loss": 0.74231899,
      "num_input_tokens_seen": 62091600,
      "step": 2890,
      "time_per_iteration": 2.6508595943450928
    },
    {
      "auxiliary_loss_clip": 0.01149119,
      "auxiliary_loss_mlp": 0.01027821,
      "balance_loss_clip": 1.05201578,
      "balance_loss_mlp": 1.01880884,
      "epoch": 0.347622196837612,
      "flos": 29095004885760.0,
      "grad_norm": 1.934230461291507,
      "language_loss": 0.75887275,
      "learning_rate": 3.031982192546238e-06,
      "loss": 0.78064221,
      "num_input_tokens_seen": 62114695,
      "step": 2891,
      "time_per_iteration": 2.6416945457458496
    },
    {
      "auxiliary_loss_clip": 0.01174891,
      "auxiliary_loss_mlp": 0.01031807,
      "balance_loss_clip": 1.05370033,
      "balance_loss_mlp": 1.02365315,
      "epoch": 0.3477424397282511,
      "flos": 22455732758400.0,
      "grad_norm": 2.5063923598455813,
      "language_loss": 0.94305003,
      "learning_rate": 3.0313148518903696e-06,
      "loss": 0.96511704,
      "num_input_tokens_seen": 62134520,
      "step": 2892,
      "time_per_iteration": 2.558743953704834
    },
    {
      "auxiliary_loss_clip": 0.01161799,
      "auxiliary_loss_mlp": 0.01029946,
      "balance_loss_clip": 1.05358434,
      "balance_loss_mlp": 1.02226019,
      "epoch": 0.34786268261889014,
      "flos": 15778790242560.0,
      "grad_norm": 1.9840017781610457,
      "language_loss": 0.81326824,
      "learning_rate": 3.030647354784859e-06,
      "loss": 0.83518577,
      "num_input_tokens_seen": 62151560,
      "step": 2893,
      "time_per_iteration": 2.529383659362793
    },
    {
      "auxiliary_loss_clip": 0.01143491,
      "auxiliary_loss_mlp": 0.01035493,
      "balance_loss_clip": 1.05040693,
      "balance_loss_mlp": 1.0278163,
      "epoch": 0.34798292550952925,
      "flos": 20777627214720.0,
      "grad_norm": 1.7967939834736704,
      "language_loss": 0.77679121,
      "learning_rate": 3.029979701330964e-06,
      "loss": 0.798581,
      "num_input_tokens_seen": 62170985,
      "step": 2894,
      "time_per_iteration": 2.5791220664978027
    },
    {
      "auxiliary_loss_clip": 0.01162493,
      "auxiliary_loss_mlp": 0.01034549,
      "balance_loss_clip": 1.05122352,
      "balance_loss_mlp": 1.02669334,
      "epoch": 0.34810316840016836,
      "flos": 19937820257280.0,
      "grad_norm": 2.389606533739234,
      "language_loss": 0.8031978,
      "learning_rate": 3.029311891629966e-06,
      "loss": 0.82516825,
      "num_input_tokens_seen": 62189440,
      "step": 2895,
      "time_per_iteration": 2.5644376277923584
    },
    {
      "auxiliary_loss_clip": 0.0116181,
      "auxiliary_loss_mlp": 0.01031157,
      "balance_loss_clip": 1.05474138,
      "balance_loss_mlp": 1.02345014,
      "epoch": 0.3482234112908074,
      "flos": 23623296341760.0,
      "grad_norm": 1.7441054471782214,
      "language_loss": 0.74288774,
      "learning_rate": 3.0286439257831744e-06,
      "loss": 0.76481742,
      "num_input_tokens_seen": 62208910,
      "step": 2896,
      "time_per_iteration": 3.414696216583252
    },
    {
      "auxiliary_loss_clip": 0.0119121,
      "auxiliary_loss_mlp": 0.01034536,
      "balance_loss_clip": 1.05538344,
      "balance_loss_mlp": 1.02366972,
      "epoch": 0.3483436541814465,
      "flos": 23986712194560.0,
      "grad_norm": 1.847366152925246,
      "language_loss": 0.7148751,
      "learning_rate": 3.0279758038919156e-06,
      "loss": 0.73713255,
      "num_input_tokens_seen": 62227135,
      "step": 2897,
      "time_per_iteration": 2.5234873294830322
    },
    {
      "auxiliary_loss_clip": 0.01176337,
      "auxiliary_loss_mlp": 0.01031439,
      "balance_loss_clip": 1.05487514,
      "balance_loss_mlp": 1.02268898,
      "epoch": 0.34846389707208564,
      "flos": 22638338524800.0,
      "grad_norm": 2.124292321172943,
      "language_loss": 0.78314304,
      "learning_rate": 3.0273075260575455e-06,
      "loss": 0.80522078,
      "num_input_tokens_seen": 62246035,
      "step": 2898,
      "time_per_iteration": 2.5328586101531982
    },
    {
      "auxiliary_loss_clip": 0.01163919,
      "auxiliary_loss_mlp": 0.01034993,
      "balance_loss_clip": 1.05322409,
      "balance_loss_mlp": 1.02617764,
      "epoch": 0.3485841399627247,
      "flos": 21792857218560.0,
      "grad_norm": 1.8180860629297626,
      "language_loss": 0.80721939,
      "learning_rate": 3.0266390923814396e-06,
      "loss": 0.82920849,
      "num_input_tokens_seen": 62264095,
      "step": 2899,
      "time_per_iteration": 2.571446657180786
    },
    {
      "auxiliary_loss_clip": 0.01163365,
      "auxiliary_loss_mlp": 0.01034264,
      "balance_loss_clip": 1.05548978,
      "balance_loss_mlp": 1.02611649,
      "epoch": 0.3487043828533638,
      "flos": 17019036996480.0,
      "grad_norm": 1.8417094361743132,
      "language_loss": 0.82101685,
      "learning_rate": 3.0259705029650008e-06,
      "loss": 0.84299314,
      "num_input_tokens_seen": 62282025,
      "step": 2900,
      "time_per_iteration": 2.515490770339966
    },
    {
      "auxiliary_loss_clip": 0.01174714,
      "auxiliary_loss_mlp": 0.01024733,
      "balance_loss_clip": 1.05273795,
      "balance_loss_mlp": 1.0166328,
      "epoch": 0.34882462574400286,
      "flos": 22601135013120.0,
      "grad_norm": 1.6746819025162287,
      "language_loss": 0.72999501,
      "learning_rate": 3.025301757909652e-06,
      "loss": 0.75198948,
      "num_input_tokens_seen": 62302220,
      "step": 2901,
      "time_per_iteration": 3.303964853286743
    },
    {
      "auxiliary_loss_clip": 0.01153466,
      "auxiliary_loss_mlp": 0.00762311,
      "balance_loss_clip": 1.05559707,
      "balance_loss_mlp": 1.00019193,
      "epoch": 0.34894486863464197,
      "flos": 29861518141440.0,
      "grad_norm": 2.2037908454618753,
      "language_loss": 0.80728281,
      "learning_rate": 3.024632857316842e-06,
      "loss": 0.82644057,
      "num_input_tokens_seen": 62323535,
      "step": 2902,
      "time_per_iteration": 2.656179189682007
    },
    {
      "auxiliary_loss_clip": 0.0117763,
      "auxiliary_loss_mlp": 0.0103502,
      "balance_loss_clip": 1.05547309,
      "balance_loss_mlp": 1.02607965,
      "epoch": 0.3490651115252811,
      "flos": 22122265870080.0,
      "grad_norm": 1.9503382955658362,
      "language_loss": 0.77595806,
      "learning_rate": 3.0239638012880412e-06,
      "loss": 0.79808462,
      "num_input_tokens_seen": 62343430,
      "step": 2903,
      "time_per_iteration": 3.2871875762939453
    },
    {
      "auxiliary_loss_clip": 0.01124763,
      "auxiliary_loss_mlp": 0.01036805,
      "balance_loss_clip": 1.04801464,
      "balance_loss_mlp": 1.0274353,
      "epoch": 0.34918535441592014,
      "flos": 12676682943360.0,
      "grad_norm": 3.055778145098356,
      "language_loss": 0.81653523,
      "learning_rate": 3.0232945899247466e-06,
      "loss": 0.83815086,
      "num_input_tokens_seen": 62360365,
      "step": 2904,
      "time_per_iteration": 2.5806732177734375
    },
    {
      "auxiliary_loss_clip": 0.01176442,
      "auxiliary_loss_mlp": 0.01033246,
      "balance_loss_clip": 1.053442,
      "balance_loss_mlp": 1.02340484,
      "epoch": 0.34930559730655925,
      "flos": 23185617120000.0,
      "grad_norm": 2.3887718298138165,
      "language_loss": 0.7776227,
      "learning_rate": 3.022625223328476e-06,
      "loss": 0.79971957,
      "num_input_tokens_seen": 62382105,
      "step": 2905,
      "time_per_iteration": 3.392902135848999
    },
    {
      "auxiliary_loss_clip": 0.01180573,
      "auxiliary_loss_mlp": 0.01035077,
      "balance_loss_clip": 1.05525625,
      "balance_loss_mlp": 1.0261898,
      "epoch": 0.34942584019719836,
      "flos": 22855023319680.0,
      "grad_norm": 1.4565634483543557,
      "language_loss": 0.69123,
      "learning_rate": 3.0219557016007723e-06,
      "loss": 0.71338654,
      "num_input_tokens_seen": 62402235,
      "step": 2906,
      "time_per_iteration": 2.5540504455566406
    },
    {
      "auxiliary_loss_clip": 0.01172369,
      "auxiliary_loss_mlp": 0.01035951,
      "balance_loss_clip": 1.05627489,
      "balance_loss_mlp": 1.02775514,
      "epoch": 0.3495460830878374,
      "flos": 24426043441920.0,
      "grad_norm": 1.8971957532594217,
      "language_loss": 0.69731164,
      "learning_rate": 3.021286024843202e-06,
      "loss": 0.71939486,
      "num_input_tokens_seen": 62420430,
      "step": 2907,
      "time_per_iteration": 2.5714049339294434
    },
    {
      "auxiliary_loss_clip": 0.01084242,
      "auxiliary_loss_mlp": 0.01002008,
      "balance_loss_clip": 1.02243543,
      "balance_loss_mlp": 1.00050628,
      "epoch": 0.3496663259784765,
      "flos": 70008749389440.0,
      "grad_norm": 1.0730551392409449,
      "language_loss": 0.64824623,
      "learning_rate": 3.0206161931573526e-06,
      "loss": 0.66910875,
      "num_input_tokens_seen": 62472980,
      "step": 2908,
      "time_per_iteration": 3.0165178775787354
    },
    {
      "auxiliary_loss_clip": 0.01160201,
      "auxiliary_loss_mlp": 0.01027775,
      "balance_loss_clip": 1.05241728,
      "balance_loss_mlp": 1.01940632,
      "epoch": 0.34978656886911563,
      "flos": 28692805322880.0,
      "grad_norm": 1.603293142953614,
      "language_loss": 0.92911679,
      "learning_rate": 3.0199462066448388e-06,
      "loss": 0.95099652,
      "num_input_tokens_seen": 62495175,
      "step": 2909,
      "time_per_iteration": 2.618485689163208
    },
    {
      "auxiliary_loss_clip": 0.01177419,
      "auxiliary_loss_mlp": 0.01034933,
      "balance_loss_clip": 1.05501533,
      "balance_loss_mlp": 1.02638614,
      "epoch": 0.3499068117597547,
      "flos": 21142156389120.0,
      "grad_norm": 1.9057643844446464,
      "language_loss": 0.69073731,
      "learning_rate": 3.019276065407296e-06,
      "loss": 0.71286088,
      "num_input_tokens_seen": 62514295,
      "step": 2910,
      "time_per_iteration": 2.54492449760437
    },
    {
      "auxiliary_loss_clip": 0.01135377,
      "auxiliary_loss_mlp": 0.01041635,
      "balance_loss_clip": 1.04908967,
      "balance_loss_mlp": 1.03248608,
      "epoch": 0.3500270546503938,
      "flos": 22782699285120.0,
      "grad_norm": 1.7276674538637886,
      "language_loss": 0.80226147,
      "learning_rate": 3.018605769546385e-06,
      "loss": 0.82403159,
      "num_input_tokens_seen": 62534850,
      "step": 2911,
      "time_per_iteration": 2.640115737915039
    },
    {
      "auxiliary_loss_clip": 0.01173031,
      "auxiliary_loss_mlp": 0.01034798,
      "balance_loss_clip": 1.05141878,
      "balance_loss_mlp": 1.0254575,
      "epoch": 0.3501472975410329,
      "flos": 22894058424960.0,
      "grad_norm": 1.838390232102658,
      "language_loss": 0.79569614,
      "learning_rate": 3.017935319163788e-06,
      "loss": 0.81777442,
      "num_input_tokens_seen": 62553810,
      "step": 2912,
      "time_per_iteration": 2.536867618560791
    },
    {
      "auxiliary_loss_clip": 0.01179602,
      "auxiliary_loss_mlp": 0.01034387,
      "balance_loss_clip": 1.05653012,
      "balance_loss_mlp": 1.02474296,
      "epoch": 0.35026754043167196,
      "flos": 25446588658560.0,
      "grad_norm": 1.612060217566858,
      "language_loss": 0.70679367,
      "learning_rate": 3.017264714361213e-06,
      "loss": 0.72893351,
      "num_input_tokens_seen": 62573460,
      "step": 2913,
      "time_per_iteration": 2.5598392486572266
    },
    {
      "auxiliary_loss_clip": 0.01165506,
      "auxiliary_loss_mlp": 0.00762733,
      "balance_loss_clip": 1.05621672,
      "balance_loss_mlp": 1.00020981,
      "epoch": 0.3503877833223111,
      "flos": 19573757959680.0,
      "grad_norm": 1.8788484529520266,
      "language_loss": 0.82255793,
      "learning_rate": 3.016593955240389e-06,
      "loss": 0.84184027,
      "num_input_tokens_seen": 62592150,
      "step": 2914,
      "time_per_iteration": 2.574967861175537
    },
    {
      "auxiliary_loss_clip": 0.01067809,
      "auxiliary_loss_mlp": 0.01003359,
      "balance_loss_clip": 1.01982403,
      "balance_loss_mlp": 1.00179768,
      "epoch": 0.3505080262129502,
      "flos": 65072075880960.0,
      "grad_norm": 0.8755624465262817,
      "language_loss": 0.63733232,
      "learning_rate": 3.015923041903071e-06,
      "loss": 0.6580441,
      "num_input_tokens_seen": 62658275,
      "step": 2915,
      "time_per_iteration": 3.158841848373413
    },
    {
      "auxiliary_loss_clip": 0.01176134,
      "auxiliary_loss_mlp": 0.01035973,
      "balance_loss_clip": 1.05754209,
      "balance_loss_mlp": 1.02739573,
      "epoch": 0.35062826910358924,
      "flos": 29314562768640.0,
      "grad_norm": 2.120033101365607,
      "language_loss": 0.83511508,
      "learning_rate": 3.0152519744510347e-06,
      "loss": 0.85723615,
      "num_input_tokens_seen": 62678075,
      "step": 2916,
      "time_per_iteration": 2.5870041847229004
    },
    {
      "auxiliary_loss_clip": 0.01148275,
      "auxiliary_loss_mlp": 0.01030226,
      "balance_loss_clip": 1.05151772,
      "balance_loss_mlp": 1.0223285,
      "epoch": 0.35074851199422835,
      "flos": 23987717775360.0,
      "grad_norm": 1.7671441554612437,
      "language_loss": 0.82473975,
      "learning_rate": 3.014580752986081e-06,
      "loss": 0.84652478,
      "num_input_tokens_seen": 62696950,
      "step": 2917,
      "time_per_iteration": 2.6295666694641113
    },
    {
      "auxiliary_loss_clip": 0.01139876,
      "auxiliary_loss_mlp": 0.0103365,
      "balance_loss_clip": 1.05331028,
      "balance_loss_mlp": 1.02477527,
      "epoch": 0.3508687548848674,
      "flos": 15224436668160.0,
      "grad_norm": 1.809485217244025,
      "language_loss": 0.78622371,
      "learning_rate": 3.0139093776100345e-06,
      "loss": 0.80795902,
      "num_input_tokens_seen": 62713540,
      "step": 2918,
      "time_per_iteration": 2.573765754699707
    },
    {
      "auxiliary_loss_clip": 0.01187152,
      "auxiliary_loss_mlp": 0.01029734,
      "balance_loss_clip": 1.05459833,
      "balance_loss_mlp": 1.02121079,
      "epoch": 0.3509889977755065,
      "flos": 21361750185600.0,
      "grad_norm": 1.7858777474090206,
      "language_loss": 0.75717229,
      "learning_rate": 3.013237848424741e-06,
      "loss": 0.77934122,
      "num_input_tokens_seen": 62732925,
      "step": 2919,
      "time_per_iteration": 2.4969635009765625
    },
    {
      "auxiliary_loss_clip": 0.0116601,
      "auxiliary_loss_mlp": 0.01032034,
      "balance_loss_clip": 1.05565119,
      "balance_loss_mlp": 1.02331424,
      "epoch": 0.35110924066614563,
      "flos": 19135360465920.0,
      "grad_norm": 2.0976379718590157,
      "language_loss": 0.75419998,
      "learning_rate": 3.012566165532072e-06,
      "loss": 0.77618045,
      "num_input_tokens_seen": 62751715,
      "step": 2920,
      "time_per_iteration": 2.5372917652130127
    },
    {
      "auxiliary_loss_clip": 0.01121798,
      "auxiliary_loss_mlp": 0.01037741,
      "balance_loss_clip": 1.04703665,
      "balance_loss_mlp": 1.02909279,
      "epoch": 0.3512294835567847,
      "flos": 21980885938560.0,
      "grad_norm": 2.362417253563913,
      "language_loss": 0.76344681,
      "learning_rate": 3.0118943290339207e-06,
      "loss": 0.78504217,
      "num_input_tokens_seen": 62771925,
      "step": 2921,
      "time_per_iteration": 2.6943583488464355
    },
    {
      "auxiliary_loss_clip": 0.01137411,
      "auxiliary_loss_mlp": 0.01032734,
      "balance_loss_clip": 1.04757547,
      "balance_loss_mlp": 1.02402592,
      "epoch": 0.3513497264474238,
      "flos": 17817294896640.0,
      "grad_norm": 1.8835208345519276,
      "language_loss": 0.68213606,
      "learning_rate": 3.011222339032204e-06,
      "loss": 0.70383757,
      "num_input_tokens_seen": 62790075,
      "step": 2922,
      "time_per_iteration": 2.566432476043701
    },
    {
      "auxiliary_loss_clip": 0.01188784,
      "auxiliary_loss_mlp": 0.01029696,
      "balance_loss_clip": 1.05564117,
      "balance_loss_mlp": 1.02138066,
      "epoch": 0.3514699693380629,
      "flos": 26943417239040.0,
      "grad_norm": 1.7066425958830471,
      "language_loss": 0.69068933,
      "learning_rate": 3.0105501956288626e-06,
      "loss": 0.71287411,
      "num_input_tokens_seen": 62810545,
      "step": 2923,
      "time_per_iteration": 3.3607406616210938
    },
    {
      "auxiliary_loss_clip": 0.01180773,
      "auxiliary_loss_mlp": 0.01036226,
      "balance_loss_clip": 1.05416286,
      "balance_loss_mlp": 1.0273087,
      "epoch": 0.35159021222870196,
      "flos": 15267565923840.0,
      "grad_norm": 2.0277476277628126,
      "language_loss": 0.72733462,
      "learning_rate": 3.0098778989258602e-06,
      "loss": 0.74950463,
      "num_input_tokens_seen": 62829155,
      "step": 2924,
      "time_per_iteration": 2.49186372756958
    },
    {
      "auxiliary_loss_clip": 0.01144304,
      "auxiliary_loss_mlp": 0.01035476,
      "balance_loss_clip": 1.05369949,
      "balance_loss_mlp": 1.02696455,
      "epoch": 0.35171045511934107,
      "flos": 13984154000640.0,
      "grad_norm": 1.889510064805231,
      "language_loss": 0.87913203,
      "learning_rate": 3.009205449025183e-06,
      "loss": 0.90092981,
      "num_input_tokens_seen": 62845350,
      "step": 2925,
      "time_per_iteration": 2.558403730392456
    },
    {
      "auxiliary_loss_clip": 0.01140903,
      "auxiliary_loss_mlp": 0.01034339,
      "balance_loss_clip": 1.04794025,
      "balance_loss_mlp": 1.02525568,
      "epoch": 0.3518306980099802,
      "flos": 14283434119680.0,
      "grad_norm": 1.7915308613918992,
      "language_loss": 0.63177657,
      "learning_rate": 3.008532846028842e-06,
      "loss": 0.65352905,
      "num_input_tokens_seen": 62862110,
      "step": 2926,
      "time_per_iteration": 2.5531232357025146
    },
    {
      "auxiliary_loss_clip": 0.01191542,
      "auxiliary_loss_mlp": 0.01033835,
      "balance_loss_clip": 1.0564661,
      "balance_loss_mlp": 1.02417362,
      "epoch": 0.35195094090061924,
      "flos": 27052872958080.0,
      "grad_norm": 3.409516309934858,
      "language_loss": 0.71690607,
      "learning_rate": 3.0078600900388694e-06,
      "loss": 0.73915988,
      "num_input_tokens_seen": 62882415,
      "step": 2927,
      "time_per_iteration": 3.291128635406494
    },
    {
      "auxiliary_loss_clip": 0.01137617,
      "auxiliary_loss_mlp": 0.01035003,
      "balance_loss_clip": 1.04903567,
      "balance_loss_mlp": 1.02615201,
      "epoch": 0.35207118379125835,
      "flos": 25629266252160.0,
      "grad_norm": 2.01244747493883,
      "language_loss": 0.74473602,
      "learning_rate": 3.007187181157323e-06,
      "loss": 0.76646227,
      "num_input_tokens_seen": 62902425,
      "step": 2928,
      "time_per_iteration": 2.6077446937561035
    },
    {
      "auxiliary_loss_clip": 0.01110074,
      "auxiliary_loss_mlp": 0.01034501,
      "balance_loss_clip": 1.04708838,
      "balance_loss_mlp": 1.02575696,
      "epoch": 0.35219142668189746,
      "flos": 18004713085440.0,
      "grad_norm": 2.312033122612613,
      "language_loss": 0.67987168,
      "learning_rate": 3.006514119486282e-06,
      "loss": 0.70131743,
      "num_input_tokens_seen": 62919255,
      "step": 2929,
      "time_per_iteration": 3.3855643272399902
    },
    {
      "auxiliary_loss_clip": 0.01141566,
      "auxiliary_loss_mlp": 0.01032619,
      "balance_loss_clip": 1.0508604,
      "balance_loss_mlp": 1.02389848,
      "epoch": 0.3523116695725365,
      "flos": 14028109269120.0,
      "grad_norm": 1.7722484288679574,
      "language_loss": 0.6946249,
      "learning_rate": 3.005840905127849e-06,
      "loss": 0.71636677,
      "num_input_tokens_seen": 62936160,
      "step": 2930,
      "time_per_iteration": 2.5466082096099854
    },
    {
      "auxiliary_loss_clip": 0.01193016,
      "auxiliary_loss_mlp": 0.01035216,
      "balance_loss_clip": 1.05859816,
      "balance_loss_mlp": 1.02678156,
      "epoch": 0.3524319124631756,
      "flos": 21433966479360.0,
      "grad_norm": 1.9581525035326448,
      "language_loss": 0.86863154,
      "learning_rate": 3.0051675381841516e-06,
      "loss": 0.89091384,
      "num_input_tokens_seen": 62953470,
      "step": 2931,
      "time_per_iteration": 3.301849603652954
    },
    {
      "auxiliary_loss_clip": 0.01105773,
      "auxiliary_loss_mlp": 0.007629,
      "balance_loss_clip": 1.04721129,
      "balance_loss_mlp": 1.00023627,
      "epoch": 0.3525521553538147,
      "flos": 26322773114880.0,
      "grad_norm": 2.0466272697636994,
      "language_loss": 0.77048445,
      "learning_rate": 3.0044940187573363e-06,
      "loss": 0.78917122,
      "num_input_tokens_seen": 62974480,
      "step": 2932,
      "time_per_iteration": 2.77781343460083
    },
    {
      "auxiliary_loss_clip": 0.01178587,
      "auxiliary_loss_mlp": 0.01033687,
      "balance_loss_clip": 1.05519295,
      "balance_loss_mlp": 1.02531815,
      "epoch": 0.3526723982444538,
      "flos": 21543314457600.0,
      "grad_norm": 2.003518602299335,
      "language_loss": 0.65293252,
      "learning_rate": 3.003820346949578e-06,
      "loss": 0.67505527,
      "num_input_tokens_seen": 62992560,
      "step": 2933,
      "time_per_iteration": 2.5514028072357178
    },
    {
      "auxiliary_loss_clip": 0.0118994,
      "auxiliary_loss_mlp": 0.01029475,
      "balance_loss_clip": 1.05498719,
      "balance_loss_mlp": 1.02052259,
      "epoch": 0.3527926411350929,
      "flos": 23733649900800.0,
      "grad_norm": 4.592078972732562,
      "language_loss": 0.79565358,
      "learning_rate": 3.003146522863071e-06,
      "loss": 0.81784773,
      "num_input_tokens_seen": 63013445,
      "step": 2934,
      "time_per_iteration": 2.521057605743408
    },
    {
      "auxiliary_loss_clip": 0.01158371,
      "auxiliary_loss_mlp": 0.01032998,
      "balance_loss_clip": 1.05329895,
      "balance_loss_mlp": 1.02453458,
      "epoch": 0.35291288402573195,
      "flos": 30445461544320.0,
      "grad_norm": 2.1492957780130686,
      "language_loss": 0.86619318,
      "learning_rate": 3.0024725466000345e-06,
      "loss": 0.88810682,
      "num_input_tokens_seen": 63033400,
      "step": 2935,
      "time_per_iteration": 2.643237829208374
    },
    {
      "auxiliary_loss_clip": 0.01175568,
      "auxiliary_loss_mlp": 0.01030614,
      "balance_loss_clip": 1.0558815,
      "balance_loss_mlp": 1.02198899,
      "epoch": 0.35303312691637107,
      "flos": 23112179763840.0,
      "grad_norm": 1.7831961873947628,
      "language_loss": 0.78956401,
      "learning_rate": 3.0017984182627087e-06,
      "loss": 0.81162584,
      "num_input_tokens_seen": 63052725,
      "step": 2936,
      "time_per_iteration": 2.5603106021881104
    },
    {
      "auxiliary_loss_clip": 0.01145684,
      "auxiliary_loss_mlp": 0.00762016,
      "balance_loss_clip": 1.05025578,
      "balance_loss_mlp": 1.00026357,
      "epoch": 0.3531533698070102,
      "flos": 21835699165440.0,
      "grad_norm": 1.9265162608489756,
      "language_loss": 0.82390761,
      "learning_rate": 3.00112413795336e-06,
      "loss": 0.84298462,
      "num_input_tokens_seen": 63072560,
      "step": 2937,
      "time_per_iteration": 2.642991065979004
    },
    {
      "auxiliary_loss_clip": 0.01157507,
      "auxiliary_loss_mlp": 0.0103255,
      "balance_loss_clip": 1.04713035,
      "balance_loss_mlp": 1.0241282,
      "epoch": 0.35327361269764923,
      "flos": 15778969810560.0,
      "grad_norm": 4.343893384890595,
      "language_loss": 0.79819477,
      "learning_rate": 3.000449705774275e-06,
      "loss": 0.8200953,
      "num_input_tokens_seen": 63090800,
      "step": 2938,
      "time_per_iteration": 2.5782315731048584
    },
    {
      "auxiliary_loss_clip": 0.0117694,
      "auxiliary_loss_mlp": 0.01034729,
      "balance_loss_clip": 1.05578876,
      "balance_loss_mlp": 1.02574611,
      "epoch": 0.35339385558828834,
      "flos": 22090413484800.0,
      "grad_norm": 2.078845268821874,
      "language_loss": 0.7118901,
      "learning_rate": 2.9997751218277654e-06,
      "loss": 0.73400682,
      "num_input_tokens_seen": 63108955,
      "step": 2939,
      "time_per_iteration": 2.5523881912231445
    },
    {
      "auxiliary_loss_clip": 0.01189812,
      "auxiliary_loss_mlp": 0.01040231,
      "balance_loss_clip": 1.05659461,
      "balance_loss_mlp": 1.03117716,
      "epoch": 0.35351409847892745,
      "flos": 24165008328960.0,
      "grad_norm": 2.044665502704415,
      "language_loss": 0.77633983,
      "learning_rate": 2.999100386216166e-06,
      "loss": 0.79864025,
      "num_input_tokens_seen": 63127895,
      "step": 2940,
      "time_per_iteration": 2.5185952186584473
    },
    {
      "auxiliary_loss_clip": 0.01160578,
      "auxiliary_loss_mlp": 0.01037936,
      "balance_loss_clip": 1.05370665,
      "balance_loss_mlp": 1.02943099,
      "epoch": 0.3536343413695665,
      "flos": 27052298340480.0,
      "grad_norm": 1.9533908391308672,
      "language_loss": 0.74176681,
      "learning_rate": 2.998425499041831e-06,
      "loss": 0.76375192,
      "num_input_tokens_seen": 63148410,
      "step": 2941,
      "time_per_iteration": 2.6211040019989014
    },
    {
      "auxiliary_loss_clip": 0.0106924,
      "auxiliary_loss_mlp": 0.01000788,
      "balance_loss_clip": 1.01942158,
      "balance_loss_mlp": 0.99932176,
      "epoch": 0.3537545842602056,
      "flos": 65991066370560.0,
      "grad_norm": 1.2767370655689447,
      "language_loss": 0.6450097,
      "learning_rate": 2.997750460407142e-06,
      "loss": 0.66571003,
      "num_input_tokens_seen": 63209765,
      "step": 2942,
      "time_per_iteration": 3.1571106910705566
    },
    {
      "auxiliary_loss_clip": 0.0115147,
      "auxiliary_loss_mlp": 0.010323,
      "balance_loss_clip": 1.05029416,
      "balance_loss_mlp": 1.02273953,
      "epoch": 0.35387482715084473,
      "flos": 18436897526400.0,
      "grad_norm": 2.8357653571367494,
      "language_loss": 0.7010628,
      "learning_rate": 2.997075270414501e-06,
      "loss": 0.72290045,
      "num_input_tokens_seen": 63226980,
      "step": 2943,
      "time_per_iteration": 2.6431119441986084
    },
    {
      "auxiliary_loss_clip": 0.0105692,
      "auxiliary_loss_mlp": 0.01000227,
      "balance_loss_clip": 1.01843476,
      "balance_loss_mlp": 0.99872512,
      "epoch": 0.3539950700414838,
      "flos": 65588579498880.0,
      "grad_norm": 0.7002394363677609,
      "language_loss": 0.57798016,
      "learning_rate": 2.9963999291663347e-06,
      "loss": 0.59855163,
      "num_input_tokens_seen": 63292760,
      "step": 2944,
      "time_per_iteration": 3.1758339405059814
    },
    {
      "auxiliary_loss_clip": 0.01135786,
      "auxiliary_loss_mlp": 0.01043116,
      "balance_loss_clip": 1.05425501,
      "balance_loss_mlp": 1.03413403,
      "epoch": 0.3541153129321229,
      "flos": 20521655919360.0,
      "grad_norm": 4.088943333125777,
      "language_loss": 0.73735374,
      "learning_rate": 2.9957244367650915e-06,
      "loss": 0.75914276,
      "num_input_tokens_seen": 63309005,
      "step": 2945,
      "time_per_iteration": 2.6610565185546875
    },
    {
      "auxiliary_loss_clip": 0.01127398,
      "auxiliary_loss_mlp": 0.01034888,
      "balance_loss_clip": 1.0508368,
      "balance_loss_mlp": 1.02632904,
      "epoch": 0.354235555822762,
      "flos": 19573578391680.0,
      "grad_norm": 1.7881194737641446,
      "language_loss": 0.83810824,
      "learning_rate": 2.9950487933132425e-06,
      "loss": 0.85973114,
      "num_input_tokens_seen": 63326420,
      "step": 2946,
      "time_per_iteration": 2.6404831409454346
    },
    {
      "auxiliary_loss_clip": 0.0117957,
      "auxiliary_loss_mlp": 0.01032255,
      "balance_loss_clip": 1.05469894,
      "balance_loss_mlp": 1.02360606,
      "epoch": 0.35435579871340106,
      "flos": 20777268078720.0,
      "grad_norm": 2.1695012996894176,
      "language_loss": 0.71374333,
      "learning_rate": 2.994372998913283e-06,
      "loss": 0.73586154,
      "num_input_tokens_seen": 63344925,
      "step": 2947,
      "time_per_iteration": 2.5512475967407227
    },
    {
      "auxiliary_loss_clip": 0.01165056,
      "auxiliary_loss_mlp": 0.01031547,
      "balance_loss_clip": 1.05547619,
      "balance_loss_mlp": 1.0229466,
      "epoch": 0.35447604160404017,
      "flos": 23951807153280.0,
      "grad_norm": 2.2683438768037174,
      "language_loss": 0.62379706,
      "learning_rate": 2.99369705366773e-06,
      "loss": 0.64576316,
      "num_input_tokens_seen": 63365170,
      "step": 2948,
      "time_per_iteration": 2.6156389713287354
    },
    {
      "auxiliary_loss_clip": 0.01159651,
      "auxiliary_loss_mlp": 0.0103517,
      "balance_loss_clip": 1.0554136,
      "balance_loss_mlp": 1.02622294,
      "epoch": 0.3545962844946792,
      "flos": 23435662671360.0,
      "grad_norm": 2.5718085886106468,
      "language_loss": 0.81652641,
      "learning_rate": 2.9930209576791244e-06,
      "loss": 0.83847463,
      "num_input_tokens_seen": 63383645,
      "step": 2949,
      "time_per_iteration": 3.344034433364868
    },
    {
      "auxiliary_loss_clip": 0.01172302,
      "auxiliary_loss_mlp": 0.0102576,
      "balance_loss_clip": 1.05344033,
      "balance_loss_mlp": 1.01729667,
      "epoch": 0.35471652738531834,
      "flos": 22085134185600.0,
      "grad_norm": 1.9243288775532592,
      "language_loss": 0.63841122,
      "learning_rate": 2.9923447110500285e-06,
      "loss": 0.66039187,
      "num_input_tokens_seen": 63402390,
      "step": 2950,
      "time_per_iteration": 2.553407907485962
    },
    {
      "auxiliary_loss_clip": 0.01165561,
      "auxiliary_loss_mlp": 0.01033064,
      "balance_loss_clip": 1.05339217,
      "balance_loss_mlp": 1.02464139,
      "epoch": 0.35483677027595745,
      "flos": 27341881787520.0,
      "grad_norm": 1.9035049127752064,
      "language_loss": 0.75243121,
      "learning_rate": 2.9916683138830295e-06,
      "loss": 0.77441746,
      "num_input_tokens_seen": 63423055,
      "step": 2951,
      "time_per_iteration": 2.6091485023498535
    },
    {
      "auxiliary_loss_clip": 0.01156955,
      "auxiliary_loss_mlp": 0.0103526,
      "balance_loss_clip": 1.05264997,
      "balance_loss_mlp": 1.02661121,
      "epoch": 0.3549570131665965,
      "flos": 13516166678400.0,
      "grad_norm": 2.5647282678311902,
      "language_loss": 0.80928093,
      "learning_rate": 2.9909917662807353e-06,
      "loss": 0.8312031,
      "num_input_tokens_seen": 63440855,
      "step": 2952,
      "time_per_iteration": 2.550844669342041
    },
    {
      "auxiliary_loss_clip": 0.01174445,
      "auxiliary_loss_mlp": 0.0103587,
      "balance_loss_clip": 1.05368924,
      "balance_loss_mlp": 1.02660751,
      "epoch": 0.3550772560572356,
      "flos": 20887549810560.0,
      "grad_norm": 2.2616642409585967,
      "language_loss": 0.6871618,
      "learning_rate": 2.9903150683457783e-06,
      "loss": 0.70926493,
      "num_input_tokens_seen": 63459400,
      "step": 2953,
      "time_per_iteration": 3.2935280799865723
    },
    {
      "auxiliary_loss_clip": 0.01159794,
      "auxiliary_loss_mlp": 0.01034287,
      "balance_loss_clip": 1.05193734,
      "balance_loss_mlp": 1.02556717,
      "epoch": 0.3551974989478747,
      "flos": 20194042947840.0,
      "grad_norm": 2.3100128594749973,
      "language_loss": 0.65141284,
      "learning_rate": 2.9896382201808126e-06,
      "loss": 0.67335367,
      "num_input_tokens_seen": 63476800,
      "step": 2954,
      "time_per_iteration": 2.5502383708953857
    },
    {
      "auxiliary_loss_clip": 0.01189377,
      "auxiliary_loss_mlp": 0.01035686,
      "balance_loss_clip": 1.05516315,
      "balance_loss_mlp": 1.02727556,
      "epoch": 0.3553177418385138,
      "flos": 19828831415040.0,
      "grad_norm": 2.3032517855367116,
      "language_loss": 0.80947065,
      "learning_rate": 2.988961221888516e-06,
      "loss": 0.83172125,
      "num_input_tokens_seen": 63493475,
      "step": 2955,
      "time_per_iteration": 3.2146048545837402
    },
    {
      "auxiliary_loss_clip": 0.01137055,
      "auxiliary_loss_mlp": 0.01032825,
      "balance_loss_clip": 1.04907727,
      "balance_loss_mlp": 1.0244565,
      "epoch": 0.3554379847291529,
      "flos": 14829132516480.0,
      "grad_norm": 2.293735216558372,
      "language_loss": 0.79562759,
      "learning_rate": 2.988284073571589e-06,
      "loss": 0.81732637,
      "num_input_tokens_seen": 63509560,
      "step": 2956,
      "time_per_iteration": 2.5867269039154053
    },
    {
      "auxiliary_loss_clip": 0.01176702,
      "auxiliary_loss_mlp": 0.00762334,
      "balance_loss_clip": 1.05434501,
      "balance_loss_mlp": 1.00024354,
      "epoch": 0.355558227619792,
      "flos": 20485350247680.0,
      "grad_norm": 2.2558053273478493,
      "language_loss": 0.7298156,
      "learning_rate": 2.9876067753327528e-06,
      "loss": 0.74920595,
      "num_input_tokens_seen": 63527290,
      "step": 2957,
      "time_per_iteration": 3.405550241470337
    },
    {
      "auxiliary_loss_clip": 0.01178228,
      "auxiliary_loss_mlp": 0.01036731,
      "balance_loss_clip": 1.05359542,
      "balance_loss_mlp": 1.02793908,
      "epoch": 0.35567847051043106,
      "flos": 37663613256960.0,
      "grad_norm": 1.813804692749209,
      "language_loss": 0.80463111,
      "learning_rate": 2.986929327274754e-06,
      "loss": 0.82678068,
      "num_input_tokens_seen": 63547870,
      "step": 2958,
      "time_per_iteration": 2.671753168106079
    },
    {
      "auxiliary_loss_clip": 0.01177592,
      "auxiliary_loss_mlp": 0.01034769,
      "balance_loss_clip": 1.05780411,
      "balance_loss_mlp": 1.02641261,
      "epoch": 0.35579871340107017,
      "flos": 26943058103040.0,
      "grad_norm": 1.6137583906021165,
      "language_loss": 0.7864399,
      "learning_rate": 2.9862517295003617e-06,
      "loss": 0.80856353,
      "num_input_tokens_seen": 63568285,
      "step": 2959,
      "time_per_iteration": 2.622518539428711
    },
    {
      "auxiliary_loss_clip": 0.01144889,
      "auxiliary_loss_mlp": 0.01031084,
      "balance_loss_clip": 1.04857564,
      "balance_loss_mlp": 1.02237535,
      "epoch": 0.3559189562917093,
      "flos": 28293335193600.0,
      "grad_norm": 1.5835469987429431,
      "language_loss": 0.72495151,
      "learning_rate": 2.9855739821123654e-06,
      "loss": 0.74671113,
      "num_input_tokens_seen": 63589865,
      "step": 2960,
      "time_per_iteration": 2.676173448562622
    },
    {
      "auxiliary_loss_clip": 0.01171813,
      "auxiliary_loss_mlp": 0.01038462,
      "balance_loss_clip": 1.05440366,
      "balance_loss_mlp": 1.03019524,
      "epoch": 0.35603919918234833,
      "flos": 25664063552640.0,
      "grad_norm": 1.7510492202817691,
      "language_loss": 0.817711,
      "learning_rate": 2.98489608521358e-06,
      "loss": 0.83981377,
      "num_input_tokens_seen": 63609805,
      "step": 2961,
      "time_per_iteration": 2.60746431350708
    },
    {
      "auxiliary_loss_clip": 0.0117817,
      "auxiliary_loss_mlp": 0.00761882,
      "balance_loss_clip": 1.0548445,
      "balance_loss_mlp": 1.00024629,
      "epoch": 0.35615944207298744,
      "flos": 23000856537600.0,
      "grad_norm": 2.0942067109982823,
      "language_loss": 0.79591548,
      "learning_rate": 2.9842180389068425e-06,
      "loss": 0.81531596,
      "num_input_tokens_seen": 63627115,
      "step": 2962,
      "time_per_iteration": 2.5739121437072754
    },
    {
      "auxiliary_loss_clip": 0.01048691,
      "auxiliary_loss_mlp": 0.01014327,
      "balance_loss_clip": 1.02378392,
      "balance_loss_mlp": 1.01273,
      "epoch": 0.35627968496362655,
      "flos": 68251283723520.0,
      "grad_norm": 0.8495407590465676,
      "language_loss": 0.59286582,
      "learning_rate": 2.98353984329501e-06,
      "loss": 0.61349607,
      "num_input_tokens_seen": 63691460,
      "step": 2963,
      "time_per_iteration": 3.207019329071045
    },
    {
      "auxiliary_loss_clip": 0.01159567,
      "auxiliary_loss_mlp": 0.0103218,
      "balance_loss_clip": 1.05335808,
      "balance_loss_mlp": 1.02323914,
      "epoch": 0.3563999278542656,
      "flos": 22641714403200.0,
      "grad_norm": 1.7273145152354126,
      "language_loss": 0.70577037,
      "learning_rate": 2.982861498480965e-06,
      "loss": 0.72768784,
      "num_input_tokens_seen": 63713840,
      "step": 2964,
      "time_per_iteration": 2.6194419860839844
    },
    {
      "auxiliary_loss_clip": 0.01141984,
      "auxiliary_loss_mlp": 0.01027007,
      "balance_loss_clip": 1.04895496,
      "balance_loss_mlp": 1.01867461,
      "epoch": 0.3565201707449047,
      "flos": 25952533678080.0,
      "grad_norm": 2.2327427914568596,
      "language_loss": 0.82850099,
      "learning_rate": 2.9821830045676122e-06,
      "loss": 0.85019088,
      "num_input_tokens_seen": 63733540,
      "step": 2965,
      "time_per_iteration": 2.645923614501953
    },
    {
      "auxiliary_loss_clip": 0.01191524,
      "auxiliary_loss_mlp": 0.01032342,
      "balance_loss_clip": 1.05873811,
      "balance_loss_mlp": 1.02355671,
      "epoch": 0.3566404136355438,
      "flos": 28475725478400.0,
      "grad_norm": 1.784800908701464,
      "language_loss": 0.73194659,
      "learning_rate": 2.9815043616578793e-06,
      "loss": 0.75418532,
      "num_input_tokens_seen": 63754335,
      "step": 2966,
      "time_per_iteration": 2.591174840927124
    },
    {
      "auxiliary_loss_clip": 0.01143833,
      "auxiliary_loss_mlp": 0.01031014,
      "balance_loss_clip": 1.04928684,
      "balance_loss_mlp": 1.02238297,
      "epoch": 0.3567606565261829,
      "flos": 38363117690880.0,
      "grad_norm": 2.175309034732082,
      "language_loss": 0.76878512,
      "learning_rate": 2.9808255698547145e-06,
      "loss": 0.79053354,
      "num_input_tokens_seen": 63777135,
      "step": 2967,
      "time_per_iteration": 2.7601420879364014
    },
    {
      "auxiliary_loss_clip": 0.01176242,
      "auxiliary_loss_mlp": 0.01034961,
      "balance_loss_clip": 1.05749846,
      "balance_loss_mlp": 1.02640784,
      "epoch": 0.356880899416822,
      "flos": 21981029592960.0,
      "grad_norm": 2.1436998181284075,
      "language_loss": 0.79607993,
      "learning_rate": 2.9801466292610913e-06,
      "loss": 0.81819201,
      "num_input_tokens_seen": 63797020,
      "step": 2968,
      "time_per_iteration": 2.564422845840454
    },
    {
      "auxiliary_loss_clip": 0.01174084,
      "auxiliary_loss_mlp": 0.01029992,
      "balance_loss_clip": 1.05403316,
      "balance_loss_mlp": 1.02167714,
      "epoch": 0.35700114230746105,
      "flos": 18989132198400.0,
      "grad_norm": 2.2936440843313104,
      "language_loss": 0.80855602,
      "learning_rate": 2.979467539980003e-06,
      "loss": 0.8305968,
      "num_input_tokens_seen": 63813810,
      "step": 2969,
      "time_per_iteration": 2.5264787673950195
    },
    {
      "auxiliary_loss_clip": 0.0117695,
      "auxiliary_loss_mlp": 0.0103043,
      "balance_loss_clip": 1.0564481,
      "balance_loss_mlp": 1.02161467,
      "epoch": 0.35712138519810016,
      "flos": 19756112330880.0,
      "grad_norm": 1.8312404226844985,
      "language_loss": 0.77181017,
      "learning_rate": 2.978788302114468e-06,
      "loss": 0.79388392,
      "num_input_tokens_seen": 63830925,
      "step": 2970,
      "time_per_iteration": 2.568730354309082
    },
    {
      "auxiliary_loss_clip": 0.01173521,
      "auxiliary_loss_mlp": 0.01025746,
      "balance_loss_clip": 1.05625439,
      "balance_loss_mlp": 1.01730025,
      "epoch": 0.35724162808873927,
      "flos": 35183012008320.0,
      "grad_norm": 2.0877345239337695,
      "language_loss": 0.82975709,
      "learning_rate": 2.9781089157675255e-06,
      "loss": 0.85174978,
      "num_input_tokens_seen": 63849385,
      "step": 2971,
      "time_per_iteration": 2.6517412662506104
    },
    {
      "auxiliary_loss_clip": 0.01168861,
      "auxiliary_loss_mlp": 0.01032811,
      "balance_loss_clip": 1.05444157,
      "balance_loss_mlp": 1.02417445,
      "epoch": 0.3573618709793783,
      "flos": 25556726736000.0,
      "grad_norm": 1.5180517220456982,
      "language_loss": 0.88257861,
      "learning_rate": 2.977429381042238e-06,
      "loss": 0.90459526,
      "num_input_tokens_seen": 63870060,
      "step": 2972,
      "time_per_iteration": 2.604482650756836
    },
    {
      "auxiliary_loss_clip": 0.01161949,
      "auxiliary_loss_mlp": 0.01030096,
      "balance_loss_clip": 1.05378127,
      "balance_loss_mlp": 1.02218056,
      "epoch": 0.35748211387001744,
      "flos": 29132352051840.0,
      "grad_norm": 2.0951629530708553,
      "language_loss": 0.88458514,
      "learning_rate": 2.9767496980416913e-06,
      "loss": 0.90650558,
      "num_input_tokens_seen": 63889355,
      "step": 2973,
      "time_per_iteration": 2.612957000732422
    },
    {
      "auxiliary_loss_clip": 0.01153241,
      "auxiliary_loss_mlp": 0.01033927,
      "balance_loss_clip": 1.04934359,
      "balance_loss_mlp": 1.02502751,
      "epoch": 0.35760235676065655,
      "flos": 13954169122560.0,
      "grad_norm": 3.262699544296505,
      "language_loss": 0.81082118,
      "learning_rate": 2.9760698668689914e-06,
      "loss": 0.8326928,
      "num_input_tokens_seen": 63905580,
      "step": 2974,
      "time_per_iteration": 2.5223615169525146
    },
    {
      "auxiliary_loss_clip": 0.01174418,
      "auxiliary_loss_mlp": 0.01030626,
      "balance_loss_clip": 1.05356789,
      "balance_loss_mlp": 1.0221802,
      "epoch": 0.3577225996512956,
      "flos": 44018688977280.0,
      "grad_norm": 3.2470123520484417,
      "language_loss": 0.71305203,
      "learning_rate": 2.975389887627269e-06,
      "loss": 0.73510247,
      "num_input_tokens_seen": 63928180,
      "step": 2975,
      "time_per_iteration": 3.444403648376465
    },
    {
      "auxiliary_loss_clip": 0.01148396,
      "auxiliary_loss_mlp": 0.01037493,
      "balance_loss_clip": 1.05117178,
      "balance_loss_mlp": 1.02978635,
      "epoch": 0.3578428425419347,
      "flos": 17055199013760.0,
      "grad_norm": 2.4954237047389833,
      "language_loss": 0.89835173,
      "learning_rate": 2.9747097604196764e-06,
      "loss": 0.9202106,
      "num_input_tokens_seen": 63944825,
      "step": 2976,
      "time_per_iteration": 2.5825753211975098
    },
    {
      "auxiliary_loss_clip": 0.01036976,
      "auxiliary_loss_mlp": 0.01006896,
      "balance_loss_clip": 1.01872742,
      "balance_loss_mlp": 1.0051434,
      "epoch": 0.3579630854325738,
      "flos": 71676550707840.0,
      "grad_norm": 0.673228561617856,
      "language_loss": 0.56643474,
      "learning_rate": 2.9740294853493875e-06,
      "loss": 0.58687347,
      "num_input_tokens_seen": 64016385,
      "step": 2977,
      "time_per_iteration": 3.401665687561035
    },
    {
      "auxiliary_loss_clip": 0.01137438,
      "auxiliary_loss_mlp": 0.01032733,
      "balance_loss_clip": 1.05121922,
      "balance_loss_mlp": 1.02446568,
      "epoch": 0.3580833283232129,
      "flos": 25046651652480.0,
      "grad_norm": 1.859368263692074,
      "language_loss": 0.6712957,
      "learning_rate": 2.9733490625196008e-06,
      "loss": 0.6929974,
      "num_input_tokens_seen": 64036245,
      "step": 2978,
      "time_per_iteration": 2.7114243507385254
    },
    {
      "auxiliary_loss_clip": 0.01132467,
      "auxiliary_loss_mlp": 0.01029218,
      "balance_loss_clip": 1.04806888,
      "balance_loss_mlp": 1.02137947,
      "epoch": 0.358203571213852,
      "flos": 13953127628160.0,
      "grad_norm": 2.688448538533833,
      "language_loss": 0.7522375,
      "learning_rate": 2.9726684920335353e-06,
      "loss": 0.77385437,
      "num_input_tokens_seen": 64054110,
      "step": 2979,
      "time_per_iteration": 3.4021072387695312
    },
    {
      "auxiliary_loss_clip": 0.0119276,
      "auxiliary_loss_mlp": 0.00762278,
      "balance_loss_clip": 1.05610502,
      "balance_loss_mlp": 1.0002811,
      "epoch": 0.35832381410449105,
      "flos": 20302457172480.0,
      "grad_norm": 2.0402853269433128,
      "language_loss": 0.81934845,
      "learning_rate": 2.971987773994432e-06,
      "loss": 0.83889884,
      "num_input_tokens_seen": 64070295,
      "step": 2980,
      "time_per_iteration": 2.528413772583008
    },
    {
      "auxiliary_loss_clip": 0.01166342,
      "auxiliary_loss_mlp": 0.01038134,
      "balance_loss_clip": 1.05132878,
      "balance_loss_mlp": 1.02924132,
      "epoch": 0.35844405699513016,
      "flos": 16983234115200.0,
      "grad_norm": 1.7976744140797722,
      "language_loss": 0.83427614,
      "learning_rate": 2.9713069085055566e-06,
      "loss": 0.85632086,
      "num_input_tokens_seen": 64088605,
      "step": 2981,
      "time_per_iteration": 3.322321653366089
    },
    {
      "auxiliary_loss_clip": 0.01148129,
      "auxiliary_loss_mlp": 0.01025828,
      "balance_loss_clip": 1.0541656,
      "balance_loss_mlp": 1.01801395,
      "epoch": 0.35856429988576927,
      "flos": 23216858974080.0,
      "grad_norm": 1.526973959550334,
      "language_loss": 0.78915197,
      "learning_rate": 2.9706258956701958e-06,
      "loss": 0.81089151,
      "num_input_tokens_seen": 64108595,
      "step": 2982,
      "time_per_iteration": 3.4460666179656982
    },
    {
      "auxiliary_loss_clip": 0.01177297,
      "auxiliary_loss_mlp": 0.0102953,
      "balance_loss_clip": 1.05557704,
      "balance_loss_mlp": 1.02101278,
      "epoch": 0.3586845427764083,
      "flos": 23034576430080.0,
      "grad_norm": 2.8122019572984067,
      "language_loss": 0.77490187,
      "learning_rate": 2.9699447355916575e-06,
      "loss": 0.79697007,
      "num_input_tokens_seen": 64127405,
      "step": 2983,
      "time_per_iteration": 2.543665885925293
    },
    {
      "auxiliary_loss_clip": 0.01187042,
      "auxiliary_loss_mlp": 0.00761806,
      "balance_loss_clip": 1.05544615,
      "balance_loss_mlp": 1.00024033,
      "epoch": 0.35880478566704743,
      "flos": 20010682995840.0,
      "grad_norm": 2.0605366396313625,
      "language_loss": 0.74082214,
      "learning_rate": 2.969263428373275e-06,
      "loss": 0.76031059,
      "num_input_tokens_seen": 64145755,
      "step": 2984,
      "time_per_iteration": 2.5341405868530273
    },
    {
      "auxiliary_loss_clip": 0.01162012,
      "auxiliary_loss_mlp": 0.01029131,
      "balance_loss_clip": 1.05232584,
      "balance_loss_mlp": 1.02140582,
      "epoch": 0.35892502855768654,
      "flos": 13699095667200.0,
      "grad_norm": 2.107418851265342,
      "language_loss": 0.79423696,
      "learning_rate": 2.9685819741184007e-06,
      "loss": 0.8161484,
      "num_input_tokens_seen": 64164195,
      "step": 2985,
      "time_per_iteration": 2.5411946773529053
    },
    {
      "auxiliary_loss_clip": 0.01139104,
      "auxiliary_loss_mlp": 0.01028242,
      "balance_loss_clip": 1.05003095,
      "balance_loss_mlp": 1.02052939,
      "epoch": 0.3590452714483256,
      "flos": 18114096977280.0,
      "grad_norm": 2.357496905791042,
      "language_loss": 0.68380487,
      "learning_rate": 2.967900372930411e-06,
      "loss": 0.70547831,
      "num_input_tokens_seen": 64182705,
      "step": 2986,
      "time_per_iteration": 2.6030380725860596
    },
    {
      "auxiliary_loss_clip": 0.01153811,
      "auxiliary_loss_mlp": 0.01038882,
      "balance_loss_clip": 1.05075741,
      "balance_loss_mlp": 1.03059125,
      "epoch": 0.3591655143389647,
      "flos": 17749352321280.0,
      "grad_norm": 2.053023237586477,
      "language_loss": 0.78678519,
      "learning_rate": 2.9672186249127046e-06,
      "loss": 0.80871212,
      "num_input_tokens_seen": 64202170,
      "step": 2987,
      "time_per_iteration": 2.5542361736297607
    },
    {
      "auxiliary_loss_clip": 0.01160766,
      "auxiliary_loss_mlp": 0.01025244,
      "balance_loss_clip": 1.05322397,
      "balance_loss_mlp": 1.01753736,
      "epoch": 0.3592857572296038,
      "flos": 25224409082880.0,
      "grad_norm": 1.9234653587917128,
      "language_loss": 0.78805345,
      "learning_rate": 2.9665367301687014e-06,
      "loss": 0.80991352,
      "num_input_tokens_seen": 64220415,
      "step": 2988,
      "time_per_iteration": 2.622885227203369
    },
    {
      "auxiliary_loss_clip": 0.01148338,
      "auxiliary_loss_mlp": 0.01031595,
      "balance_loss_clip": 1.04827297,
      "balance_loss_mlp": 1.02348328,
      "epoch": 0.3594060001202429,
      "flos": 29384408764800.0,
      "grad_norm": 1.8680717853947078,
      "language_loss": 0.76733184,
      "learning_rate": 2.965854688801845e-06,
      "loss": 0.78913116,
      "num_input_tokens_seen": 64242475,
      "step": 2989,
      "time_per_iteration": 2.6460471153259277
    },
    {
      "auxiliary_loss_clip": 0.01169141,
      "auxiliary_loss_mlp": 0.0102596,
      "balance_loss_clip": 1.04963684,
      "balance_loss_mlp": 1.01817584,
      "epoch": 0.359526243010882,
      "flos": 17052900543360.0,
      "grad_norm": 2.024081973352873,
      "language_loss": 0.76415765,
      "learning_rate": 2.9651725009156005e-06,
      "loss": 0.78610861,
      "num_input_tokens_seen": 64260220,
      "step": 2990,
      "time_per_iteration": 2.5363292694091797
    },
    {
      "auxiliary_loss_clip": 0.01148502,
      "auxiliary_loss_mlp": 0.01036132,
      "balance_loss_clip": 1.04971313,
      "balance_loss_mlp": 1.02861595,
      "epoch": 0.3596464859015211,
      "flos": 22965089569920.0,
      "grad_norm": 1.6188057072754765,
      "language_loss": 0.74199027,
      "learning_rate": 2.964490166613454e-06,
      "loss": 0.76383662,
      "num_input_tokens_seen": 64280145,
      "step": 2991,
      "time_per_iteration": 2.581660509109497
    },
    {
      "auxiliary_loss_clip": 0.01083336,
      "auxiliary_loss_mlp": 0.01003505,
      "balance_loss_clip": 1.02283013,
      "balance_loss_mlp": 1.00199747,
      "epoch": 0.35976672879216015,
      "flos": 54739462590720.0,
      "grad_norm": 0.7593158470077602,
      "language_loss": 0.57708901,
      "learning_rate": 2.963807685998917e-06,
      "loss": 0.59795743,
      "num_input_tokens_seen": 64336010,
      "step": 2992,
      "time_per_iteration": 2.9290308952331543
    },
    {
      "auxiliary_loss_clip": 0.01134086,
      "auxiliary_loss_mlp": 0.01027848,
      "balance_loss_clip": 1.04892874,
      "balance_loss_mlp": 1.02049291,
      "epoch": 0.35988697168279926,
      "flos": 43139020901760.0,
      "grad_norm": 1.5105335910541742,
      "language_loss": 0.7806412,
      "learning_rate": 2.9631250591755196e-06,
      "loss": 0.80226058,
      "num_input_tokens_seen": 64358725,
      "step": 2993,
      "time_per_iteration": 2.8622219562530518
    },
    {
      "auxiliary_loss_clip": 0.01153637,
      "auxiliary_loss_mlp": 0.0103182,
      "balance_loss_clip": 1.05298352,
      "balance_loss_mlp": 1.02390993,
      "epoch": 0.36000721457343837,
      "flos": 35845600239360.0,
      "grad_norm": 1.834005016182697,
      "language_loss": 0.57290375,
      "learning_rate": 2.962442286246817e-06,
      "loss": 0.59475827,
      "num_input_tokens_seen": 64381555,
      "step": 2994,
      "time_per_iteration": 2.7167856693267822
    },
    {
      "auxiliary_loss_clip": 0.01161339,
      "auxiliary_loss_mlp": 0.01027333,
      "balance_loss_clip": 1.05301905,
      "balance_loss_mlp": 1.01891065,
      "epoch": 0.3601274574640774,
      "flos": 18291100222080.0,
      "grad_norm": 1.5615388103299124,
      "language_loss": 0.69543469,
      "learning_rate": 2.9617593673163853e-06,
      "loss": 0.7173214,
      "num_input_tokens_seen": 64400375,
      "step": 2995,
      "time_per_iteration": 2.611388683319092
    },
    {
      "auxiliary_loss_clip": 0.01161376,
      "auxiliary_loss_mlp": 0.01027937,
      "balance_loss_clip": 1.050843,
      "balance_loss_mlp": 1.02036738,
      "epoch": 0.36024770035471654,
      "flos": 13333955961600.0,
      "grad_norm": 2.1806970324604,
      "language_loss": 0.77278101,
      "learning_rate": 2.9610763024878216e-06,
      "loss": 0.79467416,
      "num_input_tokens_seen": 64415880,
      "step": 2996,
      "time_per_iteration": 2.559418201446533
    },
    {
      "auxiliary_loss_clip": 0.01151828,
      "auxiliary_loss_mlp": 0.01028868,
      "balance_loss_clip": 1.04971063,
      "balance_loss_mlp": 1.0203979,
      "epoch": 0.3603679432453556,
      "flos": 20267013427200.0,
      "grad_norm": 1.7239202851211788,
      "language_loss": 0.91831505,
      "learning_rate": 2.960393091864747e-06,
      "loss": 0.94012201,
      "num_input_tokens_seen": 64434260,
      "step": 2997,
      "time_per_iteration": 2.577052593231201
    },
    {
      "auxiliary_loss_clip": 0.01159764,
      "auxiliary_loss_mlp": 0.01031317,
      "balance_loss_clip": 1.0534693,
      "balance_loss_mlp": 1.02330065,
      "epoch": 0.3604881861359947,
      "flos": 22451135817600.0,
      "grad_norm": 1.8620274822251992,
      "language_loss": 0.74873924,
      "learning_rate": 2.959709735550804e-06,
      "loss": 0.77065009,
      "num_input_tokens_seen": 64453855,
      "step": 2998,
      "time_per_iteration": 2.622349977493286
    },
    {
      "auxiliary_loss_clip": 0.0113265,
      "auxiliary_loss_mlp": 0.01030815,
      "balance_loss_clip": 1.04690671,
      "balance_loss_mlp": 1.02265549,
      "epoch": 0.3606084290266338,
      "flos": 22054251467520.0,
      "grad_norm": 2.1727884090421865,
      "language_loss": 0.75719935,
      "learning_rate": 2.9590262336496575e-06,
      "loss": 0.77883404,
      "num_input_tokens_seen": 64473585,
      "step": 2999,
      "time_per_iteration": 2.6729953289031982
    },
    {
      "auxiliary_loss_clip": 0.01141071,
      "auxiliary_loss_mlp": 0.01035089,
      "balance_loss_clip": 1.05153251,
      "balance_loss_mlp": 1.02623188,
      "epoch": 0.36072867191727287,
      "flos": 15632921111040.0,
      "grad_norm": 1.9370419412350122,
      "language_loss": 0.85378653,
      "learning_rate": 2.9583425862649936e-06,
      "loss": 0.87554812,
      "num_input_tokens_seen": 64491720,
      "step": 3000,
      "time_per_iteration": 2.6145782470703125
    },
    {
      "auxiliary_loss_clip": 0.01189452,
      "auxiliary_loss_mlp": 0.01033024,
      "balance_loss_clip": 1.05695057,
      "balance_loss_mlp": 1.02455389,
      "epoch": 0.360848914807912,
      "flos": 19677000625920.0,
      "grad_norm": 2.394161443086213,
      "language_loss": 0.7371887,
      "learning_rate": 2.9576587935005215e-06,
      "loss": 0.75941342,
      "num_input_tokens_seen": 64509800,
      "step": 3001,
      "time_per_iteration": 3.3059401512145996
    },
    {
      "auxiliary_loss_clip": 0.0117561,
      "auxiliary_loss_mlp": 0.01029698,
      "balance_loss_clip": 1.05473554,
      "balance_loss_mlp": 1.02119195,
      "epoch": 0.3609691576985511,
      "flos": 18877808972160.0,
      "grad_norm": 2.3422659213003714,
      "language_loss": 0.72254425,
      "learning_rate": 2.9569748554599713e-06,
      "loss": 0.74459732,
      "num_input_tokens_seen": 64525410,
      "step": 3002,
      "time_per_iteration": 2.5386037826538086
    },
    {
      "auxiliary_loss_clip": 0.01160657,
      "auxiliary_loss_mlp": 0.01029221,
      "balance_loss_clip": 1.0539484,
      "balance_loss_mlp": 1.02106071,
      "epoch": 0.36108940058919015,
      "flos": 42224088648960.0,
      "grad_norm": 1.963416834705708,
      "language_loss": 0.73280585,
      "learning_rate": 2.956290772247097e-06,
      "loss": 0.75470459,
      "num_input_tokens_seen": 64544085,
      "step": 3003,
      "time_per_iteration": 2.777763605117798
    },
    {
      "auxiliary_loss_clip": 0.01124196,
      "auxiliary_loss_mlp": 0.01033257,
      "balance_loss_clip": 1.0501498,
      "balance_loss_mlp": 1.02540123,
      "epoch": 0.36120964347982926,
      "flos": 23185150243200.0,
      "grad_norm": 2.012552454645506,
      "language_loss": 0.73042154,
      "learning_rate": 2.9556065439656724e-06,
      "loss": 0.75199604,
      "num_input_tokens_seen": 64563135,
      "step": 3004,
      "time_per_iteration": 2.6723005771636963
    },
    {
      "auxiliary_loss_clip": 0.01105358,
      "auxiliary_loss_mlp": 0.01031262,
      "balance_loss_clip": 1.04274321,
      "balance_loss_mlp": 1.02392435,
      "epoch": 0.36132988637046837,
      "flos": 18113055482880.0,
      "grad_norm": 1.6636088509979885,
      "language_loss": 0.81939244,
      "learning_rate": 2.9549221707194952e-06,
      "loss": 0.84075868,
      "num_input_tokens_seen": 64581985,
      "step": 3005,
      "time_per_iteration": 3.427976369857788
    },
    {
      "auxiliary_loss_clip": 0.01174077,
      "auxiliary_loss_mlp": 0.01026053,
      "balance_loss_clip": 1.05388761,
      "balance_loss_mlp": 1.01808429,
      "epoch": 0.3614501292611074,
      "flos": 27813101333760.0,
      "grad_norm": 6.230958789280565,
      "language_loss": 0.72599262,
      "learning_rate": 2.954237652612384e-06,
      "loss": 0.74799389,
      "num_input_tokens_seen": 64601035,
      "step": 3006,
      "time_per_iteration": 2.624509811401367
    },
    {
      "auxiliary_loss_clip": 0.01155901,
      "auxiliary_loss_mlp": 0.01031463,
      "balance_loss_clip": 1.0528264,
      "balance_loss_mlp": 1.02391732,
      "epoch": 0.36157037215174653,
      "flos": 22634926732800.0,
      "grad_norm": 2.1223855815370523,
      "language_loss": 0.84521031,
      "learning_rate": 2.9535529897481796e-06,
      "loss": 0.86708397,
      "num_input_tokens_seen": 64618580,
      "step": 3007,
      "time_per_iteration": 3.272669553756714
    },
    {
      "auxiliary_loss_clip": 0.01187191,
      "auxiliary_loss_mlp": 0.01033454,
      "balance_loss_clip": 1.05449343,
      "balance_loss_mlp": 1.02481771,
      "epoch": 0.36169061504238564,
      "flos": 12600839376000.0,
      "grad_norm": 2.343332141191926,
      "language_loss": 0.76704162,
      "learning_rate": 2.9528681822307446e-06,
      "loss": 0.78924811,
      "num_input_tokens_seen": 64635430,
      "step": 3008,
      "time_per_iteration": 3.6193058490753174
    },
    {
      "auxiliary_loss_clip": 0.01173071,
      "auxiliary_loss_mlp": 0.00760876,
      "balance_loss_clip": 1.05787468,
      "balance_loss_mlp": 1.0002985,
      "epoch": 0.3618108579330247,
      "flos": 26684644682880.0,
      "grad_norm": 1.9610629955517311,
      "language_loss": 0.82151628,
      "learning_rate": 2.952183230163964e-06,
      "loss": 0.84085578,
      "num_input_tokens_seen": 64655005,
      "step": 3009,
      "time_per_iteration": 2.6674630641937256
    },
    {
      "auxiliary_loss_clip": 0.0113949,
      "auxiliary_loss_mlp": 0.01027984,
      "balance_loss_clip": 1.04947722,
      "balance_loss_mlp": 1.02021158,
      "epoch": 0.3619311008236638,
      "flos": 22817029708800.0,
      "grad_norm": 2.1159372241094805,
      "language_loss": 0.73341405,
      "learning_rate": 2.9514981336517448e-06,
      "loss": 0.75508881,
      "num_input_tokens_seen": 64674775,
      "step": 3010,
      "time_per_iteration": 2.6961112022399902
    },
    {
      "auxiliary_loss_clip": 0.0117384,
      "auxiliary_loss_mlp": 0.01029711,
      "balance_loss_clip": 1.05710196,
      "balance_loss_mlp": 1.02140844,
      "epoch": 0.36205134371430286,
      "flos": 25919603884800.0,
      "grad_norm": 5.047529039866728,
      "language_loss": 0.81380534,
      "learning_rate": 2.950812892798015e-06,
      "loss": 0.83584082,
      "num_input_tokens_seen": 64695670,
      "step": 3011,
      "time_per_iteration": 2.631805896759033
    },
    {
      "auxiliary_loss_clip": 0.01127869,
      "auxiliary_loss_mlp": 0.00761007,
      "balance_loss_clip": 1.05235457,
      "balance_loss_mlp": 1.00029182,
      "epoch": 0.362171586604942,
      "flos": 26139592730880.0,
      "grad_norm": 1.9554608628898114,
      "language_loss": 0.87154818,
      "learning_rate": 2.9501275077067256e-06,
      "loss": 0.89043695,
      "num_input_tokens_seen": 64716290,
      "step": 3012,
      "time_per_iteration": 2.76419997215271
    },
    {
      "auxiliary_loss_clip": 0.01097907,
      "auxiliary_loss_mlp": 0.01033944,
      "balance_loss_clip": 1.043275,
      "balance_loss_mlp": 1.02600455,
      "epoch": 0.3622918294955811,
      "flos": 28074208273920.0,
      "grad_norm": 1.4234003060072937,
      "language_loss": 0.88336098,
      "learning_rate": 2.949441978481848e-06,
      "loss": 0.90467954,
      "num_input_tokens_seen": 64737190,
      "step": 3013,
      "time_per_iteration": 2.8157968521118164
    },
    {
      "auxiliary_loss_clip": 0.0114741,
      "auxiliary_loss_mlp": 0.01028575,
      "balance_loss_clip": 1.0501188,
      "balance_loss_mlp": 1.02022409,
      "epoch": 0.36241207238622014,
      "flos": 19828005402240.0,
      "grad_norm": 1.845317238759066,
      "language_loss": 0.79953539,
      "learning_rate": 2.9487563052273778e-06,
      "loss": 0.82129514,
      "num_input_tokens_seen": 64753950,
      "step": 3014,
      "time_per_iteration": 2.590996026992798
    },
    {
      "auxiliary_loss_clip": 0.0116998,
      "auxiliary_loss_mlp": 0.01028507,
      "balance_loss_clip": 1.05644679,
      "balance_loss_mlp": 1.02057362,
      "epoch": 0.36253231527685925,
      "flos": 21397158017280.0,
      "grad_norm": 1.774398609047121,
      "language_loss": 0.85661811,
      "learning_rate": 2.94807048804733e-06,
      "loss": 0.87860298,
      "num_input_tokens_seen": 64773570,
      "step": 3015,
      "time_per_iteration": 2.557276964187622
    },
    {
      "auxiliary_loss_clip": 0.01145704,
      "auxiliary_loss_mlp": 0.01029854,
      "balance_loss_clip": 1.04830551,
      "balance_loss_mlp": 1.02128267,
      "epoch": 0.36265255816749836,
      "flos": 18362885552640.0,
      "grad_norm": 1.7973603499401292,
      "language_loss": 0.90366513,
      "learning_rate": 2.9473845270457434e-06,
      "loss": 0.9254207,
      "num_input_tokens_seen": 64790385,
      "step": 3016,
      "time_per_iteration": 2.6244359016418457
    },
    {
      "auxiliary_loss_clip": 0.01152559,
      "auxiliary_loss_mlp": 0.01030347,
      "balance_loss_clip": 1.05170941,
      "balance_loss_mlp": 1.02199697,
      "epoch": 0.3627728010581374,
      "flos": 18660046769280.0,
      "grad_norm": 2.2668606074741415,
      "language_loss": 0.69512284,
      "learning_rate": 2.946698422326677e-06,
      "loss": 0.71695197,
      "num_input_tokens_seen": 64807845,
      "step": 3017,
      "time_per_iteration": 2.580044746398926
    },
    {
      "auxiliary_loss_clip": 0.01127419,
      "auxiliary_loss_mlp": 0.01038358,
      "balance_loss_clip": 1.04691052,
      "balance_loss_mlp": 1.03058541,
      "epoch": 0.36289304394877653,
      "flos": 27524272072320.0,
      "grad_norm": 2.0801829335386484,
      "language_loss": 0.79654753,
      "learning_rate": 2.946012173994213e-06,
      "loss": 0.81820536,
      "num_input_tokens_seen": 64827630,
      "step": 3018,
      "time_per_iteration": 2.6965670585632324
    },
    {
      "auxiliary_loss_clip": 0.01169839,
      "auxiliary_loss_mlp": 0.01035495,
      "balance_loss_clip": 1.05649805,
      "balance_loss_mlp": 1.0275315,
      "epoch": 0.36301328683941564,
      "flos": 34533244932480.0,
      "grad_norm": 1.4205675411924588,
      "language_loss": 0.67735088,
      "learning_rate": 2.945325782152454e-06,
      "loss": 0.69940412,
      "num_input_tokens_seen": 64850665,
      "step": 3019,
      "time_per_iteration": 2.6987364292144775
    },
    {
      "auxiliary_loss_clip": 0.01156298,
      "auxiliary_loss_mlp": 0.01027704,
      "balance_loss_clip": 1.04832327,
      "balance_loss_mlp": 1.01970553,
      "epoch": 0.3631335297300547,
      "flos": 19025976574080.0,
      "grad_norm": 2.357773480886045,
      "language_loss": 0.78539896,
      "learning_rate": 2.9446392469055257e-06,
      "loss": 0.80723894,
      "num_input_tokens_seen": 64868700,
      "step": 3020,
      "time_per_iteration": 2.5559279918670654
    },
    {
      "auxiliary_loss_clip": 0.0113941,
      "auxiliary_loss_mlp": 0.01027221,
      "balance_loss_clip": 1.05426657,
      "balance_loss_mlp": 1.0189476,
      "epoch": 0.3632537726206938,
      "flos": 19536769929600.0,
      "grad_norm": 1.8399831065155903,
      "language_loss": 0.7978043,
      "learning_rate": 2.9439525683575745e-06,
      "loss": 0.81947064,
      "num_input_tokens_seen": 64887620,
      "step": 3021,
      "time_per_iteration": 2.586909294128418
    },
    {
      "auxiliary_loss_clip": 0.01193558,
      "auxiliary_loss_mlp": 0.01032152,
      "balance_loss_clip": 1.05860984,
      "balance_loss_mlp": 1.02362847,
      "epoch": 0.3633740155113329,
      "flos": 21068611292160.0,
      "grad_norm": 2.0230933532387465,
      "language_loss": 0.75097662,
      "learning_rate": 2.9432657466127694e-06,
      "loss": 0.77323371,
      "num_input_tokens_seen": 64907190,
      "step": 3022,
      "time_per_iteration": 2.5171566009521484
    },
    {
      "auxiliary_loss_clip": 0.01133388,
      "auxiliary_loss_mlp": 0.01029043,
      "balance_loss_clip": 1.05494785,
      "balance_loss_mlp": 1.0205673,
      "epoch": 0.36349425840197197,
      "flos": 20298722158080.0,
      "grad_norm": 1.685789417394078,
      "language_loss": 0.76334143,
      "learning_rate": 2.9425787817753007e-06,
      "loss": 0.78496575,
      "num_input_tokens_seen": 64925850,
      "step": 3023,
      "time_per_iteration": 2.6426382064819336
    },
    {
      "auxiliary_loss_clip": 0.01146917,
      "auxiliary_loss_mlp": 0.01029321,
      "balance_loss_clip": 1.05270076,
      "balance_loss_mlp": 1.0210954,
      "epoch": 0.3636145012926111,
      "flos": 29716762331520.0,
      "grad_norm": 1.4920897406731028,
      "language_loss": 0.7159313,
      "learning_rate": 2.94189167394938e-06,
      "loss": 0.73769379,
      "num_input_tokens_seen": 64948285,
      "step": 3024,
      "time_per_iteration": 2.6681315898895264
    },
    {
      "auxiliary_loss_clip": 0.0119118,
      "auxiliary_loss_mlp": 0.01035404,
      "balance_loss_clip": 1.05970085,
      "balance_loss_mlp": 1.02731538,
      "epoch": 0.3637347441832502,
      "flos": 21431847576960.0,
      "grad_norm": 2.1030748232493672,
      "language_loss": 0.81061196,
      "learning_rate": 2.941204423239241e-06,
      "loss": 0.83287781,
      "num_input_tokens_seen": 64967160,
      "step": 3025,
      "time_per_iteration": 2.5022130012512207
    },
    {
      "auxiliary_loss_clip": 0.01170257,
      "auxiliary_loss_mlp": 0.01030627,
      "balance_loss_clip": 1.05501378,
      "balance_loss_mlp": 1.02262819,
      "epoch": 0.36385498707388925,
      "flos": 29533941083520.0,
      "grad_norm": 1.7712426920408302,
      "language_loss": 0.75934488,
      "learning_rate": 2.9405170297491395e-06,
      "loss": 0.78135371,
      "num_input_tokens_seen": 64987155,
      "step": 3026,
      "time_per_iteration": 2.607408285140991
    },
    {
      "auxiliary_loss_clip": 0.01110813,
      "auxiliary_loss_mlp": 0.0076235,
      "balance_loss_clip": 1.05337465,
      "balance_loss_mlp": 1.00025439,
      "epoch": 0.36397522996452836,
      "flos": 22236569925120.0,
      "grad_norm": 1.9337125131528317,
      "language_loss": 0.80228269,
      "learning_rate": 2.939829493583353e-06,
      "loss": 0.82101429,
      "num_input_tokens_seen": 65003800,
      "step": 3027,
      "time_per_iteration": 3.3907041549682617
    },
    {
      "auxiliary_loss_clip": 0.01137575,
      "auxiliary_loss_mlp": 0.01029263,
      "balance_loss_clip": 1.04636896,
      "balance_loss_mlp": 1.0215888,
      "epoch": 0.3640954728551674,
      "flos": 21506505995520.0,
      "grad_norm": 2.6109996943259115,
      "language_loss": 0.83565295,
      "learning_rate": 2.939141814846179e-06,
      "loss": 0.85732126,
      "num_input_tokens_seen": 65021215,
      "step": 3028,
      "time_per_iteration": 2.612678289413452
    },
    {
      "auxiliary_loss_clip": 0.01159242,
      "auxiliary_loss_mlp": 0.01028346,
      "balance_loss_clip": 1.05152249,
      "balance_loss_mlp": 1.02017987,
      "epoch": 0.3642157157458065,
      "flos": 17712867081600.0,
      "grad_norm": 1.9006511532324148,
      "language_loss": 0.82323146,
      "learning_rate": 2.938453993641938e-06,
      "loss": 0.84510732,
      "num_input_tokens_seen": 65039590,
      "step": 3029,
      "time_per_iteration": 2.5615835189819336
    },
    {
      "auxiliary_loss_clip": 0.01161073,
      "auxiliary_loss_mlp": 0.01037701,
      "balance_loss_clip": 1.05670655,
      "balance_loss_mlp": 1.0288918,
      "epoch": 0.36433595863644563,
      "flos": 17639537466240.0,
      "grad_norm": 2.078396979175287,
      "language_loss": 0.70102203,
      "learning_rate": 2.937766030074973e-06,
      "loss": 0.72300977,
      "num_input_tokens_seen": 65056845,
      "step": 3030,
      "time_per_iteration": 2.5462467670440674
    },
    {
      "auxiliary_loss_clip": 0.01147677,
      "auxiliary_loss_mlp": 0.01035292,
      "balance_loss_clip": 1.05021048,
      "balance_loss_mlp": 1.02633977,
      "epoch": 0.3644562015270847,
      "flos": 26833279161600.0,
      "grad_norm": 1.76962567156387,
      "language_loss": 0.8271904,
      "learning_rate": 2.937077924249646e-06,
      "loss": 0.84902012,
      "num_input_tokens_seen": 65079435,
      "step": 3031,
      "time_per_iteration": 3.5059802532196045
    },
    {
      "auxiliary_loss_clip": 0.01165657,
      "auxiliary_loss_mlp": 0.0103464,
      "balance_loss_clip": 1.05444336,
      "balance_loss_mlp": 1.0263257,
      "epoch": 0.3645764444177238,
      "flos": 14282715847680.0,
      "grad_norm": 1.9728971158033006,
      "language_loss": 0.75358099,
      "learning_rate": 2.9363896762703443e-06,
      "loss": 0.77558398,
      "num_input_tokens_seen": 65096500,
      "step": 3032,
      "time_per_iteration": 2.570932626724243
    },
    {
      "auxiliary_loss_clip": 0.01188617,
      "auxiliary_loss_mlp": 0.01031237,
      "balance_loss_clip": 1.05628633,
      "balance_loss_mlp": 1.02251053,
      "epoch": 0.3646966873083629,
      "flos": 20667489137280.0,
      "grad_norm": 1.7226941857922682,
      "language_loss": 0.84365773,
      "learning_rate": 2.9357012862414725e-06,
      "loss": 0.86585635,
      "num_input_tokens_seen": 65115860,
      "step": 3033,
      "time_per_iteration": 3.2878799438476562
    },
    {
      "auxiliary_loss_clip": 0.01173918,
      "auxiliary_loss_mlp": 0.01027106,
      "balance_loss_clip": 1.05484903,
      "balance_loss_mlp": 1.01920271,
      "epoch": 0.36481693019900197,
      "flos": 27782613665280.0,
      "grad_norm": 2.2298614783992896,
      "language_loss": 0.72076815,
      "learning_rate": 2.9350127542674593e-06,
      "loss": 0.74277836,
      "num_input_tokens_seen": 65138070,
      "step": 3034,
      "time_per_iteration": 3.477714776992798
    },
    {
      "auxiliary_loss_clip": 0.0116331,
      "auxiliary_loss_mlp": 0.0103039,
      "balance_loss_clip": 1.05413556,
      "balance_loss_mlp": 1.02211702,
      "epoch": 0.3649371730896411,
      "flos": 19712588025600.0,
      "grad_norm": 1.7836179630980373,
      "language_loss": 0.76434505,
      "learning_rate": 2.934324080452755e-06,
      "loss": 0.78628212,
      "num_input_tokens_seen": 65155860,
      "step": 3035,
      "time_per_iteration": 2.587925910949707
    },
    {
      "auxiliary_loss_clip": 0.01137674,
      "auxiliary_loss_mlp": 0.00763068,
      "balance_loss_clip": 1.04944229,
      "balance_loss_mlp": 1.00021601,
      "epoch": 0.3650574159802802,
      "flos": 24750496016640.0,
      "grad_norm": 1.681275716849011,
      "language_loss": 0.78212619,
      "learning_rate": 2.9336352649018307e-06,
      "loss": 0.80113357,
      "num_input_tokens_seen": 65175930,
      "step": 3036,
      "time_per_iteration": 2.660174608230591
    },
    {
      "auxiliary_loss_clip": 0.01160802,
      "auxiliary_loss_mlp": 0.0103413,
      "balance_loss_clip": 1.05348969,
      "balance_loss_mlp": 1.02589321,
      "epoch": 0.36517765887091924,
      "flos": 32853487363200.0,
      "grad_norm": 1.5809240629600054,
      "language_loss": 0.69975024,
      "learning_rate": 2.9329463077191783e-06,
      "loss": 0.7216996,
      "num_input_tokens_seen": 65199305,
      "step": 3037,
      "time_per_iteration": 2.6926467418670654
    },
    {
      "auxiliary_loss_clip": 0.01133627,
      "auxiliary_loss_mlp": 0.01027345,
      "balance_loss_clip": 1.05170417,
      "balance_loss_mlp": 1.01873243,
      "epoch": 0.36529790176155835,
      "flos": 20120318282880.0,
      "grad_norm": 2.325845761087175,
      "language_loss": 0.64126492,
      "learning_rate": 2.9322572090093135e-06,
      "loss": 0.6628747,
      "num_input_tokens_seen": 65218010,
      "step": 3038,
      "time_per_iteration": 2.63631010055542
    },
    {
      "auxiliary_loss_clip": 0.01130314,
      "auxiliary_loss_mlp": 0.01027033,
      "balance_loss_clip": 1.04831934,
      "balance_loss_mlp": 1.01876545,
      "epoch": 0.36541814465219746,
      "flos": 17639573379840.0,
      "grad_norm": 3.005902018464999,
      "language_loss": 0.76386923,
      "learning_rate": 2.9315679688767713e-06,
      "loss": 0.78544271,
      "num_input_tokens_seen": 65236020,
      "step": 3039,
      "time_per_iteration": 2.6230359077453613
    },
    {
      "auxiliary_loss_clip": 0.01153348,
      "auxiliary_loss_mlp": 0.01027578,
      "balance_loss_clip": 1.05054092,
      "balance_loss_mlp": 1.01977623,
      "epoch": 0.3655383875428365,
      "flos": 22674356887680.0,
      "grad_norm": 1.9792315846174076,
      "language_loss": 0.6674186,
      "learning_rate": 2.9308785874261085e-06,
      "loss": 0.68922788,
      "num_input_tokens_seen": 65256210,
      "step": 3040,
      "time_per_iteration": 2.60276198387146
    },
    {
      "auxiliary_loss_clip": 0.01190613,
      "auxiliary_loss_mlp": 0.01034016,
      "balance_loss_clip": 1.05886972,
      "balance_loss_mlp": 1.02576697,
      "epoch": 0.36565863043347563,
      "flos": 21981173247360.0,
      "grad_norm": 1.739429228584804,
      "language_loss": 0.81576943,
      "learning_rate": 2.9301890647619045e-06,
      "loss": 0.83801568,
      "num_input_tokens_seen": 65275505,
      "step": 3041,
      "time_per_iteration": 2.564736843109131
    },
    {
      "auxiliary_loss_clip": 0.01166692,
      "auxiliary_loss_mlp": 0.01032765,
      "balance_loss_clip": 1.05669737,
      "balance_loss_mlp": 1.02407515,
      "epoch": 0.36577887332411474,
      "flos": 24827632473600.0,
      "grad_norm": 1.8394353699799733,
      "language_loss": 0.80657059,
      "learning_rate": 2.929499400988759e-06,
      "loss": 0.82856512,
      "num_input_tokens_seen": 65296665,
      "step": 3042,
      "time_per_iteration": 2.6268043518066406
    },
    {
      "auxiliary_loss_clip": 0.01177283,
      "auxiliary_loss_mlp": 0.0103047,
      "balance_loss_clip": 1.05973852,
      "balance_loss_mlp": 1.02149928,
      "epoch": 0.3658991162147538,
      "flos": 28293191539200.0,
      "grad_norm": 1.75933512841636,
      "language_loss": 0.65134609,
      "learning_rate": 2.9288095962112927e-06,
      "loss": 0.67342365,
      "num_input_tokens_seen": 65317370,
      "step": 3043,
      "time_per_iteration": 2.6216440200805664
    },
    {
      "auxiliary_loss_clip": 0.01188327,
      "auxiliary_loss_mlp": 0.01030607,
      "balance_loss_clip": 1.0578146,
      "balance_loss_mlp": 1.02167225,
      "epoch": 0.3660193591053929,
      "flos": 17785550252160.0,
      "grad_norm": 1.8625630710626468,
      "language_loss": 0.85039634,
      "learning_rate": 2.9281196505341503e-06,
      "loss": 0.87258565,
      "num_input_tokens_seen": 65334540,
      "step": 3044,
      "time_per_iteration": 2.4993529319763184
    },
    {
      "auxiliary_loss_clip": 0.0112771,
      "auxiliary_loss_mlp": 0.0076214,
      "balance_loss_clip": 1.0508734,
      "balance_loss_mlp": 1.00022769,
      "epoch": 0.36613960199603196,
      "flos": 10342776839040.0,
      "grad_norm": 1.9013913213161548,
      "language_loss": 0.78615409,
      "learning_rate": 2.9274295640619946e-06,
      "loss": 0.80505258,
      "num_input_tokens_seen": 65351670,
      "step": 3045,
      "time_per_iteration": 2.618086099624634
    },
    {
      "auxiliary_loss_clip": 0.01143418,
      "auxiliary_loss_mlp": 0.01030788,
      "balance_loss_clip": 1.04897046,
      "balance_loss_mlp": 1.0229795,
      "epoch": 0.36625984488667107,
      "flos": 19755609540480.0,
      "grad_norm": 2.0480004377426564,
      "language_loss": 0.78561926,
      "learning_rate": 2.9267393368995103e-06,
      "loss": 0.80736136,
      "num_input_tokens_seen": 65370900,
      "step": 3046,
      "time_per_iteration": 2.6518871784210205
    },
    {
      "auxiliary_loss_clip": 0.01190881,
      "auxiliary_loss_mlp": 0.01028666,
      "balance_loss_clip": 1.05841458,
      "balance_loss_mlp": 1.02002907,
      "epoch": 0.3663800877773102,
      "flos": 17674262939520.0,
      "grad_norm": 2.614841880720753,
      "language_loss": 0.74498087,
      "learning_rate": 2.926048969151407e-06,
      "loss": 0.76717639,
      "num_input_tokens_seen": 65388185,
      "step": 3047,
      "time_per_iteration": 2.4901578426361084
    },
    {
      "auxiliary_loss_clip": 0.01130427,
      "auxiliary_loss_mlp": 0.01027308,
      "balance_loss_clip": 1.05459142,
      "balance_loss_mlp": 1.01883209,
      "epoch": 0.36650033066794924,
      "flos": 20303606407680.0,
      "grad_norm": 1.6388577348152669,
      "language_loss": 0.68485755,
      "learning_rate": 2.92535846092241e-06,
      "loss": 0.70643491,
      "num_input_tokens_seen": 65407200,
      "step": 3048,
      "time_per_iteration": 2.66616153717041
    },
    {
      "auxiliary_loss_clip": 0.01161495,
      "auxiliary_loss_mlp": 0.01030697,
      "balance_loss_clip": 1.05368531,
      "balance_loss_mlp": 1.02194738,
      "epoch": 0.36662057355858835,
      "flos": 24716237420160.0,
      "grad_norm": 1.5452805478719565,
      "language_loss": 0.82730567,
      "learning_rate": 2.9246678123172704e-06,
      "loss": 0.84922755,
      "num_input_tokens_seen": 65427290,
      "step": 3049,
      "time_per_iteration": 2.605611801147461
    },
    {
      "auxiliary_loss_clip": 0.01191277,
      "auxiliary_loss_mlp": 0.01035257,
      "balance_loss_clip": 1.0582962,
      "balance_loss_mlp": 1.02623308,
      "epoch": 0.36674081644922746,
      "flos": 12385267902720.0,
      "grad_norm": 2.8671481837670036,
      "language_loss": 0.74635351,
      "learning_rate": 2.9239770234407596e-06,
      "loss": 0.76861882,
      "num_input_tokens_seen": 65445595,
      "step": 3050,
      "time_per_iteration": 2.503836154937744
    },
    {
      "auxiliary_loss_clip": 0.01175483,
      "auxiliary_loss_mlp": 0.01027363,
      "balance_loss_clip": 1.05520046,
      "balance_loss_mlp": 1.01811838,
      "epoch": 0.3668610593398665,
      "flos": 21105922544640.0,
      "grad_norm": 1.555021859311062,
      "language_loss": 0.67829776,
      "learning_rate": 2.9232860943976686e-06,
      "loss": 0.7003262,
      "num_input_tokens_seen": 65466330,
      "step": 3051,
      "time_per_iteration": 2.5863325595855713
    },
    {
      "auxiliary_loss_clip": 0.01161029,
      "auxiliary_loss_mlp": 0.01031372,
      "balance_loss_clip": 1.05432987,
      "balance_loss_mlp": 1.02385259,
      "epoch": 0.3669813022305056,
      "flos": 26758082039040.0,
      "grad_norm": 1.6205452753751888,
      "language_loss": 0.84367633,
      "learning_rate": 2.9225950252928115e-06,
      "loss": 0.86560035,
      "num_input_tokens_seen": 65487180,
      "step": 3052,
      "time_per_iteration": 3.423762083053589
    },
    {
      "auxiliary_loss_clip": 0.01177108,
      "auxiliary_loss_mlp": 0.01028949,
      "balance_loss_clip": 1.05907035,
      "balance_loss_mlp": 1.01981807,
      "epoch": 0.36710154512114473,
      "flos": 19099521671040.0,
      "grad_norm": 2.806793971641526,
      "language_loss": 0.8180486,
      "learning_rate": 2.9219038162310217e-06,
      "loss": 0.84010917,
      "num_input_tokens_seen": 65505380,
      "step": 3053,
      "time_per_iteration": 2.524834632873535
    },
    {
      "auxiliary_loss_clip": 0.01103214,
      "auxiliary_loss_mlp": 0.00762307,
      "balance_loss_clip": 1.04688001,
      "balance_loss_mlp": 1.00020778,
      "epoch": 0.3672217880117838,
      "flos": 20812029465600.0,
      "grad_norm": 1.7769446040248729,
      "language_loss": 0.82772493,
      "learning_rate": 2.921212467317157e-06,
      "loss": 0.84638011,
      "num_input_tokens_seen": 65524825,
      "step": 3054,
      "time_per_iteration": 2.71394944190979
    },
    {
      "auxiliary_loss_clip": 0.01141752,
      "auxiliary_loss_mlp": 0.01038292,
      "balance_loss_clip": 1.0466361,
      "balance_loss_mlp": 1.02902913,
      "epoch": 0.3673420309024229,
      "flos": 13590394133760.0,
      "grad_norm": 1.7683938955733205,
      "language_loss": 0.79741436,
      "learning_rate": 2.920520978656093e-06,
      "loss": 0.81921482,
      "num_input_tokens_seen": 65541790,
      "step": 3055,
      "time_per_iteration": 2.5482518672943115
    },
    {
      "auxiliary_loss_clip": 0.01188595,
      "auxiliary_loss_mlp": 0.00762093,
      "balance_loss_clip": 1.05794322,
      "balance_loss_mlp": 1.00021434,
      "epoch": 0.367462273793062,
      "flos": 28986877969920.0,
      "grad_norm": 1.8278809278783392,
      "language_loss": 0.76687223,
      "learning_rate": 2.919829350352729e-06,
      "loss": 0.7863791,
      "num_input_tokens_seen": 65563395,
      "step": 3056,
      "time_per_iteration": 2.5884175300598145
    },
    {
      "auxiliary_loss_clip": 0.01082836,
      "auxiliary_loss_mlp": 0.0100887,
      "balance_loss_clip": 1.02306163,
      "balance_loss_mlp": 1.00726056,
      "epoch": 0.36758251668370107,
      "flos": 62643148346880.0,
      "grad_norm": 0.8191791098836778,
      "language_loss": 0.60094762,
      "learning_rate": 2.919137582511983e-06,
      "loss": 0.62186468,
      "num_input_tokens_seen": 65619835,
      "step": 3057,
      "time_per_iteration": 3.822255849838257
    },
    {
      "auxiliary_loss_clip": 0.01152975,
      "auxiliary_loss_mlp": 0.01032352,
      "balance_loss_clip": 1.05742431,
      "balance_loss_mlp": 1.02365518,
      "epoch": 0.3677027595743402,
      "flos": 12713886455040.0,
      "grad_norm": 1.9685655889550187,
      "language_loss": 0.64195728,
      "learning_rate": 2.918445675238797e-06,
      "loss": 0.66381061,
      "num_input_tokens_seen": 65636760,
      "step": 3058,
      "time_per_iteration": 3.3292181491851807
    },
    {
      "auxiliary_loss_clip": 0.01188056,
      "auxiliary_loss_mlp": 0.01030023,
      "balance_loss_clip": 1.0557456,
      "balance_loss_mlp": 1.02165508,
      "epoch": 0.36782300246497923,
      "flos": 25046579825280.0,
      "grad_norm": 1.9199615397620513,
      "language_loss": 0.69417918,
      "learning_rate": 2.917753628638132e-06,
      "loss": 0.71635997,
      "num_input_tokens_seen": 65657065,
      "step": 3059,
      "time_per_iteration": 2.565377712249756
    },
    {
      "auxiliary_loss_clip": 0.01162677,
      "auxiliary_loss_mlp": 0.01038534,
      "balance_loss_clip": 1.05530345,
      "balance_loss_mlp": 1.02993274,
      "epoch": 0.36794324535561834,
      "flos": 17419512706560.0,
      "grad_norm": 2.148561141928926,
      "language_loss": 0.70094121,
      "learning_rate": 2.9170614428149716e-06,
      "loss": 0.72295332,
      "num_input_tokens_seen": 65675400,
      "step": 3060,
      "time_per_iteration": 3.3701725006103516
    },
    {
      "auxiliary_loss_clip": 0.01143916,
      "auxiliary_loss_mlp": 0.01034783,
      "balance_loss_clip": 1.05254149,
      "balance_loss_mlp": 1.02533615,
      "epoch": 0.36806348824625745,
      "flos": 24089128848000.0,
      "grad_norm": 2.5323596018777814,
      "language_loss": 0.86748183,
      "learning_rate": 2.9163691178743195e-06,
      "loss": 0.88926882,
      "num_input_tokens_seen": 65694050,
      "step": 3061,
      "time_per_iteration": 2.6102726459503174
    },
    {
      "auxiliary_loss_clip": 0.01172401,
      "auxiliary_loss_mlp": 0.01029811,
      "balance_loss_clip": 1.05480886,
      "balance_loss_mlp": 1.02193129,
      "epoch": 0.3681837311368965,
      "flos": 20521871400960.0,
      "grad_norm": 2.2758273244429827,
      "language_loss": 0.77357745,
      "learning_rate": 2.9156766539212006e-06,
      "loss": 0.79559958,
      "num_input_tokens_seen": 65711695,
      "step": 3062,
      "time_per_iteration": 2.579263687133789
    },
    {
      "auxiliary_loss_clip": 0.01177576,
      "auxiliary_loss_mlp": 0.01036095,
      "balance_loss_clip": 1.05397654,
      "balance_loss_mlp": 1.0277319,
      "epoch": 0.3683039740275356,
      "flos": 21466644877440.0,
      "grad_norm": 2.075404749999038,
      "language_loss": 0.71951866,
      "learning_rate": 2.9149840510606614e-06,
      "loss": 0.74165535,
      "num_input_tokens_seen": 65730350,
      "step": 3063,
      "time_per_iteration": 2.5790624618530273
    },
    {
      "auxiliary_loss_clip": 0.01069476,
      "auxiliary_loss_mlp": 0.00752152,
      "balance_loss_clip": 1.02090478,
      "balance_loss_mlp": 1.00012386,
      "epoch": 0.36842421691817473,
      "flos": 70380999987840.0,
      "grad_norm": 1.0347194434592082,
      "language_loss": 0.64215702,
      "learning_rate": 2.914291309397769e-06,
      "loss": 0.66037327,
      "num_input_tokens_seen": 65787820,
      "step": 3064,
      "time_per_iteration": 3.246145248413086
    },
    {
      "auxiliary_loss_clip": 0.01106687,
      "auxiliary_loss_mlp": 0.01032249,
      "balance_loss_clip": 1.04713821,
      "balance_loss_mlp": 1.02320063,
      "epoch": 0.3685444598088138,
      "flos": 23331378510720.0,
      "grad_norm": 1.9023975164964235,
      "language_loss": 0.78268087,
      "learning_rate": 2.9135984290376117e-06,
      "loss": 0.80407023,
      "num_input_tokens_seen": 65806685,
      "step": 3065,
      "time_per_iteration": 2.6756486892700195
    },
    {
      "auxiliary_loss_clip": 0.01113001,
      "auxiliary_loss_mlp": 0.01037699,
      "balance_loss_clip": 1.04570687,
      "balance_loss_mlp": 1.02930677,
      "epoch": 0.3686647026994529,
      "flos": 23070271570560.0,
      "grad_norm": 2.011975139913776,
      "language_loss": 0.82539231,
      "learning_rate": 2.9129054100853e-06,
      "loss": 0.84689927,
      "num_input_tokens_seen": 65825525,
      "step": 3066,
      "time_per_iteration": 2.7358574867248535
    },
    {
      "auxiliary_loss_clip": 0.01159104,
      "auxiliary_loss_mlp": 0.01036792,
      "balance_loss_clip": 1.05203962,
      "balance_loss_mlp": 1.0278635,
      "epoch": 0.368784945590092,
      "flos": 25119909440640.0,
      "grad_norm": 1.6223965101953781,
      "language_loss": 0.75827396,
      "learning_rate": 2.912212252645963e-06,
      "loss": 0.78023291,
      "num_input_tokens_seen": 65848110,
      "step": 3067,
      "time_per_iteration": 2.6248300075531006
    },
    {
      "auxiliary_loss_clip": 0.01178366,
      "auxiliary_loss_mlp": 0.01034246,
      "balance_loss_clip": 1.05385065,
      "balance_loss_mlp": 1.02538848,
      "epoch": 0.36890518848073106,
      "flos": 18442284566400.0,
      "grad_norm": 1.9277229167586838,
      "language_loss": 0.76767355,
      "learning_rate": 2.9115189568247523e-06,
      "loss": 0.78979963,
      "num_input_tokens_seen": 65865670,
      "step": 3068,
      "time_per_iteration": 2.585052490234375
    },
    {
      "auxiliary_loss_clip": 0.0112304,
      "auxiliary_loss_mlp": 0.01034124,
      "balance_loss_clip": 1.05406666,
      "balance_loss_mlp": 1.02585936,
      "epoch": 0.36902543137137017,
      "flos": 16362446336640.0,
      "grad_norm": 2.961541685481398,
      "language_loss": 0.9197461,
      "learning_rate": 2.910825522726841e-06,
      "loss": 0.94131768,
      "num_input_tokens_seen": 65883195,
      "step": 3069,
      "time_per_iteration": 2.6364240646362305
    },
    {
      "auxiliary_loss_clip": 0.01124674,
      "auxiliary_loss_mlp": 0.01035394,
      "balance_loss_clip": 1.0478214,
      "balance_loss_mlp": 1.02716875,
      "epoch": 0.3691456742620093,
      "flos": 12275596702080.0,
      "grad_norm": 1.8801250294984104,
      "language_loss": 0.77230316,
      "learning_rate": 2.9101319504574215e-06,
      "loss": 0.79390383,
      "num_input_tokens_seen": 65899635,
      "step": 3070,
      "time_per_iteration": 2.650235891342163
    },
    {
      "auxiliary_loss_clip": 0.01163101,
      "auxiliary_loss_mlp": 0.01023635,
      "balance_loss_clip": 1.05046046,
      "balance_loss_mlp": 1.01534951,
      "epoch": 0.36926591715264834,
      "flos": 17786412178560.0,
      "grad_norm": 2.0968095878394926,
      "language_loss": 0.76134789,
      "learning_rate": 2.909438240121709e-06,
      "loss": 0.78321517,
      "num_input_tokens_seen": 65919910,
      "step": 3071,
      "time_per_iteration": 2.5852949619293213
    },
    {
      "auxiliary_loss_clip": 0.01153484,
      "auxiliary_loss_mlp": 0.01035242,
      "balance_loss_clip": 1.05448127,
      "balance_loss_mlp": 1.02719545,
      "epoch": 0.36938616004328745,
      "flos": 28948309741440.0,
      "grad_norm": 1.624665255715946,
      "language_loss": 0.7051416,
      "learning_rate": 2.908744391824939e-06,
      "loss": 0.72702885,
      "num_input_tokens_seen": 65940930,
      "step": 3072,
      "time_per_iteration": 2.6355745792388916
    },
    {
      "auxiliary_loss_clip": 0.01117484,
      "auxiliary_loss_mlp": 0.01026109,
      "balance_loss_clip": 1.04616261,
      "balance_loss_mlp": 1.01765084,
      "epoch": 0.36950640293392656,
      "flos": 29205394358400.0,
      "grad_norm": 1.6971752209306938,
      "language_loss": 0.7891084,
      "learning_rate": 2.908050405672367e-06,
      "loss": 0.81054437,
      "num_input_tokens_seen": 65960475,
      "step": 3073,
      "time_per_iteration": 2.740341901779175
    },
    {
      "auxiliary_loss_clip": 0.0116394,
      "auxiliary_loss_mlp": 0.01027415,
      "balance_loss_clip": 1.05004096,
      "balance_loss_mlp": 1.01885569,
      "epoch": 0.3696266458245656,
      "flos": 24827776128000.0,
      "grad_norm": 1.702741025165438,
      "language_loss": 0.79109931,
      "learning_rate": 2.9073562817692703e-06,
      "loss": 0.81301296,
      "num_input_tokens_seen": 65979160,
      "step": 3074,
      "time_per_iteration": 2.620823860168457
    },
    {
      "auxiliary_loss_clip": 0.01045622,
      "auxiliary_loss_mlp": 0.01001565,
      "balance_loss_clip": 1.02275658,
      "balance_loss_mlp": 0.99986076,
      "epoch": 0.3697468887152047,
      "flos": 59887257264000.0,
      "grad_norm": 0.7213313569317554,
      "language_loss": 0.56544518,
      "learning_rate": 2.9066620202209468e-06,
      "loss": 0.585917,
      "num_input_tokens_seen": 66041650,
      "step": 3075,
      "time_per_iteration": 3.2141683101654053
    },
    {
      "auxiliary_loss_clip": 0.01136179,
      "auxiliary_loss_mlp": 0.01026171,
      "balance_loss_clip": 1.05037737,
      "balance_loss_mlp": 1.01789784,
      "epoch": 0.3698671316058438,
      "flos": 26137581569280.0,
      "grad_norm": 1.8204161791523308,
      "language_loss": 0.77551782,
      "learning_rate": 2.905967621132716e-06,
      "loss": 0.79714131,
      "num_input_tokens_seen": 66059260,
      "step": 3076,
      "time_per_iteration": 2.659736394882202
    },
    {
      "auxiliary_loss_clip": 0.01158659,
      "auxiliary_loss_mlp": 0.01030346,
      "balance_loss_clip": 1.05056643,
      "balance_loss_mlp": 1.0214529,
      "epoch": 0.3699873744964829,
      "flos": 24607464059520.0,
      "grad_norm": 2.0423881445281955,
      "language_loss": 0.75529003,
      "learning_rate": 2.9052730846099172e-06,
      "loss": 0.77718008,
      "num_input_tokens_seen": 66080605,
      "step": 3077,
      "time_per_iteration": 2.614818572998047
    },
    {
      "auxiliary_loss_clip": 0.01059164,
      "auxiliary_loss_mlp": 0.01002919,
      "balance_loss_clip": 1.01953566,
      "balance_loss_mlp": 1.00122607,
      "epoch": 0.370107617387122,
      "flos": 64885340050560.0,
      "grad_norm": 0.8579539014062775,
      "language_loss": 0.6100443,
      "learning_rate": 2.9045784107579123e-06,
      "loss": 0.63066512,
      "num_input_tokens_seen": 66140710,
      "step": 3078,
      "time_per_iteration": 3.914754867553711
    },
    {
      "auxiliary_loss_clip": 0.01188831,
      "auxiliary_loss_mlp": 0.01027849,
      "balance_loss_clip": 1.05717099,
      "balance_loss_mlp": 1.02026725,
      "epoch": 0.37022786027776106,
      "flos": 15961683317760.0,
      "grad_norm": 1.8514519508278144,
      "language_loss": 0.66742313,
      "learning_rate": 2.9038835996820807e-06,
      "loss": 0.68958992,
      "num_input_tokens_seen": 66158320,
      "step": 3079,
      "time_per_iteration": 2.482332944869995
    },
    {
      "auxiliary_loss_clip": 0.01147922,
      "auxiliary_loss_mlp": 0.01040174,
      "balance_loss_clip": 1.0488236,
      "balance_loss_mlp": 1.03209162,
      "epoch": 0.37034810316840017,
      "flos": 18546927863040.0,
      "grad_norm": 1.6906608437094306,
      "language_loss": 0.79733455,
      "learning_rate": 2.903188651487826e-06,
      "loss": 0.81921548,
      "num_input_tokens_seen": 66176875,
      "step": 3080,
      "time_per_iteration": 2.6286144256591797
    },
    {
      "auxiliary_loss_clip": 0.01177393,
      "auxiliary_loss_mlp": 0.01035292,
      "balance_loss_clip": 1.05609608,
      "balance_loss_mlp": 1.0269115,
      "epoch": 0.3704683460590393,
      "flos": 17821927751040.0,
      "grad_norm": 2.026080146369784,
      "language_loss": 0.86605823,
      "learning_rate": 2.902493566280571e-06,
      "loss": 0.88818514,
      "num_input_tokens_seen": 66194980,
      "step": 3081,
      "time_per_iteration": 2.4914939403533936
    },
    {
      "auxiliary_loss_clip": 0.0115648,
      "auxiliary_loss_mlp": 0.01032753,
      "balance_loss_clip": 1.05311823,
      "balance_loss_mlp": 1.02411056,
      "epoch": 0.37058858894967833,
      "flos": 14134081368960.0,
      "grad_norm": 1.785586424285132,
      "language_loss": 0.80917335,
      "learning_rate": 2.9017983441657595e-06,
      "loss": 0.83106565,
      "num_input_tokens_seen": 66212310,
      "step": 3082,
      "time_per_iteration": 2.5235087871551514
    },
    {
      "auxiliary_loss_clip": 0.01131251,
      "auxiliary_loss_mlp": 0.01029464,
      "balance_loss_clip": 1.04671597,
      "balance_loss_mlp": 1.02097678,
      "epoch": 0.37070883184031744,
      "flos": 13954492344960.0,
      "grad_norm": 2.6531039938454777,
      "language_loss": 0.7503525,
      "learning_rate": 2.9011029852488564e-06,
      "loss": 0.77195966,
      "num_input_tokens_seen": 66229545,
      "step": 3083,
      "time_per_iteration": 3.3755571842193604
    },
    {
      "auxiliary_loss_clip": 0.01080276,
      "auxiliary_loss_mlp": 0.0100149,
      "balance_loss_clip": 1.02083635,
      "balance_loss_mlp": 0.99997008,
      "epoch": 0.37082907473095655,
      "flos": 52315419306240.0,
      "grad_norm": 0.9835191681181462,
      "language_loss": 0.62461263,
      "learning_rate": 2.9004074896353465e-06,
      "loss": 0.64543027,
      "num_input_tokens_seen": 66283545,
      "step": 3084,
      "time_per_iteration": 3.6774215698242188
    },
    {
      "auxiliary_loss_clip": 0.01186565,
      "auxiliary_loss_mlp": 0.01032013,
      "balance_loss_clip": 1.05795574,
      "balance_loss_mlp": 1.02381778,
      "epoch": 0.3709493176215956,
      "flos": 15998096730240.0,
      "grad_norm": 1.6649196559817716,
      "language_loss": 0.81604838,
      "learning_rate": 2.8997118574307362e-06,
      "loss": 0.83823419,
      "num_input_tokens_seen": 66300500,
      "step": 3085,
      "time_per_iteration": 3.300213098526001
    },
    {
      "auxiliary_loss_clip": 0.01150652,
      "auxiliary_loss_mlp": 0.01031741,
      "balance_loss_clip": 1.05212224,
      "balance_loss_mlp": 1.02330148,
      "epoch": 0.3710695605122347,
      "flos": 20959837931520.0,
      "grad_norm": 1.8370776807017108,
      "language_loss": 0.74297613,
      "learning_rate": 2.899016088740553e-06,
      "loss": 0.76480007,
      "num_input_tokens_seen": 66318610,
      "step": 3086,
      "time_per_iteration": 2.5594472885131836
    },
    {
      "auxiliary_loss_clip": 0.0113054,
      "auxiliary_loss_mlp": 0.01035854,
      "balance_loss_clip": 1.05115592,
      "balance_loss_mlp": 1.02755141,
      "epoch": 0.37118980340287383,
      "flos": 14355578586240.0,
      "grad_norm": 2.01558522223361,
      "language_loss": 0.79210949,
      "learning_rate": 2.898320183670344e-06,
      "loss": 0.81377345,
      "num_input_tokens_seen": 66336025,
      "step": 3087,
      "time_per_iteration": 2.606536626815796
    },
    {
      "auxiliary_loss_clip": 0.01130537,
      "auxiliary_loss_mlp": 0.01037087,
      "balance_loss_clip": 1.05345142,
      "balance_loss_mlp": 1.02880192,
      "epoch": 0.3713100462935129,
      "flos": 25885381201920.0,
      "grad_norm": 1.6283201077915095,
      "language_loss": 0.89131236,
      "learning_rate": 2.8976241423256767e-06,
      "loss": 0.9129886,
      "num_input_tokens_seen": 66356120,
      "step": 3088,
      "time_per_iteration": 2.6480982303619385
    },
    {
      "auxiliary_loss_clip": 0.01152222,
      "auxiliary_loss_mlp": 0.0103117,
      "balance_loss_clip": 1.05087292,
      "balance_loss_mlp": 1.02371931,
      "epoch": 0.371430289184152,
      "flos": 30518934814080.0,
      "grad_norm": 3.5707084902204342,
      "language_loss": 0.67942512,
      "learning_rate": 2.896927964812142e-06,
      "loss": 0.70125902,
      "num_input_tokens_seen": 66376685,
      "step": 3089,
      "time_per_iteration": 2.6328155994415283
    },
    {
      "auxiliary_loss_clip": 0.0115504,
      "auxiliary_loss_mlp": 0.01030157,
      "balance_loss_clip": 1.05339503,
      "balance_loss_mlp": 1.02196133,
      "epoch": 0.37155053207479105,
      "flos": 15742233175680.0,
      "grad_norm": 2.301872782142269,
      "language_loss": 0.74872816,
      "learning_rate": 2.8962316512353465e-06,
      "loss": 0.77058011,
      "num_input_tokens_seen": 66394230,
      "step": 3090,
      "time_per_iteration": 2.5298497676849365
    },
    {
      "auxiliary_loss_clip": 0.01110719,
      "auxiliary_loss_mlp": 0.01030922,
      "balance_loss_clip": 1.04499722,
      "balance_loss_mlp": 1.02294111,
      "epoch": 0.37167077496543016,
      "flos": 23404061681280.0,
      "grad_norm": 1.492641239836418,
      "language_loss": 0.74721301,
      "learning_rate": 2.8955352017009233e-06,
      "loss": 0.76862949,
      "num_input_tokens_seen": 66413475,
      "step": 3091,
      "time_per_iteration": 2.6866841316223145
    },
    {
      "auxiliary_loss_clip": 0.01157479,
      "auxiliary_loss_mlp": 0.01036611,
      "balance_loss_clip": 1.05369103,
      "balance_loss_mlp": 1.02840972,
      "epoch": 0.3717910178560693,
      "flos": 22088653718400.0,
      "grad_norm": 1.7596695269494207,
      "language_loss": 0.77213919,
      "learning_rate": 2.8948386163145212e-06,
      "loss": 0.79408002,
      "num_input_tokens_seen": 66432685,
      "step": 3092,
      "time_per_iteration": 2.5968210697174072
    },
    {
      "auxiliary_loss_clip": 0.01175526,
      "auxiliary_loss_mlp": 0.01026011,
      "balance_loss_clip": 1.0554738,
      "balance_loss_mlp": 1.01829767,
      "epoch": 0.3719112607467083,
      "flos": 26939969533440.0,
      "grad_norm": 1.7400800800950047,
      "language_loss": 0.79439336,
      "learning_rate": 2.8941418951818135e-06,
      "loss": 0.81640869,
      "num_input_tokens_seen": 66452245,
      "step": 3093,
      "time_per_iteration": 2.5818326473236084
    },
    {
      "auxiliary_loss_clip": 0.0114578,
      "auxiliary_loss_mlp": 0.01026169,
      "balance_loss_clip": 1.04969275,
      "balance_loss_mlp": 1.01872683,
      "epoch": 0.37203150363734744,
      "flos": 12166500119040.0,
      "grad_norm": 2.0993453355913756,
      "language_loss": 0.71011704,
      "learning_rate": 2.8934450384084903e-06,
      "loss": 0.7318365,
      "num_input_tokens_seen": 66469760,
      "step": 3094,
      "time_per_iteration": 2.583446741104126
    },
    {
      "auxiliary_loss_clip": 0.01149581,
      "auxiliary_loss_mlp": 0.01030282,
      "balance_loss_clip": 1.05014801,
      "balance_loss_mlp": 1.02197325,
      "epoch": 0.37215174652798655,
      "flos": 23697595624320.0,
      "grad_norm": 2.0630284855135597,
      "language_loss": 0.69538569,
      "learning_rate": 2.8927480461002653e-06,
      "loss": 0.71718436,
      "num_input_tokens_seen": 66489730,
      "step": 3095,
      "time_per_iteration": 2.5924856662750244
    },
    {
      "auxiliary_loss_clip": 0.01154355,
      "auxiliary_loss_mlp": 0.01033382,
      "balance_loss_clip": 1.04799032,
      "balance_loss_mlp": 1.02464437,
      "epoch": 0.3722719894186256,
      "flos": 17887751424000.0,
      "grad_norm": 3.6124445772948035,
      "language_loss": 0.85957086,
      "learning_rate": 2.892050918362872e-06,
      "loss": 0.88144827,
      "num_input_tokens_seen": 66504785,
      "step": 3096,
      "time_per_iteration": 2.5162241458892822
    },
    {
      "auxiliary_loss_clip": 0.01009706,
      "auxiliary_loss_mlp": 0.01004555,
      "balance_loss_clip": 1.01421881,
      "balance_loss_mlp": 1.00317764,
      "epoch": 0.3723922323092647,
      "flos": 62419891363200.0,
      "grad_norm": 0.8444655002282905,
      "language_loss": 0.5600332,
      "learning_rate": 2.8913536553020626e-06,
      "loss": 0.58017582,
      "num_input_tokens_seen": 66558840,
      "step": 3097,
      "time_per_iteration": 3.3245577812194824
    },
    {
      "auxiliary_loss_clip": 0.01119395,
      "auxiliary_loss_mlp": 0.01034725,
      "balance_loss_clip": 1.04538298,
      "balance_loss_mlp": 1.0263443,
      "epoch": 0.3725124751999038,
      "flos": 23039747988480.0,
      "grad_norm": 1.7917782439039736,
      "language_loss": 0.85026723,
      "learning_rate": 2.8906562570236137e-06,
      "loss": 0.87180841,
      "num_input_tokens_seen": 66576750,
      "step": 3098,
      "time_per_iteration": 2.7825443744659424
    },
    {
      "auxiliary_loss_clip": 0.01112119,
      "auxiliary_loss_mlp": 0.01026784,
      "balance_loss_clip": 1.04340136,
      "balance_loss_mlp": 1.01937521,
      "epoch": 0.3726327180905429,
      "flos": 20920551431040.0,
      "grad_norm": 1.5166029996182928,
      "language_loss": 0.76629043,
      "learning_rate": 2.889958723633318e-06,
      "loss": 0.78767943,
      "num_input_tokens_seen": 66595690,
      "step": 3099,
      "time_per_iteration": 2.6565101146698
    },
    {
      "auxiliary_loss_clip": 0.01143699,
      "auxiliary_loss_mlp": 0.01032107,
      "balance_loss_clip": 1.0531491,
      "balance_loss_mlp": 1.0239526,
      "epoch": 0.372752960981182,
      "flos": 30592156688640.0,
      "grad_norm": 1.6180833413597637,
      "language_loss": 0.73731536,
      "learning_rate": 2.889261055236992e-06,
      "loss": 0.75907344,
      "num_input_tokens_seen": 66617905,
      "step": 3100,
      "time_per_iteration": 3.058896064758301
    },
    {
      "auxiliary_loss_clip": 0.01157165,
      "auxiliary_loss_mlp": 0.01033167,
      "balance_loss_clip": 1.05396652,
      "balance_loss_mlp": 1.02491772,
      "epoch": 0.3728732038718211,
      "flos": 25116749043840.0,
      "grad_norm": 1.849492178690653,
      "language_loss": 0.82501423,
      "learning_rate": 2.8885632519404704e-06,
      "loss": 0.84691757,
      "num_input_tokens_seen": 66638175,
      "step": 3101,
      "time_per_iteration": 2.6088409423828125
    },
    {
      "auxiliary_loss_clip": 0.01156361,
      "auxiliary_loss_mlp": 0.01030068,
      "balance_loss_clip": 1.05325365,
      "balance_loss_mlp": 1.02203298,
      "epoch": 0.37299344676246016,
      "flos": 25302048330240.0,
      "grad_norm": 2.0051315284697098,
      "language_loss": 0.75411737,
      "learning_rate": 2.8878653138496107e-06,
      "loss": 0.77598155,
      "num_input_tokens_seen": 66658670,
      "step": 3102,
      "time_per_iteration": 2.62419056892395
    },
    {
      "auxiliary_loss_clip": 0.01111724,
      "auxiliary_loss_mlp": 0.01029633,
      "balance_loss_clip": 1.04370189,
      "balance_loss_mlp": 1.02160072,
      "epoch": 0.37311368965309927,
      "flos": 23842531002240.0,
      "grad_norm": 2.212292725787708,
      "language_loss": 0.76913017,
      "learning_rate": 2.8871672410702878e-06,
      "loss": 0.79054368,
      "num_input_tokens_seen": 66676030,
      "step": 3103,
      "time_per_iteration": 2.671839714050293
    },
    {
      "auxiliary_loss_clip": 0.01148695,
      "auxiliary_loss_mlp": 0.01028559,
      "balance_loss_clip": 1.04836142,
      "balance_loss_mlp": 1.01965976,
      "epoch": 0.3732339325437384,
      "flos": 25811943845760.0,
      "grad_norm": 1.6788766857892055,
      "language_loss": 0.81964219,
      "learning_rate": 2.8864690337084008e-06,
      "loss": 0.84141469,
      "num_input_tokens_seen": 66695305,
      "step": 3104,
      "time_per_iteration": 3.5025062561035156
    },
    {
      "auxiliary_loss_clip": 0.01165877,
      "auxiliary_loss_mlp": 0.01035587,
      "balance_loss_clip": 1.05122685,
      "balance_loss_mlp": 1.02755868,
      "epoch": 0.37335417543437743,
      "flos": 26208433146240.0,
      "grad_norm": 1.9100694158133065,
      "language_loss": 0.77847707,
      "learning_rate": 2.885770691869866e-06,
      "loss": 0.80049175,
      "num_input_tokens_seen": 66716185,
      "step": 3105,
      "time_per_iteration": 2.567470073699951
    },
    {
      "auxiliary_loss_clip": 0.01167675,
      "auxiliary_loss_mlp": 0.01033682,
      "balance_loss_clip": 1.05389857,
      "balance_loss_mlp": 1.02549219,
      "epoch": 0.37347441832501654,
      "flos": 24023879792640.0,
      "grad_norm": 2.224968857459263,
      "language_loss": 0.74490809,
      "learning_rate": 2.8850722156606207e-06,
      "loss": 0.76692164,
      "num_input_tokens_seen": 66734575,
      "step": 3106,
      "time_per_iteration": 2.5556812286376953
    },
    {
      "auxiliary_loss_clip": 0.01162036,
      "auxiliary_loss_mlp": 0.01035635,
      "balance_loss_clip": 1.04940033,
      "balance_loss_mlp": 1.02794611,
      "epoch": 0.3735946612156556,
      "flos": 19714922409600.0,
      "grad_norm": 1.6230341502152954,
      "language_loss": 0.66967607,
      "learning_rate": 2.8843736051866252e-06,
      "loss": 0.69165277,
      "num_input_tokens_seen": 66753500,
      "step": 3107,
      "time_per_iteration": 2.5148329734802246
    },
    {
      "auxiliary_loss_clip": 0.01126453,
      "auxiliary_loss_mlp": 0.00761312,
      "balance_loss_clip": 1.04712462,
      "balance_loss_mlp": 1.00029409,
      "epoch": 0.3737149041062947,
      "flos": 23039604334080.0,
      "grad_norm": 1.519579327745257,
      "language_loss": 0.69546807,
      "learning_rate": 2.8836748605538557e-06,
      "loss": 0.71434569,
      "num_input_tokens_seen": 66775140,
      "step": 3108,
      "time_per_iteration": 2.670302629470825
    },
    {
      "auxiliary_loss_clip": 0.01161312,
      "auxiliary_loss_mlp": 0.01028925,
      "balance_loss_clip": 1.05257463,
      "balance_loss_mlp": 1.02043176,
      "epoch": 0.3738351469969338,
      "flos": 34678108483200.0,
      "grad_norm": 1.961512812219949,
      "language_loss": 0.63319892,
      "learning_rate": 2.882975981868313e-06,
      "loss": 0.6551013,
      "num_input_tokens_seen": 66795525,
      "step": 3109,
      "time_per_iteration": 4.175340414047241
    },
    {
      "auxiliary_loss_clip": 0.01168867,
      "auxiliary_loss_mlp": 0.01033556,
      "balance_loss_clip": 1.05434895,
      "balance_loss_mlp": 1.02533686,
      "epoch": 0.3739553898875729,
      "flos": 43507967448960.0,
      "grad_norm": 2.7617806843406574,
      "language_loss": 0.68699944,
      "learning_rate": 2.882276969236016e-06,
      "loss": 0.70902371,
      "num_input_tokens_seen": 66816885,
      "step": 3110,
      "time_per_iteration": 3.542484760284424
    },
    {
      "auxiliary_loss_clip": 0.01152661,
      "auxiliary_loss_mlp": 0.01027832,
      "balance_loss_clip": 1.05029416,
      "balance_loss_mlp": 1.01970768,
      "epoch": 0.374075632778212,
      "flos": 12856487448960.0,
      "grad_norm": 2.0019563037806583,
      "language_loss": 0.76665866,
      "learning_rate": 2.881577822763005e-06,
      "loss": 0.78846359,
      "num_input_tokens_seen": 66834835,
      "step": 3111,
      "time_per_iteration": 2.5261940956115723
    },
    {
      "auxiliary_loss_clip": 0.01168773,
      "auxiliary_loss_mlp": 0.01029592,
      "balance_loss_clip": 1.05223024,
      "balance_loss_mlp": 1.02170062,
      "epoch": 0.3741958756688511,
      "flos": 26024031699840.0,
      "grad_norm": 1.7818054056493005,
      "language_loss": 0.87562221,
      "learning_rate": 2.880878542555338e-06,
      "loss": 0.89760584,
      "num_input_tokens_seen": 66852600,
      "step": 3112,
      "time_per_iteration": 2.550283670425415
    },
    {
      "auxiliary_loss_clip": 0.011835,
      "auxiliary_loss_mlp": 0.01039905,
      "balance_loss_clip": 1.05292261,
      "balance_loss_mlp": 1.03148293,
      "epoch": 0.37431611855949015,
      "flos": 21433894652160.0,
      "grad_norm": 1.9655739173930757,
      "language_loss": 0.80952716,
      "learning_rate": 2.8801791287190976e-06,
      "loss": 0.83176112,
      "num_input_tokens_seen": 66870595,
      "step": 3113,
      "time_per_iteration": 2.4860949516296387
    },
    {
      "auxiliary_loss_clip": 0.01171419,
      "auxiliary_loss_mlp": 0.01029833,
      "balance_loss_clip": 1.05272532,
      "balance_loss_mlp": 1.02154779,
      "epoch": 0.37443636145012926,
      "flos": 24207096090240.0,
      "grad_norm": 2.5747255725972837,
      "language_loss": 0.86600751,
      "learning_rate": 2.8794795813603817e-06,
      "loss": 0.88802004,
      "num_input_tokens_seen": 66886060,
      "step": 3114,
      "time_per_iteration": 2.529881000518799
    },
    {
      "auxiliary_loss_clip": 0.01173194,
      "auxiliary_loss_mlp": 0.01035831,
      "balance_loss_clip": 1.05213451,
      "balance_loss_mlp": 1.02775419,
      "epoch": 0.3745566043407684,
      "flos": 15378601841280.0,
      "grad_norm": 1.6795689400736182,
      "language_loss": 0.81589979,
      "learning_rate": 2.878779900585314e-06,
      "loss": 0.83799005,
      "num_input_tokens_seen": 66903900,
      "step": 3115,
      "time_per_iteration": 2.4610259532928467
    },
    {
      "auxiliary_loss_clip": 0.01162878,
      "auxiliary_loss_mlp": 0.01030351,
      "balance_loss_clip": 1.05343223,
      "balance_loss_mlp": 1.02321661,
      "epoch": 0.37467684723140743,
      "flos": 24608218245120.0,
      "grad_norm": 1.6029659177132476,
      "language_loss": 0.754053,
      "learning_rate": 2.8780800865000336e-06,
      "loss": 0.7759853,
      "num_input_tokens_seen": 66925210,
      "step": 3116,
      "time_per_iteration": 2.5944268703460693
    },
    {
      "auxiliary_loss_clip": 0.010672,
      "auxiliary_loss_mlp": 0.01003033,
      "balance_loss_clip": 1.0180819,
      "balance_loss_mlp": 1.00154924,
      "epoch": 0.37479709012204654,
      "flos": 64377491610240.0,
      "grad_norm": 0.9762708163580068,
      "language_loss": 0.59219861,
      "learning_rate": 2.877380139210702e-06,
      "loss": 0.61290085,
      "num_input_tokens_seen": 66983880,
      "step": 3117,
      "time_per_iteration": 3.058168411254883
    },
    {
      "auxiliary_loss_clip": 0.01143135,
      "auxiliary_loss_mlp": 0.01029084,
      "balance_loss_clip": 1.05072784,
      "balance_loss_mlp": 1.02097225,
      "epoch": 0.37491733301268565,
      "flos": 23803962773760.0,
      "grad_norm": 1.6924516172608204,
      "language_loss": 0.7626003,
      "learning_rate": 2.876680058823501e-06,
      "loss": 0.7843225,
      "num_input_tokens_seen": 67004280,
      "step": 3118,
      "time_per_iteration": 2.5937039852142334
    },
    {
      "auxiliary_loss_clip": 0.01146646,
      "auxiliary_loss_mlp": 0.01026606,
      "balance_loss_clip": 1.04993105,
      "balance_loss_mlp": 1.01737356,
      "epoch": 0.3750375759033247,
      "flos": 32160950167680.0,
      "grad_norm": 1.8492100168528574,
      "language_loss": 0.66320944,
      "learning_rate": 2.8759798454446314e-06,
      "loss": 0.68494201,
      "num_input_tokens_seen": 67027445,
      "step": 3119,
      "time_per_iteration": 2.627420425415039
    },
    {
      "auxiliary_loss_clip": 0.01173238,
      "auxiliary_loss_mlp": 0.01030997,
      "balance_loss_clip": 1.05419493,
      "balance_loss_mlp": 1.02276576,
      "epoch": 0.3751578187939638,
      "flos": 23367791923200.0,
      "grad_norm": 1.7829956234710356,
      "language_loss": 0.81399703,
      "learning_rate": 2.8752794991803173e-06,
      "loss": 0.83603942,
      "num_input_tokens_seen": 67045130,
      "step": 3120,
      "time_per_iteration": 2.532078504562378
    },
    {
      "auxiliary_loss_clip": 0.01156024,
      "auxiliary_loss_mlp": 0.01031941,
      "balance_loss_clip": 1.05477691,
      "balance_loss_mlp": 1.02467489,
      "epoch": 0.37527806168460287,
      "flos": 14605731878400.0,
      "grad_norm": 1.8337930466584658,
      "language_loss": 0.75310308,
      "learning_rate": 2.8745790201367976e-06,
      "loss": 0.77498275,
      "num_input_tokens_seen": 67060885,
      "step": 3121,
      "time_per_iteration": 2.5081212520599365
    },
    {
      "auxiliary_loss_clip": 0.01186766,
      "auxiliary_loss_mlp": 0.01031347,
      "balance_loss_clip": 1.05467927,
      "balance_loss_mlp": 1.02240062,
      "epoch": 0.375398304575242,
      "flos": 26390823431040.0,
      "grad_norm": 1.9795001321221324,
      "language_loss": 0.84083986,
      "learning_rate": 2.8738784084203373e-06,
      "loss": 0.86302096,
      "num_input_tokens_seen": 67080960,
      "step": 3122,
      "time_per_iteration": 2.533269166946411
    },
    {
      "auxiliary_loss_clip": 0.01149735,
      "auxiliary_loss_mlp": 0.01029931,
      "balance_loss_clip": 1.04930878,
      "balance_loss_mlp": 1.02229559,
      "epoch": 0.3755185474658811,
      "flos": 22236605838720.0,
      "grad_norm": 1.714503921975435,
      "language_loss": 0.78795302,
      "learning_rate": 2.873177664137216e-06,
      "loss": 0.80974972,
      "num_input_tokens_seen": 67101890,
      "step": 3123,
      "time_per_iteration": 2.57914400100708
    },
    {
      "auxiliary_loss_clip": 0.01138866,
      "auxiliary_loss_mlp": 0.01025507,
      "balance_loss_clip": 1.05187714,
      "balance_loss_mlp": 1.01778793,
      "epoch": 0.37563879035652015,
      "flos": 30812935633920.0,
      "grad_norm": 2.233959996422891,
      "language_loss": 0.69058216,
      "learning_rate": 2.8724767873937384e-06,
      "loss": 0.71222591,
      "num_input_tokens_seen": 67126010,
      "step": 3124,
      "time_per_iteration": 2.6755900382995605
    },
    {
      "auxiliary_loss_clip": 0.01155075,
      "auxiliary_loss_mlp": 0.01024807,
      "balance_loss_clip": 1.05157638,
      "balance_loss_mlp": 1.0170821,
      "epoch": 0.37575903324715926,
      "flos": 20773533064320.0,
      "grad_norm": 1.9348775240610623,
      "language_loss": 0.87052363,
      "learning_rate": 2.871775778296225e-06,
      "loss": 0.89232242,
      "num_input_tokens_seen": 67143100,
      "step": 3125,
      "time_per_iteration": 2.540315628051758
    },
    {
      "auxiliary_loss_clip": 0.01171018,
      "auxiliary_loss_mlp": 0.01028914,
      "balance_loss_clip": 1.05548894,
      "balance_loss_mlp": 1.02074194,
      "epoch": 0.37587927613779837,
      "flos": 18697681244160.0,
      "grad_norm": 2.045301713367826,
      "language_loss": 0.78424966,
      "learning_rate": 2.8710746369510196e-06,
      "loss": 0.80624896,
      "num_input_tokens_seen": 67161085,
      "step": 3126,
      "time_per_iteration": 2.526965856552124
    },
    {
      "auxiliary_loss_clip": 0.01151032,
      "auxiliary_loss_mlp": 0.01033561,
      "balance_loss_clip": 1.0509851,
      "balance_loss_mlp": 1.0252465,
      "epoch": 0.3759995190284374,
      "flos": 13624796384640.0,
      "grad_norm": 2.3746909476439964,
      "language_loss": 0.83908075,
      "learning_rate": 2.8703733634644846e-06,
      "loss": 0.86092669,
      "num_input_tokens_seen": 67175840,
      "step": 3127,
      "time_per_iteration": 2.510077953338623
    },
    {
      "auxiliary_loss_clip": 0.01184159,
      "auxiliary_loss_mlp": 0.01024949,
      "balance_loss_clip": 1.05629849,
      "balance_loss_mlp": 1.01733708,
      "epoch": 0.37611976191907653,
      "flos": 20484847457280.0,
      "grad_norm": 1.6197542842097143,
      "language_loss": 0.78890401,
      "learning_rate": 2.869671957943002e-06,
      "loss": 0.81099504,
      "num_input_tokens_seen": 67194995,
      "step": 3128,
      "time_per_iteration": 2.5099806785583496
    },
    {
      "auxiliary_loss_clip": 0.01154343,
      "auxiliary_loss_mlp": 0.01036077,
      "balance_loss_clip": 1.05813646,
      "balance_loss_mlp": 1.0283339,
      "epoch": 0.37624000480971564,
      "flos": 21141797253120.0,
      "grad_norm": 1.7836061818240865,
      "language_loss": 0.74326438,
      "learning_rate": 2.8689704204929747e-06,
      "loss": 0.76516855,
      "num_input_tokens_seen": 67214175,
      "step": 3129,
      "time_per_iteration": 2.5851807594299316
    },
    {
      "auxiliary_loss_clip": 0.01186337,
      "auxiliary_loss_mlp": 0.01034023,
      "balance_loss_clip": 1.05537009,
      "balance_loss_mlp": 1.02568424,
      "epoch": 0.3763602477003547,
      "flos": 22564470205440.0,
      "grad_norm": 11.064369512615885,
      "language_loss": 0.81097388,
      "learning_rate": 2.8682687512208253e-06,
      "loss": 0.83317745,
      "num_input_tokens_seen": 67233185,
      "step": 3130,
      "time_per_iteration": 3.2288777828216553
    },
    {
      "auxiliary_loss_clip": 0.01177207,
      "auxiliary_loss_mlp": 0.01031081,
      "balance_loss_clip": 1.05496669,
      "balance_loss_mlp": 1.0233686,
      "epoch": 0.3764804905909938,
      "flos": 27526857851520.0,
      "grad_norm": 2.0343474064386537,
      "language_loss": 0.80458796,
      "learning_rate": 2.8675669502329972e-06,
      "loss": 0.82667077,
      "num_input_tokens_seen": 67254715,
      "step": 3131,
      "time_per_iteration": 2.5694735050201416
    },
    {
      "auxiliary_loss_clip": 0.01166704,
      "auxiliary_loss_mlp": 0.00761321,
      "balance_loss_clip": 1.05205703,
      "balance_loss_mlp": 1.00027251,
      "epoch": 0.3766007334816329,
      "flos": 22528092706560.0,
      "grad_norm": 2.343843029903865,
      "language_loss": 0.86289328,
      "learning_rate": 2.866865017635952e-06,
      "loss": 0.88217354,
      "num_input_tokens_seen": 67272535,
      "step": 3132,
      "time_per_iteration": 2.5360028743743896
    },
    {
      "auxiliary_loss_clip": 0.01139843,
      "auxiliary_loss_mlp": 0.01032512,
      "balance_loss_clip": 1.05374122,
      "balance_loss_mlp": 1.02442348,
      "epoch": 0.376720976372272,
      "flos": 25957166532480.0,
      "grad_norm": 1.554909453421561,
      "language_loss": 0.7903167,
      "learning_rate": 2.866162953536174e-06,
      "loss": 0.81204021,
      "num_input_tokens_seen": 67293505,
      "step": 3133,
      "time_per_iteration": 2.617272138595581
    },
    {
      "auxiliary_loss_clip": 0.01154852,
      "auxiliary_loss_mlp": 0.00761359,
      "balance_loss_clip": 1.05102825,
      "balance_loss_mlp": 1.00031066,
      "epoch": 0.3768412192629111,
      "flos": 18041162411520.0,
      "grad_norm": 1.6378832987875838,
      "language_loss": 0.75217307,
      "learning_rate": 2.8654607580401634e-06,
      "loss": 0.77133518,
      "num_input_tokens_seen": 67313240,
      "step": 3134,
      "time_per_iteration": 2.568263292312622
    },
    {
      "auxiliary_loss_clip": 0.0106874,
      "auxiliary_loss_mlp": 0.01011893,
      "balance_loss_clip": 1.0207119,
      "balance_loss_mlp": 1.01042068,
      "epoch": 0.3769614621535502,
      "flos": 62989472304000.0,
      "grad_norm": 0.8797373320484281,
      "language_loss": 0.65255582,
      "learning_rate": 2.8647584312544446e-06,
      "loss": 0.67336214,
      "num_input_tokens_seen": 67378445,
      "step": 3135,
      "time_per_iteration": 3.9317147731781006
    },
    {
      "auxiliary_loss_clip": 0.01135453,
      "auxiliary_loss_mlp": 0.00760963,
      "balance_loss_clip": 1.04708123,
      "balance_loss_mlp": 1.00028825,
      "epoch": 0.37708170504418925,
      "flos": 23661685002240.0,
      "grad_norm": 1.3767434421482694,
      "language_loss": 0.85220915,
      "learning_rate": 2.864055973285559e-06,
      "loss": 0.87117332,
      "num_input_tokens_seen": 67400445,
      "step": 3136,
      "time_per_iteration": 3.4891505241394043
    },
    {
      "auxiliary_loss_clip": 0.01144211,
      "auxiliary_loss_mlp": 0.01032361,
      "balance_loss_clip": 1.04866123,
      "balance_loss_mlp": 1.02457094,
      "epoch": 0.37720194793482836,
      "flos": 24423170353920.0,
      "grad_norm": 1.8008835842251847,
      "language_loss": 0.86144984,
      "learning_rate": 2.8633533842400698e-06,
      "loss": 0.88321561,
      "num_input_tokens_seen": 67420645,
      "step": 3137,
      "time_per_iteration": 2.5921103954315186
    },
    {
      "auxiliary_loss_clip": 0.01174223,
      "auxiliary_loss_mlp": 0.00762225,
      "balance_loss_clip": 1.05582714,
      "balance_loss_mlp": 1.00029254,
      "epoch": 0.3773221908254674,
      "flos": 20996502739200.0,
      "grad_norm": 1.7841484916481138,
      "language_loss": 0.77574736,
      "learning_rate": 2.862650664224558e-06,
      "loss": 0.79511189,
      "num_input_tokens_seen": 67439495,
      "step": 3138,
      "time_per_iteration": 2.5838611125946045
    },
    {
      "auxiliary_loss_clip": 0.01171531,
      "auxiliary_loss_mlp": 0.01031878,
      "balance_loss_clip": 1.05891132,
      "balance_loss_mlp": 1.02420425,
      "epoch": 0.37744243371610653,
      "flos": 37631724958080.0,
      "grad_norm": 1.4905736616524992,
      "language_loss": 0.69560575,
      "learning_rate": 2.861947813345627e-06,
      "loss": 0.7176398,
      "num_input_tokens_seen": 67462195,
      "step": 3139,
      "time_per_iteration": 2.7531089782714844
    },
    {
      "auxiliary_loss_clip": 0.01186575,
      "auxiliary_loss_mlp": 0.00761804,
      "balance_loss_clip": 1.05543816,
      "balance_loss_mlp": 1.00029206,
      "epoch": 0.37756267660674564,
      "flos": 26140526484480.0,
      "grad_norm": 2.4100280901583724,
      "language_loss": 0.72835028,
      "learning_rate": 2.8612448317098974e-06,
      "loss": 0.74783409,
      "num_input_tokens_seen": 67482530,
      "step": 3140,
      "time_per_iteration": 2.6196014881134033
    },
    {
      "auxiliary_loss_clip": 0.01145621,
      "auxiliary_loss_mlp": 0.00762222,
      "balance_loss_clip": 1.05089688,
      "balance_loss_mlp": 1.00024819,
      "epoch": 0.3776829194973847,
      "flos": 19427888828160.0,
      "grad_norm": 2.092315608501026,
      "language_loss": 0.82962966,
      "learning_rate": 2.8605417194240114e-06,
      "loss": 0.84870809,
      "num_input_tokens_seen": 67500890,
      "step": 3141,
      "time_per_iteration": 2.6539828777313232
    },
    {
      "auxiliary_loss_clip": 0.01165856,
      "auxiliary_loss_mlp": 0.01023981,
      "balance_loss_clip": 1.0528605,
      "balance_loss_mlp": 1.01630998,
      "epoch": 0.3778031623880238,
      "flos": 17382309194880.0,
      "grad_norm": 1.743738776926079,
      "language_loss": 0.78767085,
      "learning_rate": 2.8598384765946315e-06,
      "loss": 0.80956924,
      "num_input_tokens_seen": 67519545,
      "step": 3142,
      "time_per_iteration": 2.6098129749298096
    },
    {
      "auxiliary_loss_clip": 0.01181392,
      "auxiliary_loss_mlp": 0.01022926,
      "balance_loss_clip": 1.05155158,
      "balance_loss_mlp": 1.01551092,
      "epoch": 0.3779234052786629,
      "flos": 27125843437440.0,
      "grad_norm": 1.7569908574497826,
      "language_loss": 0.7130307,
      "learning_rate": 2.8591351033284377e-06,
      "loss": 0.7350738,
      "num_input_tokens_seen": 67539275,
      "step": 3143,
      "time_per_iteration": 2.5902416706085205
    },
    {
      "auxiliary_loss_clip": 0.01170619,
      "auxiliary_loss_mlp": 0.01024386,
      "balance_loss_clip": 1.05182374,
      "balance_loss_mlp": 1.01644707,
      "epoch": 0.37804364816930197,
      "flos": 19682639061120.0,
      "grad_norm": 1.990746806410882,
      "language_loss": 0.83823824,
      "learning_rate": 2.8584315997321325e-06,
      "loss": 0.86018836,
      "num_input_tokens_seen": 67558280,
      "step": 3144,
      "time_per_iteration": 2.5905869007110596
    },
    {
      "auxiliary_loss_clip": 0.01185299,
      "auxiliary_loss_mlp": 0.0103195,
      "balance_loss_clip": 1.055444,
      "balance_loss_mlp": 1.02391505,
      "epoch": 0.3781638910599411,
      "flos": 22702905221760.0,
      "grad_norm": 2.2653390669488354,
      "language_loss": 0.7767424,
      "learning_rate": 2.8577279659124356e-06,
      "loss": 0.79891491,
      "num_input_tokens_seen": 67575955,
      "step": 3145,
      "time_per_iteration": 2.5280532836914062
    },
    {
      "auxiliary_loss_clip": 0.01167693,
      "auxiliary_loss_mlp": 0.01027801,
      "balance_loss_clip": 1.05249465,
      "balance_loss_mlp": 1.02066064,
      "epoch": 0.3782841339505802,
      "flos": 14647604158080.0,
      "grad_norm": 1.7749965364269122,
      "language_loss": 0.83329207,
      "learning_rate": 2.857024201976089e-06,
      "loss": 0.85524702,
      "num_input_tokens_seen": 67593515,
      "step": 3146,
      "time_per_iteration": 2.5365240573883057
    },
    {
      "auxiliary_loss_clip": 0.01151806,
      "auxiliary_loss_mlp": 0.01026084,
      "balance_loss_clip": 1.05099189,
      "balance_loss_mlp": 1.01743531,
      "epoch": 0.37840437684121925,
      "flos": 32818223185920.0,
      "grad_norm": 1.889390990166829,
      "language_loss": 0.72530037,
      "learning_rate": 2.8563203080298516e-06,
      "loss": 0.74707925,
      "num_input_tokens_seen": 67614290,
      "step": 3147,
      "time_per_iteration": 2.6815083026885986
    },
    {
      "auxiliary_loss_clip": 0.01156595,
      "auxiliary_loss_mlp": 0.00761299,
      "balance_loss_clip": 1.05404425,
      "balance_loss_mlp": 1.0002352,
      "epoch": 0.37852461973185836,
      "flos": 18369206346240.0,
      "grad_norm": 2.0786215497717646,
      "language_loss": 0.89359266,
      "learning_rate": 2.855616284180505e-06,
      "loss": 0.91277152,
      "num_input_tokens_seen": 67631340,
      "step": 3148,
      "time_per_iteration": 2.553753137588501
    },
    {
      "auxiliary_loss_clip": 0.01068229,
      "auxiliary_loss_mlp": 0.01003789,
      "balance_loss_clip": 1.01840603,
      "balance_loss_mlp": 1.0024178,
      "epoch": 0.37864486262249747,
      "flos": 59500680117120.0,
      "grad_norm": 0.8774801176984057,
      "language_loss": 0.66142684,
      "learning_rate": 2.8549121305348477e-06,
      "loss": 0.68214703,
      "num_input_tokens_seen": 67691125,
      "step": 3149,
      "time_per_iteration": 3.1097583770751953
    },
    {
      "auxiliary_loss_clip": 0.0117071,
      "auxiliary_loss_mlp": 0.01025978,
      "balance_loss_clip": 1.05391324,
      "balance_loss_mlp": 1.01875949,
      "epoch": 0.3787651055131365,
      "flos": 23363015414400.0,
      "grad_norm": 2.524443136308612,
      "language_loss": 0.83408594,
      "learning_rate": 2.8542078471997006e-06,
      "loss": 0.85605282,
      "num_input_tokens_seen": 67708740,
      "step": 3150,
      "time_per_iteration": 2.557471752166748
    },
    {
      "auxiliary_loss_clip": 0.01172112,
      "auxiliary_loss_mlp": 0.01026567,
      "balance_loss_clip": 1.05404723,
      "balance_loss_mlp": 1.01880097,
      "epoch": 0.37888534840377563,
      "flos": 24601394661120.0,
      "grad_norm": 1.7804519151607685,
      "language_loss": 0.75692034,
      "learning_rate": 2.8535034342819013e-06,
      "loss": 0.77890718,
      "num_input_tokens_seen": 67726150,
      "step": 3151,
      "time_per_iteration": 2.5591392517089844
    },
    {
      "auxiliary_loss_clip": 0.01182957,
      "auxiliary_loss_mlp": 0.0102993,
      "balance_loss_clip": 1.05559611,
      "balance_loss_mlp": 1.02227664,
      "epoch": 0.37900559129441475,
      "flos": 23986891762560.0,
      "grad_norm": 1.7389360927010975,
      "language_loss": 0.72433835,
      "learning_rate": 2.85279889188831e-06,
      "loss": 0.74646717,
      "num_input_tokens_seen": 67746525,
      "step": 3152,
      "time_per_iteration": 2.5101706981658936
    },
    {
      "auxiliary_loss_clip": 0.01136632,
      "auxiliary_loss_mlp": 0.01029365,
      "balance_loss_clip": 1.04517746,
      "balance_loss_mlp": 1.02178943,
      "epoch": 0.3791258341850538,
      "flos": 24644667571200.0,
      "grad_norm": 1.8769795387577275,
      "language_loss": 0.81059456,
      "learning_rate": 2.852094220125805e-06,
      "loss": 0.83225453,
      "num_input_tokens_seen": 67766035,
      "step": 3153,
      "time_per_iteration": 2.642052173614502
    },
    {
      "auxiliary_loss_clip": 0.01171219,
      "auxiliary_loss_mlp": 0.01029597,
      "balance_loss_clip": 1.05546272,
      "balance_loss_mlp": 1.02172875,
      "epoch": 0.3792460770756929,
      "flos": 17420841509760.0,
      "grad_norm": 2.05260689234353,
      "language_loss": 0.71095759,
      "learning_rate": 2.8513894191012846e-06,
      "loss": 0.73296577,
      "num_input_tokens_seen": 67785015,
      "step": 3154,
      "time_per_iteration": 2.537792682647705
    },
    {
      "auxiliary_loss_clip": 0.01184139,
      "auxiliary_loss_mlp": 0.01032792,
      "balance_loss_clip": 1.05467153,
      "balance_loss_mlp": 1.02552056,
      "epoch": 0.37936631996633197,
      "flos": 24206557386240.0,
      "grad_norm": 1.562213547718513,
      "language_loss": 0.78750598,
      "learning_rate": 2.8506844889216664e-06,
      "loss": 0.80967534,
      "num_input_tokens_seen": 67804400,
      "step": 3155,
      "time_per_iteration": 2.525606155395508
    },
    {
      "auxiliary_loss_clip": 0.01077716,
      "auxiliary_loss_mlp": 0.01009999,
      "balance_loss_clip": 1.03342938,
      "balance_loss_mlp": 1.00852704,
      "epoch": 0.3794865628569711,
      "flos": 70297114752000.0,
      "grad_norm": 0.8701397599867261,
      "language_loss": 0.62858891,
      "learning_rate": 2.849979429693887e-06,
      "loss": 0.64946616,
      "num_input_tokens_seen": 67865385,
      "step": 3156,
      "time_per_iteration": 3.936729907989502
    },
    {
      "auxiliary_loss_clip": 0.01182027,
      "auxiliary_loss_mlp": 0.01031431,
      "balance_loss_clip": 1.05470777,
      "balance_loss_mlp": 1.02356887,
      "epoch": 0.3796068057476102,
      "flos": 15779364860160.0,
      "grad_norm": 2.1041560094794707,
      "language_loss": 0.74110687,
      "learning_rate": 2.8492742415249042e-06,
      "loss": 0.76324141,
      "num_input_tokens_seen": 67883030,
      "step": 3157,
      "time_per_iteration": 2.471731662750244
    },
    {
      "auxiliary_loss_clip": 0.01180051,
      "auxiliary_loss_mlp": 0.01028495,
      "balance_loss_clip": 1.05170882,
      "balance_loss_mlp": 1.02069926,
      "epoch": 0.37972704863824924,
      "flos": 25191694771200.0,
      "grad_norm": 1.557968926099854,
      "language_loss": 0.76303136,
      "learning_rate": 2.848568924521694e-06,
      "loss": 0.78511685,
      "num_input_tokens_seen": 67903810,
      "step": 3158,
      "time_per_iteration": 2.542872905731201
    },
    {
      "auxiliary_loss_clip": 0.01161862,
      "auxiliary_loss_mlp": 0.01035443,
      "balance_loss_clip": 1.04961884,
      "balance_loss_mlp": 1.0278914,
      "epoch": 0.37984729152888835,
      "flos": 26210372480640.0,
      "grad_norm": 2.2822294027474612,
      "language_loss": 0.73342919,
      "learning_rate": 2.8478634787912526e-06,
      "loss": 0.75540233,
      "num_input_tokens_seen": 67921865,
      "step": 3159,
      "time_per_iteration": 2.5565409660339355
    },
    {
      "auxiliary_loss_clip": 0.01165177,
      "auxiliary_loss_mlp": 0.01031415,
      "balance_loss_clip": 1.0499506,
      "balance_loss_mlp": 1.02365148,
      "epoch": 0.37996753441952746,
      "flos": 25629302165760.0,
      "grad_norm": 3.5289695506326577,
      "language_loss": 0.77036244,
      "learning_rate": 2.847157904440596e-06,
      "loss": 0.79232836,
      "num_input_tokens_seen": 67941595,
      "step": 3160,
      "time_per_iteration": 2.5633418560028076
    },
    {
      "auxiliary_loss_clip": 0.0116577,
      "auxiliary_loss_mlp": 0.01028591,
      "balance_loss_clip": 1.05259085,
      "balance_loss_mlp": 1.02106929,
      "epoch": 0.3800877773101665,
      "flos": 20118414862080.0,
      "grad_norm": 1.5344910303159796,
      "language_loss": 0.73976779,
      "learning_rate": 2.846452201576759e-06,
      "loss": 0.76171136,
      "num_input_tokens_seen": 67960970,
      "step": 3161,
      "time_per_iteration": 3.34928297996521
    },
    {
      "auxiliary_loss_clip": 0.01057928,
      "auxiliary_loss_mlp": 0.01007916,
      "balance_loss_clip": 1.01784778,
      "balance_loss_mlp": 1.00636625,
      "epoch": 0.38020802020080563,
      "flos": 63053608037760.0,
      "grad_norm": 0.8410393461646315,
      "language_loss": 0.62780994,
      "learning_rate": 2.845746370306795e-06,
      "loss": 0.64846843,
      "num_input_tokens_seen": 68026160,
      "step": 3162,
      "time_per_iteration": 4.092927932739258
    },
    {
      "auxiliary_loss_clip": 0.01167163,
      "auxiliary_loss_mlp": 0.01028534,
      "balance_loss_clip": 1.05101895,
      "balance_loss_mlp": 1.0214355,
      "epoch": 0.38032826309144474,
      "flos": 21288420570240.0,
      "grad_norm": 2.008761971764891,
      "language_loss": 0.78544438,
      "learning_rate": 2.84504041073778e-06,
      "loss": 0.8074013,
      "num_input_tokens_seen": 68044575,
      "step": 3163,
      "time_per_iteration": 2.5568647384643555
    },
    {
      "auxiliary_loss_clip": 0.01148688,
      "auxiliary_loss_mlp": 0.01034712,
      "balance_loss_clip": 1.05163682,
      "balance_loss_mlp": 1.02665091,
      "epoch": 0.3804485059820838,
      "flos": 18954119416320.0,
      "grad_norm": 1.692562232449905,
      "language_loss": 0.79089928,
      "learning_rate": 2.844334322976806e-06,
      "loss": 0.81273335,
      "num_input_tokens_seen": 68064790,
      "step": 3164,
      "time_per_iteration": 2.578557014465332
    },
    {
      "auxiliary_loss_clip": 0.0112749,
      "auxiliary_loss_mlp": 0.01028563,
      "balance_loss_clip": 1.04561555,
      "balance_loss_mlp": 1.01998568,
      "epoch": 0.3805687488727229,
      "flos": 21833759831040.0,
      "grad_norm": 1.9973479501681453,
      "language_loss": 0.83172077,
      "learning_rate": 2.8436281071309866e-06,
      "loss": 0.85328132,
      "num_input_tokens_seen": 68083330,
      "step": 3165,
      "time_per_iteration": 2.6526529788970947
    },
    {
      "auxiliary_loss_clip": 0.01035761,
      "auxiliary_loss_mlp": 0.01011579,
      "balance_loss_clip": 1.01473498,
      "balance_loss_mlp": 1.01007128,
      "epoch": 0.380688991763362,
      "flos": 58546209968640.0,
      "grad_norm": 0.7328859531460182,
      "language_loss": 0.52987236,
      "learning_rate": 2.842921763307455e-06,
      "loss": 0.55034578,
      "num_input_tokens_seen": 68146140,
      "step": 3166,
      "time_per_iteration": 3.266592502593994
    },
    {
      "auxiliary_loss_clip": 0.01145744,
      "auxiliary_loss_mlp": 0.01027574,
      "balance_loss_clip": 1.04730248,
      "balance_loss_mlp": 1.01999271,
      "epoch": 0.38080923465400107,
      "flos": 23799509487360.0,
      "grad_norm": 2.3996404307369423,
      "language_loss": 0.82391667,
      "learning_rate": 2.842215291613361e-06,
      "loss": 0.84564984,
      "num_input_tokens_seen": 68164520,
      "step": 3167,
      "time_per_iteration": 2.7710354328155518
    },
    {
      "auxiliary_loss_clip": 0.01001178,
      "auxiliary_loss_mlp": 0.01010946,
      "balance_loss_clip": 1.0139327,
      "balance_loss_mlp": 1.00921714,
      "epoch": 0.3809294775446402,
      "flos": 54969866380800.0,
      "grad_norm": 0.8232133339993828,
      "language_loss": 0.59262639,
      "learning_rate": 2.8415086921558774e-06,
      "loss": 0.61274767,
      "num_input_tokens_seen": 68227945,
      "step": 3168,
      "time_per_iteration": 3.3661868572235107
    },
    {
      "auxiliary_loss_clip": 0.01138537,
      "auxiliary_loss_mlp": 0.01024192,
      "balance_loss_clip": 1.04443073,
      "balance_loss_mlp": 1.01689017,
      "epoch": 0.38104972043527924,
      "flos": 24643697904000.0,
      "grad_norm": 1.5857258567404975,
      "language_loss": 0.78783059,
      "learning_rate": 2.840801965042194e-06,
      "loss": 0.8094579,
      "num_input_tokens_seen": 68247405,
      "step": 3169,
      "time_per_iteration": 2.9885551929473877
    },
    {
      "auxiliary_loss_clip": 0.01143365,
      "auxiliary_loss_mlp": 0.01026002,
      "balance_loss_clip": 1.04579723,
      "balance_loss_mlp": 1.01838446,
      "epoch": 0.38116996332591835,
      "flos": 22856783086080.0,
      "grad_norm": 1.7506435320198142,
      "language_loss": 0.84105092,
      "learning_rate": 2.840095110379521e-06,
      "loss": 0.86274451,
      "num_input_tokens_seen": 68266925,
      "step": 3170,
      "time_per_iteration": 2.6298749446868896
    },
    {
      "auxiliary_loss_clip": 0.01028779,
      "auxiliary_loss_mlp": 0.0100342,
      "balance_loss_clip": 1.01491737,
      "balance_loss_mlp": 1.0021143,
      "epoch": 0.38129020621655746,
      "flos": 60836160804480.0,
      "grad_norm": 0.726946192976291,
      "language_loss": 0.53961122,
      "learning_rate": 2.8393881282750884e-06,
      "loss": 0.55993319,
      "num_input_tokens_seen": 68329755,
      "step": 3171,
      "time_per_iteration": 3.1572351455688477
    },
    {
      "auxiliary_loss_clip": 0.01150418,
      "auxiliary_loss_mlp": 0.01032318,
      "balance_loss_clip": 1.05018103,
      "balance_loss_mlp": 1.02405095,
      "epoch": 0.3814104491071965,
      "flos": 21648101408640.0,
      "grad_norm": 1.8497110440144855,
      "language_loss": 0.79083824,
      "learning_rate": 2.838681018836144e-06,
      "loss": 0.81266558,
      "num_input_tokens_seen": 68347075,
      "step": 3172,
      "time_per_iteration": 2.5744926929473877
    },
    {
      "auxiliary_loss_clip": 0.01140184,
      "auxiliary_loss_mlp": 0.00760483,
      "balance_loss_clip": 1.04757977,
      "balance_loss_mlp": 1.00022674,
      "epoch": 0.3815306919978356,
      "flos": 19099090707840.0,
      "grad_norm": 4.5010796943094435,
      "language_loss": 0.78067625,
      "learning_rate": 2.837973782169955e-06,
      "loss": 0.79968286,
      "num_input_tokens_seen": 68365450,
      "step": 3173,
      "time_per_iteration": 2.5888988971710205
    },
    {
      "auxiliary_loss_clip": 0.01073525,
      "auxiliary_loss_mlp": 0.01001866,
      "balance_loss_clip": 1.01537478,
      "balance_loss_mlp": 1.00045967,
      "epoch": 0.38165093488847474,
      "flos": 67067918156160.0,
      "grad_norm": 0.8058867931929685,
      "language_loss": 0.59184951,
      "learning_rate": 2.8372664183838096e-06,
      "loss": 0.61260343,
      "num_input_tokens_seen": 68428470,
      "step": 3174,
      "time_per_iteration": 3.1438207626342773
    },
    {
      "auxiliary_loss_clip": 0.0118087,
      "auxiliary_loss_mlp": 0.01030365,
      "balance_loss_clip": 1.05410421,
      "balance_loss_mlp": 1.0227474,
      "epoch": 0.3817711777791138,
      "flos": 22341105480960.0,
      "grad_norm": 2.2981575972290815,
      "language_loss": 0.68607616,
      "learning_rate": 2.836558927585015e-06,
      "loss": 0.70818853,
      "num_input_tokens_seen": 68445440,
      "step": 3175,
      "time_per_iteration": 2.5110583305358887
    },
    {
      "auxiliary_loss_clip": 0.01169841,
      "auxiliary_loss_mlp": 0.01032639,
      "balance_loss_clip": 1.05263937,
      "balance_loss_mlp": 1.02492046,
      "epoch": 0.3818914206697529,
      "flos": 22820621068800.0,
      "grad_norm": 1.741719237994178,
      "language_loss": 0.82579315,
      "learning_rate": 2.8358513098808957e-06,
      "loss": 0.84781796,
      "num_input_tokens_seen": 68465755,
      "step": 3176,
      "time_per_iteration": 2.563683271408081
    },
    {
      "auxiliary_loss_clip": 0.01114822,
      "auxiliary_loss_mlp": 0.01029367,
      "balance_loss_clip": 1.04664588,
      "balance_loss_mlp": 1.02146983,
      "epoch": 0.382011663560392,
      "flos": 24386074583040.0,
      "grad_norm": 1.7075545654893276,
      "language_loss": 0.76702464,
      "learning_rate": 2.835143565378798e-06,
      "loss": 0.78846645,
      "num_input_tokens_seen": 68486220,
      "step": 3177,
      "time_per_iteration": 2.6704649925231934
    },
    {
      "auxiliary_loss_clip": 0.0110571,
      "auxiliary_loss_mlp": 0.01025597,
      "balance_loss_clip": 1.04483175,
      "balance_loss_mlp": 1.01767302,
      "epoch": 0.38213190645103107,
      "flos": 21981568296960.0,
      "grad_norm": 1.770877751482208,
      "language_loss": 0.78048712,
      "learning_rate": 2.8344356941860847e-06,
      "loss": 0.80180019,
      "num_input_tokens_seen": 68505850,
      "step": 3178,
      "time_per_iteration": 2.68534779548645
    },
    {
      "auxiliary_loss_clip": 0.01137199,
      "auxiliary_loss_mlp": 0.01033743,
      "balance_loss_clip": 1.05034316,
      "balance_loss_mlp": 1.0259943,
      "epoch": 0.3822521493416702,
      "flos": 35516945773440.0,
      "grad_norm": 2.018592747069429,
      "language_loss": 0.65893912,
      "learning_rate": 2.8337276964101403e-06,
      "loss": 0.68064857,
      "num_input_tokens_seen": 68526290,
      "step": 3179,
      "time_per_iteration": 2.734175682067871
    },
    {
      "auxiliary_loss_clip": 0.01171084,
      "auxiliary_loss_mlp": 0.0103468,
      "balance_loss_clip": 1.05352879,
      "balance_loss_mlp": 1.02695203,
      "epoch": 0.3823723922323093,
      "flos": 21069904181760.0,
      "grad_norm": 1.961656341265008,
      "language_loss": 0.76465887,
      "learning_rate": 2.833019572158367e-06,
      "loss": 0.78671646,
      "num_input_tokens_seen": 68544725,
      "step": 3180,
      "time_per_iteration": 2.5567846298217773
    },
    {
      "auxiliary_loss_clip": 0.01153843,
      "auxiliary_loss_mlp": 0.01028496,
      "balance_loss_clip": 1.05212271,
      "balance_loss_mlp": 1.02107823,
      "epoch": 0.38249263512294834,
      "flos": 19789149864960.0,
      "grad_norm": 1.6560811282426453,
      "language_loss": 0.80065393,
      "learning_rate": 2.8323113215381872e-06,
      "loss": 0.82247734,
      "num_input_tokens_seen": 68563070,
      "step": 3181,
      "time_per_iteration": 3.447103261947632
    },
    {
      "auxiliary_loss_clip": 0.01139709,
      "auxiliary_loss_mlp": 0.01033404,
      "balance_loss_clip": 1.04897213,
      "balance_loss_mlp": 1.02495182,
      "epoch": 0.38261287801358745,
      "flos": 21433930565760.0,
      "grad_norm": 2.2063295582998776,
      "language_loss": 0.76183581,
      "learning_rate": 2.831602944657042e-06,
      "loss": 0.78356689,
      "num_input_tokens_seen": 68581150,
      "step": 3182,
      "time_per_iteration": 2.614678382873535
    },
    {
      "auxiliary_loss_clip": 0.01158063,
      "auxiliary_loss_mlp": 0.01031475,
      "balance_loss_clip": 1.05024314,
      "balance_loss_mlp": 1.02379775,
      "epoch": 0.38273312090422656,
      "flos": 21981568296960.0,
      "grad_norm": 4.2601465907195495,
      "language_loss": 0.74315912,
      "learning_rate": 2.830894441622391e-06,
      "loss": 0.76505446,
      "num_input_tokens_seen": 68597800,
      "step": 3183,
      "time_per_iteration": 2.627392053604126
    },
    {
      "auxiliary_loss_clip": 0.01134288,
      "auxiliary_loss_mlp": 0.00761488,
      "balance_loss_clip": 1.04383349,
      "balance_loss_mlp": 1.00021768,
      "epoch": 0.3828533637948656,
      "flos": 24790895838720.0,
      "grad_norm": 1.8478201634458247,
      "language_loss": 0.79997826,
      "learning_rate": 2.8301858125417134e-06,
      "loss": 0.81893599,
      "num_input_tokens_seen": 68617640,
      "step": 3184,
      "time_per_iteration": 2.6682589054107666
    },
    {
      "auxiliary_loss_clip": 0.01153812,
      "auxiliary_loss_mlp": 0.01025583,
      "balance_loss_clip": 1.05209708,
      "balance_loss_mlp": 1.01850498,
      "epoch": 0.38297360668550473,
      "flos": 22455445449600.0,
      "grad_norm": 1.7050549723848305,
      "language_loss": 0.74028885,
      "learning_rate": 2.8294770575225082e-06,
      "loss": 0.76208282,
      "num_input_tokens_seen": 68637770,
      "step": 3185,
      "time_per_iteration": 2.5887250900268555
    },
    {
      "auxiliary_loss_clip": 0.01169411,
      "auxiliary_loss_mlp": 0.01038444,
      "balance_loss_clip": 1.05421352,
      "balance_loss_mlp": 1.03052258,
      "epoch": 0.3830938495761438,
      "flos": 24896903852160.0,
      "grad_norm": 1.7173340416909162,
      "language_loss": 0.84119129,
      "learning_rate": 2.828768176672293e-06,
      "loss": 0.86326981,
      "num_input_tokens_seen": 68656885,
      "step": 3186,
      "time_per_iteration": 3.3614981174468994
    },
    {
      "auxiliary_loss_clip": 0.01136011,
      "auxiliary_loss_mlp": 0.01040471,
      "balance_loss_clip": 1.04660833,
      "balance_loss_mlp": 1.0320425,
      "epoch": 0.3832140924667829,
      "flos": 33036236784000.0,
      "grad_norm": 1.720422409927139,
      "language_loss": 0.71573353,
      "learning_rate": 2.8280591700986044e-06,
      "loss": 0.73749828,
      "num_input_tokens_seen": 68678750,
      "step": 3187,
      "time_per_iteration": 4.3570005893707275
    },
    {
      "auxiliary_loss_clip": 0.01156623,
      "auxiliary_loss_mlp": 0.01030804,
      "balance_loss_clip": 1.04885459,
      "balance_loss_mlp": 1.02306104,
      "epoch": 0.383334335357422,
      "flos": 31903721896320.0,
      "grad_norm": 6.630667151384367,
      "language_loss": 0.74690795,
      "learning_rate": 2.827350037908999e-06,
      "loss": 0.7687822,
      "num_input_tokens_seen": 68698190,
      "step": 3188,
      "time_per_iteration": 2.69952654838562
    },
    {
      "auxiliary_loss_clip": 0.01146119,
      "auxiliary_loss_mlp": 0.01029759,
      "balance_loss_clip": 1.04970288,
      "balance_loss_mlp": 1.02148008,
      "epoch": 0.38345457824806106,
      "flos": 19791915212160.0,
      "grad_norm": 2.051121564887161,
      "language_loss": 0.79186571,
      "learning_rate": 2.8266407802110496e-06,
      "loss": 0.81362444,
      "num_input_tokens_seen": 68716445,
      "step": 3189,
      "time_per_iteration": 2.5902163982391357
    },
    {
      "auxiliary_loss_clip": 0.01102982,
      "auxiliary_loss_mlp": 0.0103912,
      "balance_loss_clip": 1.0441494,
      "balance_loss_mlp": 1.0311625,
      "epoch": 0.3835748211387002,
      "flos": 22419391173120.0,
      "grad_norm": 2.259900841120351,
      "language_loss": 0.7592839,
      "learning_rate": 2.8259313971123515e-06,
      "loss": 0.78070486,
      "num_input_tokens_seen": 68737565,
      "step": 3190,
      "time_per_iteration": 2.743882894515991
    },
    {
      "auxiliary_loss_clip": 0.01169456,
      "auxiliary_loss_mlp": 0.0102449,
      "balance_loss_clip": 1.05671239,
      "balance_loss_mlp": 1.01698017,
      "epoch": 0.3836950640293393,
      "flos": 25118436983040.0,
      "grad_norm": 1.5002690176233273,
      "language_loss": 0.78207505,
      "learning_rate": 2.8252218887205166e-06,
      "loss": 0.8040145,
      "num_input_tokens_seen": 68758255,
      "step": 3191,
      "time_per_iteration": 2.5763626098632812
    },
    {
      "auxiliary_loss_clip": 0.01116226,
      "auxiliary_loss_mlp": 0.01033513,
      "balance_loss_clip": 1.04848003,
      "balance_loss_mlp": 1.02579975,
      "epoch": 0.38381530691997834,
      "flos": 21799213925760.0,
      "grad_norm": 1.7150154740770813,
      "language_loss": 0.80969393,
      "learning_rate": 2.824512255143178e-06,
      "loss": 0.8311913,
      "num_input_tokens_seen": 68777490,
      "step": 3192,
      "time_per_iteration": 2.7162322998046875
    },
    {
      "auxiliary_loss_clip": 0.01142962,
      "auxiliary_loss_mlp": 0.01026173,
      "balance_loss_clip": 1.04978836,
      "balance_loss_mlp": 1.0182333,
      "epoch": 0.38393554981061745,
      "flos": 21252689516160.0,
      "grad_norm": 1.6540776847310479,
      "language_loss": 0.79402709,
      "learning_rate": 2.8238024964879855e-06,
      "loss": 0.81571841,
      "num_input_tokens_seen": 68798385,
      "step": 3193,
      "time_per_iteration": 2.601266384124756
    },
    {
      "auxiliary_loss_clip": 0.01185461,
      "auxiliary_loss_mlp": 0.01028307,
      "balance_loss_clip": 1.05555105,
      "balance_loss_mlp": 1.02059126,
      "epoch": 0.38405579270125656,
      "flos": 17019360218880.0,
      "grad_norm": 2.616518730027013,
      "language_loss": 0.76894891,
      "learning_rate": 2.8230926128626095e-06,
      "loss": 0.79108661,
      "num_input_tokens_seen": 68816880,
      "step": 3194,
      "time_per_iteration": 2.513286828994751
    },
    {
      "auxiliary_loss_clip": 0.01147183,
      "auxiliary_loss_mlp": 0.01028218,
      "balance_loss_clip": 1.04752803,
      "balance_loss_mlp": 1.02023101,
      "epoch": 0.3841760355918956,
      "flos": 21835375943040.0,
      "grad_norm": 2.0993437529251673,
      "language_loss": 0.79327828,
      "learning_rate": 2.822382604374738e-06,
      "loss": 0.8150323,
      "num_input_tokens_seen": 68835805,
      "step": 3195,
      "time_per_iteration": 2.575206756591797
    },
    {
      "auxiliary_loss_clip": 0.01155034,
      "auxiliary_loss_mlp": 0.01030909,
      "balance_loss_clip": 1.05496407,
      "balance_loss_mlp": 1.02264762,
      "epoch": 0.3842962784825347,
      "flos": 25915114684800.0,
      "grad_norm": 2.3246083327926605,
      "language_loss": 0.6538856,
      "learning_rate": 2.8216724711320793e-06,
      "loss": 0.67574501,
      "num_input_tokens_seen": 68854930,
      "step": 3196,
      "time_per_iteration": 2.6148626804351807
    },
    {
      "auxiliary_loss_clip": 0.01180301,
      "auxiliary_loss_mlp": 0.00760917,
      "balance_loss_clip": 1.05277216,
      "balance_loss_mlp": 1.00021362,
      "epoch": 0.38441652137317384,
      "flos": 25337492075520.0,
      "grad_norm": 3.723461568977277,
      "language_loss": 0.79567111,
      "learning_rate": 2.820962213242361e-06,
      "loss": 0.81508327,
      "num_input_tokens_seen": 68874260,
      "step": 3197,
      "time_per_iteration": 2.5601460933685303
    },
    {
      "auxiliary_loss_clip": 0.01164267,
      "auxiliary_loss_mlp": 0.0103345,
      "balance_loss_clip": 1.05300593,
      "balance_loss_mlp": 1.02567172,
      "epoch": 0.3845367642638129,
      "flos": 18113486446080.0,
      "grad_norm": 2.346647241130003,
      "language_loss": 0.83936942,
      "learning_rate": 2.8202518308133264e-06,
      "loss": 0.86134648,
      "num_input_tokens_seen": 68891535,
      "step": 3198,
      "time_per_iteration": 2.6107635498046875
    },
    {
      "auxiliary_loss_clip": 0.01185154,
      "auxiliary_loss_mlp": 0.01030303,
      "balance_loss_clip": 1.05418682,
      "balance_loss_mlp": 1.02210116,
      "epoch": 0.384657007154452,
      "flos": 25228395492480.0,
      "grad_norm": 1.8170567036206642,
      "language_loss": 0.73287165,
      "learning_rate": 2.8195413239527426e-06,
      "loss": 0.75502616,
      "num_input_tokens_seen": 68911275,
      "step": 3199,
      "time_per_iteration": 2.535554885864258
    },
    {
      "auxiliary_loss_clip": 0.0116592,
      "auxiliary_loss_mlp": 0.0102866,
      "balance_loss_clip": 1.05254185,
      "balance_loss_mlp": 1.02111447,
      "epoch": 0.38477725004509106,
      "flos": 19865855358720.0,
      "grad_norm": 1.8280198223192647,
      "language_loss": 0.80225551,
      "learning_rate": 2.8188306927683906e-06,
      "loss": 0.82420135,
      "num_input_tokens_seen": 68930745,
      "step": 3200,
      "time_per_iteration": 2.5393307209014893
    },
    {
      "auxiliary_loss_clip": 0.01155892,
      "auxiliary_loss_mlp": 0.01030924,
      "balance_loss_clip": 1.05300486,
      "balance_loss_mlp": 1.02335417,
      "epoch": 0.38489749293573017,
      "flos": 18259391491200.0,
      "grad_norm": 1.8564501128206192,
      "language_loss": 0.7456764,
      "learning_rate": 2.818119937368074e-06,
      "loss": 0.76754463,
      "num_input_tokens_seen": 68949380,
      "step": 3201,
      "time_per_iteration": 2.580806255340576
    },
    {
      "auxiliary_loss_clip": 0.01173078,
      "auxiliary_loss_mlp": 0.01032357,
      "balance_loss_clip": 1.05266225,
      "balance_loss_mlp": 1.02407229,
      "epoch": 0.3850177358263693,
      "flos": 24389163152640.0,
      "grad_norm": 1.897645435154649,
      "language_loss": 0.65535355,
      "learning_rate": 2.817409057859613e-06,
      "loss": 0.67740786,
      "num_input_tokens_seen": 68968370,
      "step": 3202,
      "time_per_iteration": 2.560251474380493
    },
    {
      "auxiliary_loss_clip": 0.01122366,
      "auxiliary_loss_mlp": 0.01034227,
      "balance_loss_clip": 1.04767954,
      "balance_loss_mlp": 1.02495873,
      "epoch": 0.38513797871700833,
      "flos": 17671533505920.0,
      "grad_norm": 2.0679698832971645,
      "language_loss": 0.79169786,
      "learning_rate": 2.8166980543508482e-06,
      "loss": 0.81326377,
      "num_input_tokens_seen": 68984260,
      "step": 3203,
      "time_per_iteration": 2.6013872623443604
    },
    {
      "auxiliary_loss_clip": 0.01189083,
      "auxiliary_loss_mlp": 0.01033587,
      "balance_loss_clip": 1.05870807,
      "balance_loss_mlp": 1.02535534,
      "epoch": 0.38525822160764744,
      "flos": 25739583897600.0,
      "grad_norm": 1.70447955417919,
      "language_loss": 0.79579139,
      "learning_rate": 2.815986926949638e-06,
      "loss": 0.81801808,
      "num_input_tokens_seen": 69002760,
      "step": 3204,
      "time_per_iteration": 2.5315709114074707
    },
    {
      "auxiliary_loss_clip": 0.01168676,
      "auxiliary_loss_mlp": 0.01033028,
      "balance_loss_clip": 1.05433273,
      "balance_loss_mlp": 1.02514839,
      "epoch": 0.38537846449828655,
      "flos": 20193647898240.0,
      "grad_norm": 1.7709256279851064,
      "language_loss": 0.8012827,
      "learning_rate": 2.8152756757638597e-06,
      "loss": 0.82329977,
      "num_input_tokens_seen": 69021260,
      "step": 3205,
      "time_per_iteration": 2.5194461345672607
    },
    {
      "auxiliary_loss_clip": 0.01166757,
      "auxiliary_loss_mlp": 0.01030373,
      "balance_loss_clip": 1.05370855,
      "balance_loss_mlp": 1.02237988,
      "epoch": 0.3854987073889256,
      "flos": 23039352938880.0,
      "grad_norm": 1.9795042599300814,
      "language_loss": 0.84469151,
      "learning_rate": 2.8145643009014093e-06,
      "loss": 0.86666274,
      "num_input_tokens_seen": 69039755,
      "step": 3206,
      "time_per_iteration": 2.5414235591888428
    },
    {
      "auxiliary_loss_clip": 0.01171327,
      "auxiliary_loss_mlp": 0.01031845,
      "balance_loss_clip": 1.05381763,
      "balance_loss_mlp": 1.02462697,
      "epoch": 0.3856189502795647,
      "flos": 20190631155840.0,
      "grad_norm": 2.0862367526245915,
      "language_loss": 0.79215789,
      "learning_rate": 2.813852802470202e-06,
      "loss": 0.81418955,
      "num_input_tokens_seen": 69057650,
      "step": 3207,
      "time_per_iteration": 2.52268123626709
    },
    {
      "auxiliary_loss_clip": 0.01151802,
      "auxiliary_loss_mlp": 0.01029077,
      "balance_loss_clip": 1.05160427,
      "balance_loss_mlp": 1.02072012,
      "epoch": 0.38573919317020383,
      "flos": 25702631781120.0,
      "grad_norm": 1.6747224770050817,
      "language_loss": 0.72519374,
      "learning_rate": 2.8131411805781717e-06,
      "loss": 0.7470026,
      "num_input_tokens_seen": 69077775,
      "step": 3208,
      "time_per_iteration": 3.405616283416748
    },
    {
      "auxiliary_loss_clip": 0.01161074,
      "auxiliary_loss_mlp": 0.01032413,
      "balance_loss_clip": 1.05637848,
      "balance_loss_mlp": 1.02386558,
      "epoch": 0.3858594360608429,
      "flos": 29821405628160.0,
      "grad_norm": 2.134008584707854,
      "language_loss": 0.64308453,
      "learning_rate": 2.8124294353332707e-06,
      "loss": 0.66501939,
      "num_input_tokens_seen": 69096450,
      "step": 3209,
      "time_per_iteration": 2.6237499713897705
    },
    {
      "auxiliary_loss_clip": 0.01145433,
      "auxiliary_loss_mlp": 0.01028123,
      "balance_loss_clip": 1.05010819,
      "balance_loss_mlp": 1.02013612,
      "epoch": 0.385979678951482,
      "flos": 24790428961920.0,
      "grad_norm": 1.571109918653736,
      "language_loss": 0.77315277,
      "learning_rate": 2.8117175668434713e-06,
      "loss": 0.79488832,
      "num_input_tokens_seen": 69116110,
      "step": 3210,
      "time_per_iteration": 2.6237189769744873
    },
    {
      "auxiliary_loss_clip": 0.01185455,
      "auxiliary_loss_mlp": 0.01030286,
      "balance_loss_clip": 1.05648303,
      "balance_loss_mlp": 1.02172065,
      "epoch": 0.3860999218421211,
      "flos": 21287881866240.0,
      "grad_norm": 2.377936303948983,
      "language_loss": 0.6999588,
      "learning_rate": 2.811005575216762e-06,
      "loss": 0.72211623,
      "num_input_tokens_seen": 69134825,
      "step": 3211,
      "time_per_iteration": 2.5008528232574463
    },
    {
      "auxiliary_loss_clip": 0.01134204,
      "auxiliary_loss_mlp": 0.01031,
      "balance_loss_clip": 1.04786479,
      "balance_loss_mlp": 1.02306092,
      "epoch": 0.38622016473276016,
      "flos": 24536720223360.0,
      "grad_norm": 1.3735740215581913,
      "language_loss": 0.78862917,
      "learning_rate": 2.8102934605611513e-06,
      "loss": 0.81028128,
      "num_input_tokens_seen": 69156460,
      "step": 3212,
      "time_per_iteration": 3.4102485179901123
    },
    {
      "auxiliary_loss_clip": 0.01164409,
      "auxiliary_loss_mlp": 0.01037551,
      "balance_loss_clip": 1.05579066,
      "balance_loss_mlp": 1.02927804,
      "epoch": 0.3863404076233993,
      "flos": 20558212986240.0,
      "grad_norm": 1.8724514536521324,
      "language_loss": 0.67309213,
      "learning_rate": 2.8095812229846665e-06,
      "loss": 0.69511175,
      "num_input_tokens_seen": 69176420,
      "step": 3213,
      "time_per_iteration": 3.3979415893554688
    },
    {
      "auxiliary_loss_clip": 0.0115829,
      "auxiliary_loss_mlp": 0.01029013,
      "balance_loss_clip": 1.0519774,
      "balance_loss_mlp": 1.02116883,
      "epoch": 0.3864606505140384,
      "flos": 22346277039360.0,
      "grad_norm": 2.2409379589455405,
      "language_loss": 0.68583167,
      "learning_rate": 2.808868862595355e-06,
      "loss": 0.70770466,
      "num_input_tokens_seen": 69196665,
      "step": 3214,
      "time_per_iteration": 2.5950615406036377
    },
    {
      "auxiliary_loss_clip": 0.01175955,
      "auxiliary_loss_mlp": 0.01036268,
      "balance_loss_clip": 1.056319,
      "balance_loss_mlp": 1.02845085,
      "epoch": 0.38658089340467744,
      "flos": 25703601448320.0,
      "grad_norm": 1.7329662004284174,
      "language_loss": 0.79644591,
      "learning_rate": 2.8081563795012795e-06,
      "loss": 0.81856817,
      "num_input_tokens_seen": 69216290,
      "step": 3215,
      "time_per_iteration": 2.5582728385925293
    },
    {
      "auxiliary_loss_clip": 0.01162647,
      "auxiliary_loss_mlp": 0.01032012,
      "balance_loss_clip": 1.05226576,
      "balance_loss_mlp": 1.02371526,
      "epoch": 0.38670113629531655,
      "flos": 33802534558080.0,
      "grad_norm": 1.8685675237690929,
      "language_loss": 0.73556638,
      "learning_rate": 2.807443773810524e-06,
      "loss": 0.75751305,
      "num_input_tokens_seen": 69237550,
      "step": 3216,
      "time_per_iteration": 2.69624924659729
    },
    {
      "auxiliary_loss_clip": 0.0114114,
      "auxiliary_loss_mlp": 0.01026062,
      "balance_loss_clip": 1.05107331,
      "balance_loss_mlp": 1.01806927,
      "epoch": 0.3868213791859556,
      "flos": 23331522165120.0,
      "grad_norm": 1.7447337428549612,
      "language_loss": 0.89402711,
      "learning_rate": 2.80673104563119e-06,
      "loss": 0.91569912,
      "num_input_tokens_seen": 69258175,
      "step": 3217,
      "time_per_iteration": 2.748058557510376
    },
    {
      "auxiliary_loss_clip": 0.011669,
      "auxiliary_loss_mlp": 0.01035821,
      "balance_loss_clip": 1.05385423,
      "balance_loss_mlp": 1.02830458,
      "epoch": 0.3869416220765947,
      "flos": 18441530380800.0,
      "grad_norm": 1.969753005943133,
      "language_loss": 0.79100776,
      "learning_rate": 2.8060181950713976e-06,
      "loss": 0.81303495,
      "num_input_tokens_seen": 69274965,
      "step": 3218,
      "time_per_iteration": 2.5027782917022705
    },
    {
      "auxiliary_loss_clip": 0.01137952,
      "auxiliary_loss_mlp": 0.01028167,
      "balance_loss_clip": 1.04760849,
      "balance_loss_mlp": 1.01924443,
      "epoch": 0.3870618649672338,
      "flos": 15632992938240.0,
      "grad_norm": 1.9444174122914126,
      "language_loss": 0.81034541,
      "learning_rate": 2.805305222239286e-06,
      "loss": 0.83200657,
      "num_input_tokens_seen": 69292220,
      "step": 3219,
      "time_per_iteration": 2.576223134994507
    },
    {
      "auxiliary_loss_clip": 0.01152114,
      "auxiliary_loss_mlp": 0.01031545,
      "balance_loss_clip": 1.05209756,
      "balance_loss_mlp": 1.02368331,
      "epoch": 0.3871821078578729,
      "flos": 23513804709120.0,
      "grad_norm": 1.7141292661656335,
      "language_loss": 0.7375254,
      "learning_rate": 2.8045921272430118e-06,
      "loss": 0.75936198,
      "num_input_tokens_seen": 69311900,
      "step": 3220,
      "time_per_iteration": 2.6178882122039795
    },
    {
      "auxiliary_loss_clip": 0.01175583,
      "auxiliary_loss_mlp": 0.01033122,
      "balance_loss_clip": 1.05281734,
      "balance_loss_mlp": 1.02506959,
      "epoch": 0.387302350748512,
      "flos": 17778259791360.0,
      "grad_norm": 2.0322488080583234,
      "language_loss": 0.76448017,
      "learning_rate": 2.803878910190753e-06,
      "loss": 0.78656727,
      "num_input_tokens_seen": 69328820,
      "step": 3221,
      "time_per_iteration": 2.496628761291504
    },
    {
      "auxiliary_loss_clip": 0.01174061,
      "auxiliary_loss_mlp": 0.01031264,
      "balance_loss_clip": 1.05425978,
      "balance_loss_mlp": 1.02347422,
      "epoch": 0.3874225936391511,
      "flos": 11503409097600.0,
      "grad_norm": 2.3220962156436378,
      "language_loss": 0.81860518,
      "learning_rate": 2.8031655711907017e-06,
      "loss": 0.84065843,
      "num_input_tokens_seen": 69342525,
      "step": 3222,
      "time_per_iteration": 2.489475965499878
    },
    {
      "auxiliary_loss_clip": 0.01175339,
      "auxiliary_loss_mlp": 0.01036923,
      "balance_loss_clip": 1.05675018,
      "balance_loss_mlp": 1.02871573,
      "epoch": 0.38754283652979016,
      "flos": 21945154884480.0,
      "grad_norm": 2.0879624834716597,
      "language_loss": 0.80654544,
      "learning_rate": 2.8024521103510723e-06,
      "loss": 0.82866806,
      "num_input_tokens_seen": 69359295,
      "step": 3223,
      "time_per_iteration": 2.527952194213867
    },
    {
      "auxiliary_loss_clip": 0.01167637,
      "auxiliary_loss_mlp": 0.01032091,
      "balance_loss_clip": 1.05054593,
      "balance_loss_mlp": 1.02381158,
      "epoch": 0.38766307942042927,
      "flos": 21175984022400.0,
      "grad_norm": 2.2051621238181087,
      "language_loss": 0.7503497,
      "learning_rate": 2.8017385277800952e-06,
      "loss": 0.77234697,
      "num_input_tokens_seen": 69377650,
      "step": 3224,
      "time_per_iteration": 2.535352945327759
    },
    {
      "auxiliary_loss_clip": 0.01147569,
      "auxiliary_loss_mlp": 0.01035857,
      "balance_loss_clip": 1.05290866,
      "balance_loss_mlp": 1.02766132,
      "epoch": 0.3877833223110684,
      "flos": 27417294391680.0,
      "grad_norm": 3.1782239351241555,
      "language_loss": 0.75343406,
      "learning_rate": 2.8010248235860213e-06,
      "loss": 0.77526832,
      "num_input_tokens_seen": 69397765,
      "step": 3225,
      "time_per_iteration": 2.6397595405578613
    },
    {
      "auxiliary_loss_clip": 0.01057263,
      "auxiliary_loss_mlp": 0.0075204,
      "balance_loss_clip": 1.01655769,
      "balance_loss_mlp": 1.00021076,
      "epoch": 0.38790356520170743,
      "flos": 64500019879680.0,
      "grad_norm": 0.8542399354502178,
      "language_loss": 0.62776875,
      "learning_rate": 2.8003109978771192e-06,
      "loss": 0.64586174,
      "num_input_tokens_seen": 69458930,
      "step": 3226,
      "time_per_iteration": 3.217353582382202
    },
    {
      "auxiliary_loss_clip": 0.01132065,
      "auxiliary_loss_mlp": 0.01029243,
      "balance_loss_clip": 1.04496861,
      "balance_loss_mlp": 1.02145267,
      "epoch": 0.38802380809234654,
      "flos": 22345415112960.0,
      "grad_norm": 1.8257176357712448,
      "language_loss": 0.78634548,
      "learning_rate": 2.799597050761674e-06,
      "loss": 0.8079586,
      "num_input_tokens_seen": 69475135,
      "step": 3227,
      "time_per_iteration": 2.6161904335021973
    },
    {
      "auxiliary_loss_clip": 0.01187093,
      "auxiliary_loss_mlp": 0.01027118,
      "balance_loss_clip": 1.05687141,
      "balance_loss_mlp": 1.01894665,
      "epoch": 0.38814405098298566,
      "flos": 25261361199360.0,
      "grad_norm": 1.8319235918043548,
      "language_loss": 0.79166847,
      "learning_rate": 2.7988829823479924e-06,
      "loss": 0.81381059,
      "num_input_tokens_seen": 69493525,
      "step": 3228,
      "time_per_iteration": 2.5282504558563232
    },
    {
      "auxiliary_loss_clip": 0.01147575,
      "auxiliary_loss_mlp": 0.01033696,
      "balance_loss_clip": 1.04751766,
      "balance_loss_mlp": 1.02592349,
      "epoch": 0.3882642938736247,
      "flos": 18841180078080.0,
      "grad_norm": 1.8634495262081499,
      "language_loss": 0.64134574,
      "learning_rate": 2.7981687927443976e-06,
      "loss": 0.66315854,
      "num_input_tokens_seen": 69510325,
      "step": 3229,
      "time_per_iteration": 2.542198657989502
    },
    {
      "auxiliary_loss_clip": 0.01169023,
      "auxiliary_loss_mlp": 0.01026763,
      "balance_loss_clip": 1.05080342,
      "balance_loss_mlp": 1.01938653,
      "epoch": 0.3883845367642638,
      "flos": 21652806090240.0,
      "grad_norm": 2.5990740733545272,
      "language_loss": 0.85583186,
      "learning_rate": 2.797454482059231e-06,
      "loss": 0.87778968,
      "num_input_tokens_seen": 69530480,
      "step": 3230,
      "time_per_iteration": 2.556319236755371
    },
    {
      "auxiliary_loss_clip": 0.01186695,
      "auxiliary_loss_mlp": 0.01028999,
      "balance_loss_clip": 1.05638611,
      "balance_loss_mlp": 1.02079785,
      "epoch": 0.3885047796549029,
      "flos": 20557530627840.0,
      "grad_norm": 1.6803844136771386,
      "language_loss": 0.84465158,
      "learning_rate": 2.7967400504008537e-06,
      "loss": 0.86680853,
      "num_input_tokens_seen": 69549780,
      "step": 3231,
      "time_per_iteration": 2.48679518699646
    },
    {
      "auxiliary_loss_clip": 0.01044955,
      "auxiliary_loss_mlp": 0.01016988,
      "balance_loss_clip": 1.02924347,
      "balance_loss_mlp": 1.01564682,
      "epoch": 0.388625022545542,
      "flos": 64325491695360.0,
      "grad_norm": 0.8243668827206219,
      "language_loss": 0.57504642,
      "learning_rate": 2.7960254978776456e-06,
      "loss": 0.59566581,
      "num_input_tokens_seen": 69611870,
      "step": 3232,
      "time_per_iteration": 3.2004919052124023
    },
    {
      "auxiliary_loss_clip": 0.01186616,
      "auxiliary_loss_mlp": 0.01033257,
      "balance_loss_clip": 1.05658269,
      "balance_loss_mlp": 1.02491832,
      "epoch": 0.3887452654361811,
      "flos": 18113881495680.0,
      "grad_norm": 2.049442158131418,
      "language_loss": 0.81632876,
      "learning_rate": 2.7953108245980006e-06,
      "loss": 0.8385275,
      "num_input_tokens_seen": 69630385,
      "step": 3233,
      "time_per_iteration": 2.480072021484375
    },
    {
      "auxiliary_loss_clip": 0.01149849,
      "auxiliary_loss_mlp": 0.01026738,
      "balance_loss_clip": 1.05305195,
      "balance_loss_mlp": 1.01921582,
      "epoch": 0.38886550832682015,
      "flos": 24975261371520.0,
      "grad_norm": 1.7231215691533082,
      "language_loss": 0.73714697,
      "learning_rate": 2.7945960306703365e-06,
      "loss": 0.7589128,
      "num_input_tokens_seen": 69653370,
      "step": 3234,
      "time_per_iteration": 3.423224449157715
    },
    {
      "auxiliary_loss_clip": 0.01172552,
      "auxiliary_loss_mlp": 0.01028604,
      "balance_loss_clip": 1.05309439,
      "balance_loss_mlp": 1.02129054,
      "epoch": 0.38898575121745926,
      "flos": 27199496275200.0,
      "grad_norm": 1.7315721883142428,
      "language_loss": 0.65436977,
      "learning_rate": 2.7938811162030865e-06,
      "loss": 0.67638135,
      "num_input_tokens_seen": 69673635,
      "step": 3235,
      "time_per_iteration": 2.5831713676452637
    },
    {
      "auxiliary_loss_clip": 0.01168905,
      "auxiliary_loss_mlp": 0.01022528,
      "balance_loss_clip": 1.05413294,
      "balance_loss_mlp": 1.01517868,
      "epoch": 0.3891059941080984,
      "flos": 28763728727040.0,
      "grad_norm": 1.7140281533287955,
      "language_loss": 0.82570899,
      "learning_rate": 2.793166081304702e-06,
      "loss": 0.84762335,
      "num_input_tokens_seen": 69694130,
      "step": 3236,
      "time_per_iteration": 2.601123094558716
    },
    {
      "auxiliary_loss_clip": 0.01144153,
      "auxiliary_loss_mlp": 0.01037988,
      "balance_loss_clip": 1.04788649,
      "balance_loss_mlp": 1.02953637,
      "epoch": 0.38922623699873743,
      "flos": 22893447893760.0,
      "grad_norm": 2.1550561064969553,
      "language_loss": 0.82250309,
      "learning_rate": 2.7924509260836543e-06,
      "loss": 0.84432447,
      "num_input_tokens_seen": 69713255,
      "step": 3237,
      "time_per_iteration": 2.608346700668335
    },
    {
      "auxiliary_loss_clip": 0.01142908,
      "auxiliary_loss_mlp": 0.01031108,
      "balance_loss_clip": 1.05112755,
      "balance_loss_mlp": 1.02360368,
      "epoch": 0.38934647988937654,
      "flos": 19792418002560.0,
      "grad_norm": 1.5261054476345841,
      "language_loss": 0.6843884,
      "learning_rate": 2.791735650648431e-06,
      "loss": 0.70612854,
      "num_input_tokens_seen": 69732375,
      "step": 3238,
      "time_per_iteration": 3.4508845806121826
    },
    {
      "auxiliary_loss_clip": 0.01152826,
      "auxiliary_loss_mlp": 0.01029852,
      "balance_loss_clip": 1.05173981,
      "balance_loss_mlp": 1.0221988,
      "epoch": 0.38946672278001565,
      "flos": 19202081978880.0,
      "grad_norm": 1.7812232190187807,
      "language_loss": 0.74393767,
      "learning_rate": 2.791020255107538e-06,
      "loss": 0.76576447,
      "num_input_tokens_seen": 69749745,
      "step": 3239,
      "time_per_iteration": 4.14414381980896
    },
    {
      "auxiliary_loss_clip": 0.01133573,
      "auxiliary_loss_mlp": 0.01030912,
      "balance_loss_clip": 1.04400384,
      "balance_loss_mlp": 1.02294874,
      "epoch": 0.3895869656706547,
      "flos": 24936477661440.0,
      "grad_norm": 1.54670484075314,
      "language_loss": 0.80433655,
      "learning_rate": 2.7903047395695023e-06,
      "loss": 0.82598138,
      "num_input_tokens_seen": 69769645,
      "step": 3240,
      "time_per_iteration": 2.6313257217407227
    },
    {
      "auxiliary_loss_clip": 0.01170655,
      "auxiliary_loss_mlp": 0.00761787,
      "balance_loss_clip": 1.05544007,
      "balance_loss_mlp": 1.00022674,
      "epoch": 0.3897072085612938,
      "flos": 24133622820480.0,
      "grad_norm": 3.3450481839136548,
      "language_loss": 0.89830005,
      "learning_rate": 2.789589104142865e-06,
      "loss": 0.91762447,
      "num_input_tokens_seen": 69787270,
      "step": 3241,
      "time_per_iteration": 2.565805673599243
    },
    {
      "auxiliary_loss_clip": 0.01143255,
      "auxiliary_loss_mlp": 0.01032956,
      "balance_loss_clip": 1.05040479,
      "balance_loss_mlp": 1.02498722,
      "epoch": 0.3898274514519329,
      "flos": 17166342672000.0,
      "grad_norm": 1.6787353610362246,
      "language_loss": 0.76628661,
      "learning_rate": 2.7888733489361895e-06,
      "loss": 0.78804874,
      "num_input_tokens_seen": 69805685,
      "step": 3242,
      "time_per_iteration": 2.610018014907837
    },
    {
      "auxiliary_loss_clip": 0.01071529,
      "auxiliary_loss_mlp": 0.01002085,
      "balance_loss_clip": 1.01368737,
      "balance_loss_mlp": 1.00066018,
      "epoch": 0.389947694342572,
      "flos": 66074807952000.0,
      "grad_norm": 0.7313670663724887,
      "language_loss": 0.58720273,
      "learning_rate": 2.788157474058054e-06,
      "loss": 0.60793889,
      "num_input_tokens_seen": 69867960,
      "step": 3243,
      "time_per_iteration": 3.1786117553710938
    },
    {
      "auxiliary_loss_clip": 0.01180056,
      "auxiliary_loss_mlp": 0.01037427,
      "balance_loss_clip": 1.05389822,
      "balance_loss_mlp": 1.02942824,
      "epoch": 0.3900679372332111,
      "flos": 25740912700800.0,
      "grad_norm": 1.5157742346654828,
      "language_loss": 0.698551,
      "learning_rate": 2.7874414796170555e-06,
      "loss": 0.72072577,
      "num_input_tokens_seen": 69889450,
      "step": 3244,
      "time_per_iteration": 2.546280860900879
    },
    {
      "auxiliary_loss_clip": 0.01165346,
      "auxiliary_loss_mlp": 0.01029829,
      "balance_loss_clip": 1.05168879,
      "balance_loss_mlp": 1.02123404,
      "epoch": 0.3901881801238502,
      "flos": 11801611808640.0,
      "grad_norm": 2.6295418538196493,
      "language_loss": 0.83954549,
      "learning_rate": 2.7867253657218113e-06,
      "loss": 0.86149728,
      "num_input_tokens_seen": 69903340,
      "step": 3245,
      "time_per_iteration": 2.4912102222442627
    },
    {
      "auxiliary_loss_clip": 0.0115172,
      "auxiliary_loss_mlp": 0.00761536,
      "balance_loss_clip": 1.0488956,
      "balance_loss_mlp": 1.0002327,
      "epoch": 0.39030842301448926,
      "flos": 27308951994240.0,
      "grad_norm": 1.589552503655944,
      "language_loss": 0.7289483,
      "learning_rate": 2.7860091324809544e-06,
      "loss": 0.74808085,
      "num_input_tokens_seen": 69924400,
      "step": 3246,
      "time_per_iteration": 2.6118240356445312
    },
    {
      "auxiliary_loss_clip": 0.0116889,
      "auxiliary_loss_mlp": 0.01028791,
      "balance_loss_clip": 1.05438185,
      "balance_loss_mlp": 1.02130473,
      "epoch": 0.39042866590512837,
      "flos": 27163334257920.0,
      "grad_norm": 1.9764949992350458,
      "language_loss": 0.80992258,
      "learning_rate": 2.7852927800031377e-06,
      "loss": 0.8318994,
      "num_input_tokens_seen": 69944565,
      "step": 3247,
      "time_per_iteration": 2.584456205368042
    },
    {
      "auxiliary_loss_clip": 0.01155526,
      "auxiliary_loss_mlp": 0.01033756,
      "balance_loss_clip": 1.04942966,
      "balance_loss_mlp": 1.02587605,
      "epoch": 0.3905489087957674,
      "flos": 29716115886720.0,
      "grad_norm": 1.7277352806985147,
      "language_loss": 0.82823861,
      "learning_rate": 2.7845763083970298e-06,
      "loss": 0.85013139,
      "num_input_tokens_seen": 69964965,
      "step": 3248,
      "time_per_iteration": 2.610621690750122
    },
    {
      "auxiliary_loss_clip": 0.0115915,
      "auxiliary_loss_mlp": 0.01029642,
      "balance_loss_clip": 1.04895723,
      "balance_loss_mlp": 1.02110636,
      "epoch": 0.39066915168640653,
      "flos": 24498618871680.0,
      "grad_norm": 2.049054980445831,
      "language_loss": 0.8217243,
      "learning_rate": 2.7838597177713205e-06,
      "loss": 0.84361219,
      "num_input_tokens_seen": 69986055,
      "step": 3249,
      "time_per_iteration": 2.5462067127227783
    },
    {
      "auxiliary_loss_clip": 0.01102515,
      "auxiliary_loss_mlp": 0.01033797,
      "balance_loss_clip": 1.04747891,
      "balance_loss_mlp": 1.02555919,
      "epoch": 0.39078939457704565,
      "flos": 20558572122240.0,
      "grad_norm": 1.6570631191782041,
      "language_loss": 0.73468232,
      "learning_rate": 2.7831430082347143e-06,
      "loss": 0.75604546,
      "num_input_tokens_seen": 70005260,
      "step": 3250,
      "time_per_iteration": 2.653055191040039
    },
    {
      "auxiliary_loss_clip": 0.01171748,
      "auxiliary_loss_mlp": 0.00760647,
      "balance_loss_clip": 1.05418539,
      "balance_loss_mlp": 1.00023413,
      "epoch": 0.3909096374676847,
      "flos": 22783417557120.0,
      "grad_norm": 2.033095262267449,
      "language_loss": 0.81821829,
      "learning_rate": 2.7824261798959373e-06,
      "loss": 0.83754218,
      "num_input_tokens_seen": 70023440,
      "step": 3251,
      "time_per_iteration": 2.537879228591919
    },
    {
      "auxiliary_loss_clip": 0.01156076,
      "auxiliary_loss_mlp": 0.01033835,
      "balance_loss_clip": 1.04867005,
      "balance_loss_mlp": 1.02543056,
      "epoch": 0.3910298803583238,
      "flos": 23003119094400.0,
      "grad_norm": 1.920746822916225,
      "language_loss": 0.79810292,
      "learning_rate": 2.78170923286373e-06,
      "loss": 0.82000202,
      "num_input_tokens_seen": 70043040,
      "step": 3252,
      "time_per_iteration": 2.579200506210327
    },
    {
      "auxiliary_loss_clip": 0.01088983,
      "auxiliary_loss_mlp": 0.01031777,
      "balance_loss_clip": 1.04511571,
      "balance_loss_mlp": 1.02284873,
      "epoch": 0.3911501232489629,
      "flos": 24316264500480.0,
      "grad_norm": 3.3187037399120114,
      "language_loss": 0.83940423,
      "learning_rate": 2.780992167246854e-06,
      "loss": 0.8606118,
      "num_input_tokens_seen": 70060565,
      "step": 3253,
      "time_per_iteration": 2.8390085697174072
    },
    {
      "auxiliary_loss_clip": 0.01052243,
      "auxiliary_loss_mlp": 0.01006461,
      "balance_loss_clip": 1.01272607,
      "balance_loss_mlp": 1.00492287,
      "epoch": 0.391270366139602,
      "flos": 60869054684160.0,
      "grad_norm": 0.9880709424166326,
      "language_loss": 0.72163969,
      "learning_rate": 2.7802749831540883e-06,
      "loss": 0.74222672,
      "num_input_tokens_seen": 70119465,
      "step": 3254,
      "time_per_iteration": 3.17475962638855
    },
    {
      "auxiliary_loss_clip": 0.01131857,
      "auxiliary_loss_mlp": 0.01029311,
      "balance_loss_clip": 1.05002081,
      "balance_loss_mlp": 1.02241516,
      "epoch": 0.3913906090302411,
      "flos": 21543494025600.0,
      "grad_norm": 2.3666916695159808,
      "language_loss": 0.81954634,
      "learning_rate": 2.7795576806942268e-06,
      "loss": 0.84115803,
      "num_input_tokens_seen": 70138270,
      "step": 3255,
      "time_per_iteration": 2.6309120655059814
    },
    {
      "auxiliary_loss_clip": 0.01052229,
      "auxiliary_loss_mlp": 0.01011248,
      "balance_loss_clip": 1.02009797,
      "balance_loss_mlp": 1.00961447,
      "epoch": 0.3915108519208802,
      "flos": 49839953702400.0,
      "grad_norm": 0.7612720311580335,
      "language_loss": 0.54884338,
      "learning_rate": 2.778840259976085e-06,
      "loss": 0.56947815,
      "num_input_tokens_seen": 70193500,
      "step": 3256,
      "time_per_iteration": 3.1145927906036377
    },
    {
      "auxiliary_loss_clip": 0.01170151,
      "auxiliary_loss_mlp": 0.01038557,
      "balance_loss_clip": 1.05212569,
      "balance_loss_mlp": 1.02948475,
      "epoch": 0.39163109481151925,
      "flos": 16506447960960.0,
      "grad_norm": 2.2852955717342778,
      "language_loss": 0.7750895,
      "learning_rate": 2.778122721108495e-06,
      "loss": 0.79717648,
      "num_input_tokens_seen": 70211730,
      "step": 3257,
      "time_per_iteration": 2.52213454246521
    },
    {
      "auxiliary_loss_clip": 0.01167518,
      "auxiliary_loss_mlp": 0.01031742,
      "balance_loss_clip": 1.05392969,
      "balance_loss_mlp": 1.02386248,
      "epoch": 0.39175133770215836,
      "flos": 26067484177920.0,
      "grad_norm": 1.954630301929467,
      "language_loss": 0.8884896,
      "learning_rate": 2.7774050642003076e-06,
      "loss": 0.91048223,
      "num_input_tokens_seen": 70232540,
      "step": 3258,
      "time_per_iteration": 2.5816099643707275
    },
    {
      "auxiliary_loss_clip": 0.01188228,
      "auxiliary_loss_mlp": 0.01037495,
      "balance_loss_clip": 1.05711007,
      "balance_loss_mlp": 1.02876329,
      "epoch": 0.3918715805927975,
      "flos": 21872076664320.0,
      "grad_norm": 2.1342684447936326,
      "language_loss": 0.93349922,
      "learning_rate": 2.7766872893603896e-06,
      "loss": 0.95575649,
      "num_input_tokens_seen": 70252515,
      "step": 3259,
      "time_per_iteration": 3.335916519165039
    },
    {
      "auxiliary_loss_clip": 0.01170113,
      "auxiliary_loss_mlp": 0.01026637,
      "balance_loss_clip": 1.05253899,
      "balance_loss_mlp": 1.01929116,
      "epoch": 0.39199182348343653,
      "flos": 20376181837440.0,
      "grad_norm": 1.905819965160923,
      "language_loss": 0.73136735,
      "learning_rate": 2.7759693966976275e-06,
      "loss": 0.75333482,
      "num_input_tokens_seen": 70271020,
      "step": 3260,
      "time_per_iteration": 2.536393165588379
    },
    {
      "auxiliary_loss_clip": 0.01139192,
      "auxiliary_loss_mlp": 0.01034649,
      "balance_loss_clip": 1.04852343,
      "balance_loss_mlp": 1.02607751,
      "epoch": 0.39211206637407564,
      "flos": 21683545153920.0,
      "grad_norm": 1.8057346217744474,
      "language_loss": 0.85341215,
      "learning_rate": 2.7752513863209242e-06,
      "loss": 0.87515068,
      "num_input_tokens_seen": 70289600,
      "step": 3261,
      "time_per_iteration": 2.569824695587158
    },
    {
      "auxiliary_loss_clip": 0.01154311,
      "auxiliary_loss_mlp": 0.00761434,
      "balance_loss_clip": 1.05439138,
      "balance_loss_mlp": 1.0002563,
      "epoch": 0.39223230926471475,
      "flos": 21066276908160.0,
      "grad_norm": 1.5864133084335696,
      "language_loss": 0.84424812,
      "learning_rate": 2.774533258339203e-06,
      "loss": 0.86340559,
      "num_input_tokens_seen": 70307060,
      "step": 3262,
      "time_per_iteration": 2.570836067199707
    },
    {
      "auxiliary_loss_clip": 0.01126303,
      "auxiliary_loss_mlp": 0.01036758,
      "balance_loss_clip": 1.04465139,
      "balance_loss_mlp": 1.02800763,
      "epoch": 0.3923525521553538,
      "flos": 17603016312960.0,
      "grad_norm": 2.123510192311751,
      "language_loss": 0.79586822,
      "learning_rate": 2.7738150128614014e-06,
      "loss": 0.8174988,
      "num_input_tokens_seen": 70324465,
      "step": 3263,
      "time_per_iteration": 2.5871896743774414
    },
    {
      "auxiliary_loss_clip": 0.01135282,
      "auxiliary_loss_mlp": 0.01031118,
      "balance_loss_clip": 1.05268812,
      "balance_loss_mlp": 1.02290416,
      "epoch": 0.3924727950459929,
      "flos": 20558284813440.0,
      "grad_norm": 1.7772748716760023,
      "language_loss": 0.89541495,
      "learning_rate": 2.7730966499964777e-06,
      "loss": 0.91707897,
      "num_input_tokens_seen": 70341415,
      "step": 3264,
      "time_per_iteration": 2.5985493659973145
    },
    {
      "auxiliary_loss_clip": 0.01185018,
      "auxiliary_loss_mlp": 0.01030495,
      "balance_loss_clip": 1.05404353,
      "balance_loss_mlp": 1.02178049,
      "epoch": 0.39259303793663197,
      "flos": 16216110328320.0,
      "grad_norm": 2.295951458867603,
      "language_loss": 0.80236995,
      "learning_rate": 2.772378169853408e-06,
      "loss": 0.82452506,
      "num_input_tokens_seen": 70358985,
      "step": 3265,
      "time_per_iteration": 4.094515085220337
    },
    {
      "auxiliary_loss_clip": 0.01145654,
      "auxiliary_loss_mlp": 0.01034374,
      "balance_loss_clip": 1.053671,
      "balance_loss_mlp": 1.02649474,
      "epoch": 0.3927132808272711,
      "flos": 16797001075200.0,
      "grad_norm": 2.0678292150447812,
      "language_loss": 0.74176633,
      "learning_rate": 2.771659572541183e-06,
      "loss": 0.76356661,
      "num_input_tokens_seen": 70376915,
      "step": 3266,
      "time_per_iteration": 2.5727617740631104
    },
    {
      "auxiliary_loss_clip": 0.01177775,
      "auxiliary_loss_mlp": 0.01031477,
      "balance_loss_clip": 1.05920088,
      "balance_loss_mlp": 1.02429748,
      "epoch": 0.3928335237179102,
      "flos": 20267228908800.0,
      "grad_norm": 3.350030591112571,
      "language_loss": 0.86953664,
      "learning_rate": 2.7709408581688143e-06,
      "loss": 0.8916291,
      "num_input_tokens_seen": 70396900,
      "step": 3267,
      "time_per_iteration": 2.5279858112335205
    },
    {
      "auxiliary_loss_clip": 0.01147091,
      "auxiliary_loss_mlp": 0.01030275,
      "balance_loss_clip": 1.05213046,
      "balance_loss_mlp": 1.02210045,
      "epoch": 0.39295376660854925,
      "flos": 24973250209920.0,
      "grad_norm": 2.1337871185043897,
      "language_loss": 0.87686706,
      "learning_rate": 2.7702220268453307e-06,
      "loss": 0.89864075,
      "num_input_tokens_seen": 70417260,
      "step": 3268,
      "time_per_iteration": 2.6120400428771973
    },
    {
      "auxiliary_loss_clip": 0.01162462,
      "auxiliary_loss_mlp": 0.01029668,
      "balance_loss_clip": 1.05328667,
      "balance_loss_mlp": 1.02171075,
      "epoch": 0.39307400949918836,
      "flos": 18697788984960.0,
      "grad_norm": 2.3648190785906134,
      "language_loss": 0.84555864,
      "learning_rate": 2.7695030786797785e-06,
      "loss": 0.86747992,
      "num_input_tokens_seen": 70433155,
      "step": 3269,
      "time_per_iteration": 2.5464043617248535
    },
    {
      "auxiliary_loss_clip": 0.01124103,
      "auxiliary_loss_mlp": 0.01035165,
      "balance_loss_clip": 1.04748297,
      "balance_loss_mlp": 1.02693975,
      "epoch": 0.39319425238982747,
      "flos": 22415476590720.0,
      "grad_norm": 1.9286264774730058,
      "language_loss": 0.74821699,
      "learning_rate": 2.7687840137812206e-06,
      "loss": 0.7698096,
      "num_input_tokens_seen": 70451240,
      "step": 3270,
      "time_per_iteration": 2.6429104804992676
    },
    {
      "auxiliary_loss_clip": 0.01057816,
      "auxiliary_loss_mlp": 0.0100833,
      "balance_loss_clip": 1.0129385,
      "balance_loss_mlp": 1.00675666,
      "epoch": 0.3933144952804665,
      "flos": 66192954762240.0,
      "grad_norm": 0.7978615358698185,
      "language_loss": 0.62118399,
      "learning_rate": 2.7680648322587395e-06,
      "loss": 0.64184546,
      "num_input_tokens_seen": 70516115,
      "step": 3271,
      "time_per_iteration": 3.1447014808654785
    },
    {
      "auxiliary_loss_clip": 0.01185082,
      "auxiliary_loss_mlp": 0.01028792,
      "balance_loss_clip": 1.05548275,
      "balance_loss_mlp": 1.02041137,
      "epoch": 0.39343473817110564,
      "flos": 15487159720320.0,
      "grad_norm": 3.2010334533145963,
      "language_loss": 0.81020206,
      "learning_rate": 2.7673455342214334e-06,
      "loss": 0.83234084,
      "num_input_tokens_seen": 70533105,
      "step": 3272,
      "time_per_iteration": 2.4701485633850098
    },
    {
      "auxiliary_loss_clip": 0.0117424,
      "auxiliary_loss_mlp": 0.01032628,
      "balance_loss_clip": 1.0556078,
      "balance_loss_mlp": 1.02489114,
      "epoch": 0.39355498106174475,
      "flos": 21324905809920.0,
      "grad_norm": 2.1309175649715897,
      "language_loss": 0.76431823,
      "learning_rate": 2.7666261197784198e-06,
      "loss": 0.78638697,
      "num_input_tokens_seen": 70551920,
      "step": 3273,
      "time_per_iteration": 2.5363118648529053
    },
    {
      "auxiliary_loss_clip": 0.01150212,
      "auxiliary_loss_mlp": 0.01026077,
      "balance_loss_clip": 1.05297697,
      "balance_loss_mlp": 1.01857853,
      "epoch": 0.3936752239523838,
      "flos": 13296357400320.0,
      "grad_norm": 1.9844937439558237,
      "language_loss": 0.76837862,
      "learning_rate": 2.7659065890388336e-06,
      "loss": 0.79014146,
      "num_input_tokens_seen": 70567920,
      "step": 3274,
      "time_per_iteration": 2.577951431274414
    },
    {
      "auxiliary_loss_clip": 0.01160623,
      "auxiliary_loss_mlp": 0.01029492,
      "balance_loss_clip": 1.05368352,
      "balance_loss_mlp": 1.02164793,
      "epoch": 0.3937954668430229,
      "flos": 16800161472000.0,
      "grad_norm": 2.2151005199006804,
      "language_loss": 0.85047209,
      "learning_rate": 2.7651869421118266e-06,
      "loss": 0.87237322,
      "num_input_tokens_seen": 70584530,
      "step": 3275,
      "time_per_iteration": 2.566519260406494
    },
    {
      "auxiliary_loss_clip": 0.01174999,
      "auxiliary_loss_mlp": 0.0103017,
      "balance_loss_clip": 1.05638647,
      "balance_loss_mlp": 1.02199173,
      "epoch": 0.393915709733662,
      "flos": 21064229832960.0,
      "grad_norm": 1.6342999295201055,
      "language_loss": 0.82774508,
      "learning_rate": 2.76446717910657e-06,
      "loss": 0.84979671,
      "num_input_tokens_seen": 70605235,
      "step": 3276,
      "time_per_iteration": 2.546081066131592
    },
    {
      "auxiliary_loss_clip": 0.01166774,
      "auxiliary_loss_mlp": 0.01027753,
      "balance_loss_clip": 1.05252433,
      "balance_loss_mlp": 1.01981974,
      "epoch": 0.3940359526243011,
      "flos": 17165265264000.0,
      "grad_norm": 2.331022270642384,
      "language_loss": 0.76435333,
      "learning_rate": 2.763747300132249e-06,
      "loss": 0.78629863,
      "num_input_tokens_seen": 70622675,
      "step": 3277,
      "time_per_iteration": 2.485366106033325
    },
    {
      "auxiliary_loss_clip": 0.01187725,
      "auxiliary_loss_mlp": 0.01035974,
      "balance_loss_clip": 1.05767298,
      "balance_loss_mlp": 1.02789736,
      "epoch": 0.3941561955149402,
      "flos": 20995856294400.0,
      "grad_norm": 1.6703445428036323,
      "language_loss": 0.86632192,
      "learning_rate": 2.7630273052980704e-06,
      "loss": 0.88855898,
      "num_input_tokens_seen": 70643265,
      "step": 3278,
      "time_per_iteration": 2.4960572719573975
    },
    {
      "auxiliary_loss_clip": 0.01148258,
      "auxiliary_loss_mlp": 0.01033317,
      "balance_loss_clip": 1.05256605,
      "balance_loss_mlp": 1.02515125,
      "epoch": 0.39427643840557924,
      "flos": 18843406721280.0,
      "grad_norm": 2.3811959633599935,
      "language_loss": 0.67455351,
      "learning_rate": 2.7623071947132554e-06,
      "loss": 0.69636929,
      "num_input_tokens_seen": 70660295,
      "step": 3279,
      "time_per_iteration": 2.546435594558716
    },
    {
      "auxiliary_loss_clip": 0.01162527,
      "auxiliary_loss_mlp": 0.01031639,
      "balance_loss_clip": 1.05239069,
      "balance_loss_mlp": 1.02348542,
      "epoch": 0.39439668129621835,
      "flos": 23258659426560.0,
      "grad_norm": 2.0566377453372477,
      "language_loss": 0.78740907,
      "learning_rate": 2.7615869684870458e-06,
      "loss": 0.80935067,
      "num_input_tokens_seen": 70679605,
      "step": 3280,
      "time_per_iteration": 2.573173761367798
    },
    {
      "auxiliary_loss_clip": 0.01171875,
      "auxiliary_loss_mlp": 0.01033458,
      "balance_loss_clip": 1.05555797,
      "balance_loss_mlp": 1.02532792,
      "epoch": 0.39451692418685746,
      "flos": 26652289507200.0,
      "grad_norm": 1.6225039422081649,
      "language_loss": 0.8419075,
      "learning_rate": 2.7608666267286986e-06,
      "loss": 0.8639608,
      "num_input_tokens_seen": 70699835,
      "step": 3281,
      "time_per_iteration": 2.5638794898986816
    },
    {
      "auxiliary_loss_clip": 0.01110819,
      "auxiliary_loss_mlp": 0.01030454,
      "balance_loss_clip": 1.04595852,
      "balance_loss_mlp": 1.02205586,
      "epoch": 0.3946371670774965,
      "flos": 18258709132800.0,
      "grad_norm": 2.0236008008002995,
      "language_loss": 0.86290181,
      "learning_rate": 2.760146169547489e-06,
      "loss": 0.88431454,
      "num_input_tokens_seen": 70716600,
      "step": 3282,
      "time_per_iteration": 2.6307291984558105
    },
    {
      "auxiliary_loss_clip": 0.0115997,
      "auxiliary_loss_mlp": 0.01042985,
      "balance_loss_clip": 1.05572939,
      "balance_loss_mlp": 1.03413987,
      "epoch": 0.39475740996813563,
      "flos": 24206126423040.0,
      "grad_norm": 1.4231476928590283,
      "language_loss": 0.76266265,
      "learning_rate": 2.75942559705271e-06,
      "loss": 0.78469217,
      "num_input_tokens_seen": 70736335,
      "step": 3283,
      "time_per_iteration": 2.5898656845092773
    },
    {
      "auxiliary_loss_clip": 0.01167257,
      "auxiliary_loss_mlp": 0.0102965,
      "balance_loss_clip": 1.05377078,
      "balance_loss_mlp": 1.0212338,
      "epoch": 0.39487765285877474,
      "flos": 19317858491520.0,
      "grad_norm": 1.9167890177164317,
      "language_loss": 0.8911733,
      "learning_rate": 2.7587049093536713e-06,
      "loss": 0.91314238,
      "num_input_tokens_seen": 70752665,
      "step": 3284,
      "time_per_iteration": 2.5155842304229736
    },
    {
      "auxiliary_loss_clip": 0.01173582,
      "auxiliary_loss_mlp": 0.01037013,
      "balance_loss_clip": 1.05436468,
      "balance_loss_mlp": 1.02945256,
      "epoch": 0.3949978957494138,
      "flos": 17311744926720.0,
      "grad_norm": 1.8266131499792055,
      "language_loss": 0.80683261,
      "learning_rate": 2.757984106559701e-06,
      "loss": 0.82893848,
      "num_input_tokens_seen": 70771650,
      "step": 3285,
      "time_per_iteration": 2.520636796951294
    },
    {
      "auxiliary_loss_clip": 0.01151428,
      "auxiliary_loss_mlp": 0.01036935,
      "balance_loss_clip": 1.05452752,
      "balance_loss_mlp": 1.02863252,
      "epoch": 0.3951181386400529,
      "flos": 36317861280000.0,
      "grad_norm": 2.892297521277485,
      "language_loss": 0.71352828,
      "learning_rate": 2.7572631887801446e-06,
      "loss": 0.73541194,
      "num_input_tokens_seen": 70793275,
      "step": 3286,
      "time_per_iteration": 3.4727320671081543
    },
    {
      "auxiliary_loss_clip": 0.01171379,
      "auxiliary_loss_mlp": 0.01026225,
      "balance_loss_clip": 1.05533028,
      "balance_loss_mlp": 1.01757634,
      "epoch": 0.395238381530692,
      "flos": 23110348170240.0,
      "grad_norm": 1.7365804446576287,
      "language_loss": 0.76525402,
      "learning_rate": 2.7565421561243654e-06,
      "loss": 0.78723007,
      "num_input_tokens_seen": 70811440,
      "step": 3287,
      "time_per_iteration": 2.5473058223724365
    },
    {
      "auxiliary_loss_clip": 0.0113478,
      "auxiliary_loss_mlp": 0.01030105,
      "balance_loss_clip": 1.04849851,
      "balance_loss_mlp": 1.02239203,
      "epoch": 0.3953586244213311,
      "flos": 24347614095360.0,
      "grad_norm": 2.3900522875107746,
      "language_loss": 0.82140088,
      "learning_rate": 2.7558210087017413e-06,
      "loss": 0.84304976,
      "num_input_tokens_seen": 70831375,
      "step": 3288,
      "time_per_iteration": 2.616666078567505
    },
    {
      "auxiliary_loss_clip": 0.01139168,
      "auxiliary_loss_mlp": 0.01028302,
      "balance_loss_clip": 1.05337,
      "balance_loss_mlp": 1.01939106,
      "epoch": 0.3954788673119702,
      "flos": 23440080044160.0,
      "grad_norm": 2.008667809664639,
      "language_loss": 0.73117793,
      "learning_rate": 2.7550997466216724e-06,
      "loss": 0.75285262,
      "num_input_tokens_seen": 70849170,
      "step": 3289,
      "time_per_iteration": 2.5922436714172363
    },
    {
      "auxiliary_loss_clip": 0.01153873,
      "auxiliary_loss_mlp": 0.01028839,
      "balance_loss_clip": 1.05578983,
      "balance_loss_mlp": 1.02035761,
      "epoch": 0.3955991102026093,
      "flos": 17494063384320.0,
      "grad_norm": 2.457065533447961,
      "language_loss": 0.80941987,
      "learning_rate": 2.7543783699935714e-06,
      "loss": 0.83124697,
      "num_input_tokens_seen": 70867200,
      "step": 3290,
      "time_per_iteration": 3.4038445949554443
    },
    {
      "auxiliary_loss_clip": 0.01171445,
      "auxiliary_loss_mlp": 0.01026741,
      "balance_loss_clip": 1.05879951,
      "balance_loss_mlp": 1.0189209,
      "epoch": 0.39571935309324835,
      "flos": 18221326053120.0,
      "grad_norm": 2.3971123935362137,
      "language_loss": 0.85789573,
      "learning_rate": 2.753656878926872e-06,
      "loss": 0.87987757,
      "num_input_tokens_seen": 70883080,
      "step": 3291,
      "time_per_iteration": 4.071390151977539
    },
    {
      "auxiliary_loss_clip": 0.01145009,
      "auxiliary_loss_mlp": 0.01036039,
      "balance_loss_clip": 1.04919922,
      "balance_loss_mlp": 1.02733636,
      "epoch": 0.39583959598388746,
      "flos": 17748813617280.0,
      "grad_norm": 1.9408406446712727,
      "language_loss": 0.74345124,
      "learning_rate": 2.752935273531023e-06,
      "loss": 0.76526165,
      "num_input_tokens_seen": 70901230,
      "step": 3292,
      "time_per_iteration": 2.532723903656006
    },
    {
      "auxiliary_loss_clip": 0.01174694,
      "auxiliary_loss_mlp": 0.01033203,
      "balance_loss_clip": 1.0588572,
      "balance_loss_mlp": 1.0243156,
      "epoch": 0.39595983887452657,
      "flos": 19352368483200.0,
      "grad_norm": 1.7887562831309047,
      "language_loss": 0.78467679,
      "learning_rate": 2.752213553915492e-06,
      "loss": 0.80675572,
      "num_input_tokens_seen": 70919585,
      "step": 3293,
      "time_per_iteration": 2.5259244441986084
    },
    {
      "auxiliary_loss_clip": 0.01052108,
      "auxiliary_loss_mlp": 0.010026,
      "balance_loss_clip": 1.0171566,
      "balance_loss_mlp": 1.00117588,
      "epoch": 0.3960800817651656,
      "flos": 60682282940160.0,
      "grad_norm": 0.8129821457787784,
      "language_loss": 0.66054308,
      "learning_rate": 2.751491720189762e-06,
      "loss": 0.68109012,
      "num_input_tokens_seen": 70977695,
      "step": 3294,
      "time_per_iteration": 3.082036018371582
    },
    {
      "auxiliary_loss_clip": 0.0115708,
      "auxiliary_loss_mlp": 0.00761619,
      "balance_loss_clip": 1.05355525,
      "balance_loss_mlp": 1.00032139,
      "epoch": 0.39620032465580474,
      "flos": 16836718538880.0,
      "grad_norm": 2.030760166406356,
      "language_loss": 0.91710925,
      "learning_rate": 2.7507697724633364e-06,
      "loss": 0.93629628,
      "num_input_tokens_seen": 70994455,
      "step": 3295,
      "time_per_iteration": 2.5268332958221436
    },
    {
      "auxiliary_loss_clip": 0.01042586,
      "auxiliary_loss_mlp": 0.01003243,
      "balance_loss_clip": 1.02218175,
      "balance_loss_mlp": 1.00170541,
      "epoch": 0.3963205675464438,
      "flos": 69071445941760.0,
      "grad_norm": 0.7735171579267955,
      "language_loss": 0.54697073,
      "learning_rate": 2.7500477108457327e-06,
      "loss": 0.56742907,
      "num_input_tokens_seen": 71046465,
      "step": 3296,
      "time_per_iteration": 2.972369432449341
    },
    {
      "auxiliary_loss_clip": 0.01167161,
      "auxiliary_loss_mlp": 0.01030607,
      "balance_loss_clip": 1.05281234,
      "balance_loss_mlp": 1.02270377,
      "epoch": 0.3964408104370829,
      "flos": 25667439431040.0,
      "grad_norm": 1.9110668256920045,
      "language_loss": 0.80341047,
      "learning_rate": 2.7493255354464877e-06,
      "loss": 0.82538813,
      "num_input_tokens_seen": 71064275,
      "step": 3297,
      "time_per_iteration": 2.5592949390411377
    },
    {
      "auxiliary_loss_clip": 0.01058324,
      "auxiliary_loss_mlp": 0.01031381,
      "balance_loss_clip": 1.03888309,
      "balance_loss_mlp": 1.02319145,
      "epoch": 0.396561053327722,
      "flos": 24277480790400.0,
      "grad_norm": 1.7489507792176195,
      "language_loss": 0.75932407,
      "learning_rate": 2.748603246375156e-06,
      "loss": 0.78022116,
      "num_input_tokens_seen": 71082290,
      "step": 3298,
      "time_per_iteration": 2.9225733280181885
    },
    {
      "auxiliary_loss_clip": 0.01185642,
      "auxiliary_loss_mlp": 0.01033435,
      "balance_loss_clip": 1.0567919,
      "balance_loss_mlp": 1.02559066,
      "epoch": 0.39668129621836107,
      "flos": 20522302364160.0,
      "grad_norm": 2.4727723494217444,
      "language_loss": 0.69935441,
      "learning_rate": 2.7478808437413055e-06,
      "loss": 0.72154522,
      "num_input_tokens_seen": 71101700,
      "step": 3299,
      "time_per_iteration": 2.6937520503997803
    },
    {
      "auxiliary_loss_clip": 0.01127575,
      "auxiliary_loss_mlp": 0.01030009,
      "balance_loss_clip": 1.05220294,
      "balance_loss_mlp": 1.02206409,
      "epoch": 0.3968015391090002,
      "flos": 27052585649280.0,
      "grad_norm": 2.3214455503447873,
      "language_loss": 0.65805173,
      "learning_rate": 2.7471583276545263e-06,
      "loss": 0.6796276,
      "num_input_tokens_seen": 71122360,
      "step": 3300,
      "time_per_iteration": 2.6683387756347656
    },
    {
      "auxiliary_loss_clip": 0.01157527,
      "auxiliary_loss_mlp": 0.01028493,
      "balance_loss_clip": 1.05259764,
      "balance_loss_mlp": 1.02069128,
      "epoch": 0.3969217819996393,
      "flos": 12531819392640.0,
      "grad_norm": 1.9290801131675008,
      "language_loss": 0.70496756,
      "learning_rate": 2.7464356982244224e-06,
      "loss": 0.72682774,
      "num_input_tokens_seen": 71140360,
      "step": 3301,
      "time_per_iteration": 2.560967206954956
    },
    {
      "auxiliary_loss_clip": 0.01064903,
      "auxiliary_loss_mlp": 0.01001034,
      "balance_loss_clip": 1.02266634,
      "balance_loss_mlp": 0.99963969,
      "epoch": 0.39704202489027834,
      "flos": 66241399230720.0,
      "grad_norm": 0.7988742560867907,
      "language_loss": 0.61715102,
      "learning_rate": 2.745712955560617e-06,
      "loss": 0.63781041,
      "num_input_tokens_seen": 71196565,
      "step": 3302,
      "time_per_iteration": 3.0620484352111816
    },
    {
      "auxiliary_loss_clip": 0.01115167,
      "auxiliary_loss_mlp": 0.01026017,
      "balance_loss_clip": 1.0485487,
      "balance_loss_mlp": 1.01763105,
      "epoch": 0.39716226778091746,
      "flos": 16982982720000.0,
      "grad_norm": 2.6489340734253477,
      "language_loss": 0.76873058,
      "learning_rate": 2.7449900997727496e-06,
      "loss": 0.79014242,
      "num_input_tokens_seen": 71214675,
      "step": 3303,
      "time_per_iteration": 2.6332030296325684
    },
    {
      "auxiliary_loss_clip": 0.01151733,
      "auxiliary_loss_mlp": 0.01029901,
      "balance_loss_clip": 1.05304015,
      "balance_loss_mlp": 1.02197385,
      "epoch": 0.39728251067155657,
      "flos": 23477139901440.0,
      "grad_norm": 1.6853211034812077,
      "language_loss": 0.8411541,
      "learning_rate": 2.744267130970476e-06,
      "loss": 0.86297047,
      "num_input_tokens_seen": 71234400,
      "step": 3304,
      "time_per_iteration": 2.5914056301116943
    },
    {
      "auxiliary_loss_clip": 0.01152548,
      "auxiliary_loss_mlp": 0.01028894,
      "balance_loss_clip": 1.05313015,
      "balance_loss_mlp": 1.02047741,
      "epoch": 0.3974027535621956,
      "flos": 20704441253760.0,
      "grad_norm": 1.927759408763687,
      "language_loss": 0.77471602,
      "learning_rate": 2.7435440492634697e-06,
      "loss": 0.79653049,
      "num_input_tokens_seen": 71253725,
      "step": 3305,
      "time_per_iteration": 2.5737717151641846
    },
    {
      "auxiliary_loss_clip": 0.01155242,
      "auxiliary_loss_mlp": 0.01030262,
      "balance_loss_clip": 1.0521692,
      "balance_loss_mlp": 1.02075553,
      "epoch": 0.39752299645283473,
      "flos": 21543278544000.0,
      "grad_norm": 1.9592512502219983,
      "language_loss": 0.67389464,
      "learning_rate": 2.7428208547614228e-06,
      "loss": 0.69574964,
      "num_input_tokens_seen": 71273220,
      "step": 3306,
      "time_per_iteration": 2.573829412460327
    },
    {
      "auxiliary_loss_clip": 0.01171515,
      "auxiliary_loss_mlp": 0.01031713,
      "balance_loss_clip": 1.05674767,
      "balance_loss_mlp": 1.02351725,
      "epoch": 0.39764323934347384,
      "flos": 19208295031680.0,
      "grad_norm": 1.9418032805684147,
      "language_loss": 0.77249783,
      "learning_rate": 2.742097547574043e-06,
      "loss": 0.79453015,
      "num_input_tokens_seen": 71291445,
      "step": 3307,
      "time_per_iteration": 2.5258736610412598
    },
    {
      "auxiliary_loss_clip": 0.01160599,
      "auxiliary_loss_mlp": 0.00761514,
      "balance_loss_clip": 1.05258822,
      "balance_loss_mlp": 1.00033844,
      "epoch": 0.3977634822341129,
      "flos": 20850202644480.0,
      "grad_norm": 2.1180484698303172,
      "language_loss": 0.77222586,
      "learning_rate": 2.7413741278110544e-06,
      "loss": 0.79144698,
      "num_input_tokens_seen": 71310135,
      "step": 3308,
      "time_per_iteration": 2.568763256072998
    },
    {
      "auxiliary_loss_clip": 0.01159932,
      "auxiliary_loss_mlp": 0.01030806,
      "balance_loss_clip": 1.05487466,
      "balance_loss_mlp": 1.02262247,
      "epoch": 0.397883725124752,
      "flos": 39786042038400.0,
      "grad_norm": 2.1326684185653253,
      "language_loss": 0.68663478,
      "learning_rate": 2.7406505955822016e-06,
      "loss": 0.70854211,
      "num_input_tokens_seen": 71331160,
      "step": 3309,
      "time_per_iteration": 2.7219078540802
    },
    {
      "auxiliary_loss_clip": 0.01152981,
      "auxiliary_loss_mlp": 0.010309,
      "balance_loss_clip": 1.05048513,
      "balance_loss_mlp": 1.02288949,
      "epoch": 0.39800396801539106,
      "flos": 17379507934080.0,
      "grad_norm": 3.2728819154551605,
      "language_loss": 0.66288781,
      "learning_rate": 2.7399269509972415e-06,
      "loss": 0.68472666,
      "num_input_tokens_seen": 71345315,
      "step": 3310,
      "time_per_iteration": 2.518404960632324
    },
    {
      "auxiliary_loss_clip": 0.01145745,
      "auxiliary_loss_mlp": 0.01032356,
      "balance_loss_clip": 1.04636765,
      "balance_loss_mlp": 1.02318907,
      "epoch": 0.3981242109060302,
      "flos": 19202764337280.0,
      "grad_norm": 2.420787134398533,
      "language_loss": 0.85568476,
      "learning_rate": 2.7392031941659514e-06,
      "loss": 0.87746578,
      "num_input_tokens_seen": 71363160,
      "step": 3311,
      "time_per_iteration": 3.3499300479888916
    },
    {
      "auxiliary_loss_clip": 0.01157677,
      "auxiliary_loss_mlp": 0.01028967,
      "balance_loss_clip": 1.05691862,
      "balance_loss_mlp": 1.02098048,
      "epoch": 0.3982444537966693,
      "flos": 24565124903040.0,
      "grad_norm": 2.027237978263829,
      "language_loss": 0.85940158,
      "learning_rate": 2.7384793251981244e-06,
      "loss": 0.88126802,
      "num_input_tokens_seen": 71382145,
      "step": 3312,
      "time_per_iteration": 2.6079113483428955
    },
    {
      "auxiliary_loss_clip": 0.01174385,
      "auxiliary_loss_mlp": 0.01031654,
      "balance_loss_clip": 1.05407584,
      "balance_loss_mlp": 1.02394152,
      "epoch": 0.39836469668730834,
      "flos": 26213856099840.0,
      "grad_norm": 1.7148447482353042,
      "language_loss": 0.80605197,
      "learning_rate": 2.737755344203571e-06,
      "loss": 0.82811236,
      "num_input_tokens_seen": 71402095,
      "step": 3313,
      "time_per_iteration": 2.717622756958008
    },
    {
      "auxiliary_loss_clip": 0.01171739,
      "auxiliary_loss_mlp": 0.01031686,
      "balance_loss_clip": 1.05520117,
      "balance_loss_mlp": 1.02332354,
      "epoch": 0.39848493957794745,
      "flos": 27636134002560.0,
      "grad_norm": 1.5056439760959839,
      "language_loss": 0.79993337,
      "learning_rate": 2.7370312512921186e-06,
      "loss": 0.8219676,
      "num_input_tokens_seen": 71423875,
      "step": 3314,
      "time_per_iteration": 2.5878875255584717
    },
    {
      "auxiliary_loss_clip": 0.01156642,
      "auxiliary_loss_mlp": 0.01028516,
      "balance_loss_clip": 1.05029678,
      "balance_loss_mlp": 1.0195868,
      "epoch": 0.39860518246858656,
      "flos": 12239326944000.0,
      "grad_norm": 2.191526585801904,
      "language_loss": 0.76624763,
      "learning_rate": 2.736307046573611e-06,
      "loss": 0.78809923,
      "num_input_tokens_seen": 71439745,
      "step": 3315,
      "time_per_iteration": 2.5447206497192383
    },
    {
      "auxiliary_loss_clip": 0.01180907,
      "auxiliary_loss_mlp": 0.01029272,
      "balance_loss_clip": 1.05361164,
      "balance_loss_mlp": 1.02136827,
      "epoch": 0.3987254253592256,
      "flos": 22379135005440.0,
      "grad_norm": 1.7069022897227117,
      "language_loss": 0.81826562,
      "learning_rate": 2.73558273015791e-06,
      "loss": 0.84036744,
      "num_input_tokens_seen": 71459575,
      "step": 3316,
      "time_per_iteration": 3.293494939804077
    },
    {
      "auxiliary_loss_clip": 0.0118664,
      "auxiliary_loss_mlp": 0.01031079,
      "balance_loss_clip": 1.05639482,
      "balance_loss_mlp": 1.02276373,
      "epoch": 0.3988456682498647,
      "flos": 23514020190720.0,
      "grad_norm": 2.266805566956406,
      "language_loss": 0.71389675,
      "learning_rate": 2.734858302154894e-06,
      "loss": 0.73607397,
      "num_input_tokens_seen": 71481075,
      "step": 3317,
      "time_per_iteration": 4.075145959854126
    },
    {
      "auxiliary_loss_clip": 0.01151747,
      "auxiliary_loss_mlp": 0.01028068,
      "balance_loss_clip": 1.05188441,
      "balance_loss_mlp": 1.02024746,
      "epoch": 0.39896591114050384,
      "flos": 19208761908480.0,
      "grad_norm": 4.478554232385496,
      "language_loss": 0.76242644,
      "learning_rate": 2.734133762674457e-06,
      "loss": 0.78422457,
      "num_input_tokens_seen": 71500665,
      "step": 3318,
      "time_per_iteration": 2.555310010910034
    },
    {
      "auxiliary_loss_clip": 0.01158735,
      "auxiliary_loss_mlp": 0.01028487,
      "balance_loss_clip": 1.05432987,
      "balance_loss_mlp": 1.02046382,
      "epoch": 0.3990861540311429,
      "flos": 28401031146240.0,
      "grad_norm": 1.9297441681361427,
      "language_loss": 0.70583892,
      "learning_rate": 2.7334091118265124e-06,
      "loss": 0.7277112,
      "num_input_tokens_seen": 71522560,
      "step": 3319,
      "time_per_iteration": 2.666428327560425
    },
    {
      "auxiliary_loss_clip": 0.0106552,
      "auxiliary_loss_mlp": 0.01010123,
      "balance_loss_clip": 1.01775646,
      "balance_loss_mlp": 1.0086863,
      "epoch": 0.399206396921782,
      "flos": 61758563086080.0,
      "grad_norm": 0.6817551335941444,
      "language_loss": 0.5782308,
      "learning_rate": 2.732684349720989e-06,
      "loss": 0.59898722,
      "num_input_tokens_seen": 71590520,
      "step": 3320,
      "time_per_iteration": 3.121854305267334
    },
    {
      "auxiliary_loss_clip": 0.01142268,
      "auxiliary_loss_mlp": 0.01030544,
      "balance_loss_clip": 1.04986453,
      "balance_loss_mlp": 1.02214551,
      "epoch": 0.3993266398124211,
      "flos": 28074567409920.0,
      "grad_norm": 1.7137408975460526,
      "language_loss": 0.7505728,
      "learning_rate": 2.7319594764678318e-06,
      "loss": 0.77230096,
      "num_input_tokens_seen": 71612620,
      "step": 3321,
      "time_per_iteration": 2.6383843421936035
    },
    {
      "auxiliary_loss_clip": 0.01128991,
      "auxiliary_loss_mlp": 0.01035486,
      "balance_loss_clip": 1.04957819,
      "balance_loss_mlp": 1.02675426,
      "epoch": 0.39944688270306017,
      "flos": 23225083188480.0,
      "grad_norm": 11.017534996621418,
      "language_loss": 0.83142871,
      "learning_rate": 2.7312344921770044e-06,
      "loss": 0.85307348,
      "num_input_tokens_seen": 71634320,
      "step": 3322,
      "time_per_iteration": 2.659432888031006
    },
    {
      "auxiliary_loss_clip": 0.01154859,
      "auxiliary_loss_mlp": 0.01025745,
      "balance_loss_clip": 1.05069232,
      "balance_loss_mlp": 1.01819897,
      "epoch": 0.3995671255936993,
      "flos": 19390433921280.0,
      "grad_norm": 1.8086302134730927,
      "language_loss": 0.78571028,
      "learning_rate": 2.7305093969584857e-06,
      "loss": 0.80751634,
      "num_input_tokens_seen": 71653145,
      "step": 3323,
      "time_per_iteration": 2.5431511402130127
    },
    {
      "auxiliary_loss_clip": 0.01163549,
      "auxiliary_loss_mlp": 0.01031459,
      "balance_loss_clip": 1.05208802,
      "balance_loss_mlp": 1.02368653,
      "epoch": 0.3996873684843384,
      "flos": 23842638743040.0,
      "grad_norm": 1.9090835762648626,
      "language_loss": 0.80103266,
      "learning_rate": 2.729784190922272e-06,
      "loss": 0.82298273,
      "num_input_tokens_seen": 71674580,
      "step": 3324,
      "time_per_iteration": 2.7020463943481445
    },
    {
      "auxiliary_loss_clip": 0.01048096,
      "auxiliary_loss_mlp": 0.01002681,
      "balance_loss_clip": 1.01411486,
      "balance_loss_mlp": 1.0011189,
      "epoch": 0.39980761137497745,
      "flos": 66576877280640.0,
      "grad_norm": 0.9460573816378759,
      "language_loss": 0.57178736,
      "learning_rate": 2.729058874178378e-06,
      "loss": 0.59229517,
      "num_input_tokens_seen": 71745260,
      "step": 3325,
      "time_per_iteration": 3.199441432952881
    },
    {
      "auxiliary_loss_clip": 0.01160955,
      "auxiliary_loss_mlp": 0.01028019,
      "balance_loss_clip": 1.0541923,
      "balance_loss_mlp": 1.0194838,
      "epoch": 0.39992785426561656,
      "flos": 28549162834560.0,
      "grad_norm": 2.0373415527164767,
      "language_loss": 0.68971497,
      "learning_rate": 2.7283334468368315e-06,
      "loss": 0.71160471,
      "num_input_tokens_seen": 71766540,
      "step": 3326,
      "time_per_iteration": 2.6941609382629395
    },
    {
      "auxiliary_loss_clip": 0.01081471,
      "auxiliary_loss_mlp": 0.01027437,
      "balance_loss_clip": 1.03998208,
      "balance_loss_mlp": 1.01893127,
      "epoch": 0.4000480971562556,
      "flos": 15049408671360.0,
      "grad_norm": 1.871062557544671,
      "language_loss": 0.72889221,
      "learning_rate": 2.72760790900768e-06,
      "loss": 0.74998128,
      "num_input_tokens_seen": 71783125,
      "step": 3327,
      "time_per_iteration": 2.7893972396850586
    },
    {
      "auxiliary_loss_clip": 0.01187092,
      "auxiliary_loss_mlp": 0.01029321,
      "balance_loss_clip": 1.05846369,
      "balance_loss_mlp": 1.0211848,
      "epoch": 0.4001683400468947,
      "flos": 23915609222400.0,
      "grad_norm": 1.8215921715513685,
      "language_loss": 0.79121959,
      "learning_rate": 2.7268822608009875e-06,
      "loss": 0.8133837,
      "num_input_tokens_seen": 71802500,
      "step": 3328,
      "time_per_iteration": 2.8155364990234375
    },
    {
      "auxiliary_loss_clip": 0.01147576,
      "auxiliary_loss_mlp": 0.01030505,
      "balance_loss_clip": 1.05193067,
      "balance_loss_mlp": 1.02252376,
      "epoch": 0.40028858293753383,
      "flos": 24352677912960.0,
      "grad_norm": 1.9811420240404083,
      "language_loss": 0.78053224,
      "learning_rate": 2.726156502326834e-06,
      "loss": 0.80231303,
      "num_input_tokens_seen": 71823800,
      "step": 3329,
      "time_per_iteration": 2.7047863006591797
    },
    {
      "auxiliary_loss_clip": 0.01020533,
      "auxiliary_loss_mlp": 0.01001206,
      "balance_loss_clip": 1.01619411,
      "balance_loss_mlp": 0.99963284,
      "epoch": 0.4004088258281729,
      "flos": 66787025800320.0,
      "grad_norm": 0.69538167002949,
      "language_loss": 0.60260153,
      "learning_rate": 2.725430633695316e-06,
      "loss": 0.62281895,
      "num_input_tokens_seen": 71886880,
      "step": 3330,
      "time_per_iteration": 3.2746493816375732
    },
    {
      "auxiliary_loss_clip": 0.01071055,
      "auxiliary_loss_mlp": 0.01004128,
      "balance_loss_clip": 1.01403213,
      "balance_loss_mlp": 1.00280428,
      "epoch": 0.400529068718812,
      "flos": 58598386473600.0,
      "grad_norm": 0.9087665369101099,
      "language_loss": 0.58001769,
      "learning_rate": 2.7247046550165485e-06,
      "loss": 0.60076952,
      "num_input_tokens_seen": 71939005,
      "step": 3331,
      "time_per_iteration": 3.109802484512329
    },
    {
      "auxiliary_loss_clip": 0.01186647,
      "auxiliary_loss_mlp": 0.01033389,
      "balance_loss_clip": 1.05826902,
      "balance_loss_mlp": 1.02427006,
      "epoch": 0.4006493116094511,
      "flos": 25377460934400.0,
      "grad_norm": 1.4113942120544534,
      "language_loss": 0.75682831,
      "learning_rate": 2.7239785664006606e-06,
      "loss": 0.77902859,
      "num_input_tokens_seen": 71962545,
      "step": 3332,
      "time_per_iteration": 2.5876948833465576
    },
    {
      "auxiliary_loss_clip": 0.01061021,
      "auxiliary_loss_mlp": 0.01005247,
      "balance_loss_clip": 1.01361299,
      "balance_loss_mlp": 1.00396562,
      "epoch": 0.40076955450009016,
      "flos": 60280729822080.0,
      "grad_norm": 0.9866681568630786,
      "language_loss": 0.61797071,
      "learning_rate": 2.7232523679578002e-06,
      "loss": 0.63863337,
      "num_input_tokens_seen": 72025625,
      "step": 3333,
      "time_per_iteration": 3.1671290397644043
    },
    {
      "auxiliary_loss_clip": 0.01172267,
      "auxiliary_loss_mlp": 0.01037208,
      "balance_loss_clip": 1.05733252,
      "balance_loss_mlp": 1.02914906,
      "epoch": 0.4008897973907293,
      "flos": 16617268396800.0,
      "grad_norm": 2.123020252981161,
      "language_loss": 0.79410017,
      "learning_rate": 2.7225260597981295e-06,
      "loss": 0.81619489,
      "num_input_tokens_seen": 72043330,
      "step": 3334,
      "time_per_iteration": 2.5278756618499756
    },
    {
      "auxiliary_loss_clip": 0.01141591,
      "auxiliary_loss_mlp": 0.00762415,
      "balance_loss_clip": 1.0525912,
      "balance_loss_mlp": 1.00034368,
      "epoch": 0.4010100402813684,
      "flos": 15377344865280.0,
      "grad_norm": 2.858855117760107,
      "language_loss": 0.78497851,
      "learning_rate": 2.721799642031831e-06,
      "loss": 0.80401862,
      "num_input_tokens_seen": 72059500,
      "step": 3335,
      "time_per_iteration": 2.5957112312316895
    },
    {
      "auxiliary_loss_clip": 0.01161648,
      "auxiliary_loss_mlp": 0.01028496,
      "balance_loss_clip": 1.05198014,
      "balance_loss_mlp": 1.02041376,
      "epoch": 0.40113028317200744,
      "flos": 13298835438720.0,
      "grad_norm": 2.0133398303646404,
      "language_loss": 0.77450228,
      "learning_rate": 2.721073114769101e-06,
      "loss": 0.79640365,
      "num_input_tokens_seen": 72077175,
      "step": 3336,
      "time_per_iteration": 2.560335159301758
    },
    {
      "auxiliary_loss_clip": 0.01141674,
      "auxiliary_loss_mlp": 0.01029928,
      "balance_loss_clip": 1.05259871,
      "balance_loss_mlp": 1.02225733,
      "epoch": 0.40125052606264655,
      "flos": 20668027841280.0,
      "grad_norm": 1.8403792208800422,
      "language_loss": 0.74809957,
      "learning_rate": 2.7203464781201523e-06,
      "loss": 0.76981556,
      "num_input_tokens_seen": 72096490,
      "step": 3337,
      "time_per_iteration": 3.7842981815338135
    },
    {
      "auxiliary_loss_clip": 0.01185894,
      "auxiliary_loss_mlp": 0.01033408,
      "balance_loss_clip": 1.05662525,
      "balance_loss_mlp": 1.02587438,
      "epoch": 0.40137076895328566,
      "flos": 24607679541120.0,
      "grad_norm": 2.043070867849423,
      "language_loss": 0.78008741,
      "learning_rate": 2.719619732195215e-06,
      "loss": 0.80228043,
      "num_input_tokens_seen": 72118130,
      "step": 3338,
      "time_per_iteration": 2.618217706680298
    },
    {
      "auxiliary_loss_clip": 0.01142262,
      "auxiliary_loss_mlp": 0.01034634,
      "balance_loss_clip": 1.05210829,
      "balance_loss_mlp": 1.02658105,
      "epoch": 0.4014910118439247,
      "flos": 24206593299840.0,
      "grad_norm": 1.3947879171429154,
      "language_loss": 0.72701442,
      "learning_rate": 2.7188928771045377e-06,
      "loss": 0.74878335,
      "num_input_tokens_seen": 72139450,
      "step": 3339,
      "time_per_iteration": 2.6293091773986816
    },
    {
      "auxiliary_loss_clip": 0.011374,
      "auxiliary_loss_mlp": 0.0102858,
      "balance_loss_clip": 1.04968345,
      "balance_loss_mlp": 1.02026486,
      "epoch": 0.4016112547345638,
      "flos": 26725080418560.0,
      "grad_norm": 1.60668316265841,
      "language_loss": 0.80158603,
      "learning_rate": 2.7181659129583815e-06,
      "loss": 0.82324582,
      "num_input_tokens_seen": 72159040,
      "step": 3340,
      "time_per_iteration": 2.6706857681274414
    },
    {
      "auxiliary_loss_clip": 0.01146988,
      "auxiliary_loss_mlp": 0.01024279,
      "balance_loss_clip": 1.04792571,
      "balance_loss_mlp": 1.01620317,
      "epoch": 0.4017314976252029,
      "flos": 21288025520640.0,
      "grad_norm": 1.6364106908211977,
      "language_loss": 0.7549026,
      "learning_rate": 2.7174388398670276e-06,
      "loss": 0.77661526,
      "num_input_tokens_seen": 72178220,
      "step": 3341,
      "time_per_iteration": 2.577338933944702
    },
    {
      "auxiliary_loss_clip": 0.01182108,
      "auxiliary_loss_mlp": 0.01028732,
      "balance_loss_clip": 1.05257559,
      "balance_loss_mlp": 1.02037537,
      "epoch": 0.401851740515842,
      "flos": 25484690010240.0,
      "grad_norm": 1.9986596620399,
      "language_loss": 0.91736746,
      "learning_rate": 2.716711657940773e-06,
      "loss": 0.93947577,
      "num_input_tokens_seen": 72199230,
      "step": 3342,
      "time_per_iteration": 4.088343620300293
    },
    {
      "auxiliary_loss_clip": 0.01039451,
      "auxiliary_loss_mlp": 0.01002545,
      "balance_loss_clip": 1.0159111,
      "balance_loss_mlp": 1.00129974,
      "epoch": 0.4019719834064811,
      "flos": 55395334978560.0,
      "grad_norm": 0.8148904899020343,
      "language_loss": 0.5656327,
      "learning_rate": 2.7159843672899284e-06,
      "loss": 0.58605266,
      "num_input_tokens_seen": 72263430,
      "step": 3343,
      "time_per_iteration": 4.11398458480835
    },
    {
      "auxiliary_loss_clip": 0.01174022,
      "auxiliary_loss_mlp": 0.01030054,
      "balance_loss_clip": 1.05736732,
      "balance_loss_mlp": 1.02201939,
      "epoch": 0.40209222629712016,
      "flos": 18180100218240.0,
      "grad_norm": 1.9537745856659812,
      "language_loss": 0.81331533,
      "learning_rate": 2.715256968024825e-06,
      "loss": 0.83535612,
      "num_input_tokens_seen": 72280505,
      "step": 3344,
      "time_per_iteration": 2.516437292098999
    },
    {
      "auxiliary_loss_clip": 0.01161655,
      "auxiliary_loss_mlp": 0.01031792,
      "balance_loss_clip": 1.05416083,
      "balance_loss_mlp": 1.02399015,
      "epoch": 0.40221246918775927,
      "flos": 25961009287680.0,
      "grad_norm": 1.7446634914750148,
      "language_loss": 0.82099319,
      "learning_rate": 2.7145294602558083e-06,
      "loss": 0.84292769,
      "num_input_tokens_seen": 72301215,
      "step": 3345,
      "time_per_iteration": 2.6400160789489746
    },
    {
      "auxiliary_loss_clip": 0.0117163,
      "auxiliary_loss_mlp": 0.01029109,
      "balance_loss_clip": 1.0546658,
      "balance_loss_mlp": 1.02030516,
      "epoch": 0.4023327120783984,
      "flos": 33838912056960.0,
      "grad_norm": 1.8938298206673478,
      "language_loss": 0.70647985,
      "learning_rate": 2.713801844093241e-06,
      "loss": 0.72848725,
      "num_input_tokens_seen": 72322365,
      "step": 3346,
      "time_per_iteration": 2.6522574424743652
    },
    {
      "auxiliary_loss_clip": 0.01171881,
      "auxiliary_loss_mlp": 0.01026741,
      "balance_loss_clip": 1.05360007,
      "balance_loss_mlp": 1.01845562,
      "epoch": 0.40245295496903744,
      "flos": 26900252069760.0,
      "grad_norm": 1.9235033156095243,
      "language_loss": 0.88549435,
      "learning_rate": 2.7130741196475014e-06,
      "loss": 0.90748054,
      "num_input_tokens_seen": 72340495,
      "step": 3347,
      "time_per_iteration": 2.5933449268341064
    },
    {
      "auxiliary_loss_clip": 0.01159105,
      "auxiliary_loss_mlp": 0.01028449,
      "balance_loss_clip": 1.05608964,
      "balance_loss_mlp": 1.02011597,
      "epoch": 0.40257319785967655,
      "flos": 36902738436480.0,
      "grad_norm": 1.8713785663782259,
      "language_loss": 0.79032999,
      "learning_rate": 2.7123462870289848e-06,
      "loss": 0.81220555,
      "num_input_tokens_seen": 72360545,
      "step": 3348,
      "time_per_iteration": 2.696370840072632
    },
    {
      "auxiliary_loss_clip": 0.01159482,
      "auxiliary_loss_mlp": 0.01029999,
      "balance_loss_clip": 1.05209994,
      "balance_loss_mlp": 1.02203608,
      "epoch": 0.40269344075031566,
      "flos": 24353180703360.0,
      "grad_norm": 1.6503283992832696,
      "language_loss": 0.80953014,
      "learning_rate": 2.711618346348102e-06,
      "loss": 0.83142495,
      "num_input_tokens_seen": 72381070,
      "step": 3349,
      "time_per_iteration": 2.617626190185547
    },
    {
      "auxiliary_loss_clip": 0.01153183,
      "auxiliary_loss_mlp": 0.01031329,
      "balance_loss_clip": 1.05446875,
      "balance_loss_mlp": 1.02366388,
      "epoch": 0.4028136836409547,
      "flos": 14389657614720.0,
      "grad_norm": 2.562961298300526,
      "language_loss": 0.63350964,
      "learning_rate": 2.7108902977152825e-06,
      "loss": 0.65535474,
      "num_input_tokens_seen": 72398970,
      "step": 3350,
      "time_per_iteration": 2.582167863845825
    },
    {
      "auxiliary_loss_clip": 0.01169044,
      "auxiliary_loss_mlp": 0.01033643,
      "balance_loss_clip": 1.05374706,
      "balance_loss_mlp": 1.02502418,
      "epoch": 0.4029339265315938,
      "flos": 26136037284480.0,
      "grad_norm": 2.602953884498685,
      "language_loss": 0.74741685,
      "learning_rate": 2.7101621412409704e-06,
      "loss": 0.76944375,
      "num_input_tokens_seen": 72418455,
      "step": 3351,
      "time_per_iteration": 2.602501153945923
    },
    {
      "auxiliary_loss_clip": 0.01186417,
      "auxiliary_loss_mlp": 0.01026525,
      "balance_loss_clip": 1.05664051,
      "balance_loss_mlp": 1.01819241,
      "epoch": 0.40305416942223293,
      "flos": 23256325042560.0,
      "grad_norm": 1.7932023289001278,
      "language_loss": 0.85906827,
      "learning_rate": 2.7094338770356256e-06,
      "loss": 0.88119769,
      "num_input_tokens_seen": 72437540,
      "step": 3352,
      "time_per_iteration": 2.5133907794952393
    },
    {
      "auxiliary_loss_clip": 0.01153097,
      "auxiliary_loss_mlp": 0.01030257,
      "balance_loss_clip": 1.05299795,
      "balance_loss_mlp": 1.02175784,
      "epoch": 0.403174412312872,
      "flos": 27089645506560.0,
      "grad_norm": 2.1525940304014806,
      "language_loss": 0.64131194,
      "learning_rate": 2.708705505209726e-06,
      "loss": 0.66314554,
      "num_input_tokens_seen": 72458315,
      "step": 3353,
      "time_per_iteration": 2.646333694458008
    },
    {
      "auxiliary_loss_clip": 0.01121275,
      "auxiliary_loss_mlp": 0.01038338,
      "balance_loss_clip": 1.04793608,
      "balance_loss_mlp": 1.03017175,
      "epoch": 0.4032946552035111,
      "flos": 21756336065280.0,
      "grad_norm": 2.3315895273689895,
      "language_loss": 0.91552693,
      "learning_rate": 2.7079770258737646e-06,
      "loss": 0.93712306,
      "num_input_tokens_seen": 72476225,
      "step": 3354,
      "time_per_iteration": 2.663471221923828
    },
    {
      "auxiliary_loss_clip": 0.01138265,
      "auxiliary_loss_mlp": 0.01032407,
      "balance_loss_clip": 1.04902697,
      "balance_loss_mlp": 1.02328706,
      "epoch": 0.4034148980941502,
      "flos": 17343956448000.0,
      "grad_norm": 2.2391837588450767,
      "language_loss": 0.75504506,
      "learning_rate": 2.707248439138251e-06,
      "loss": 0.77675176,
      "num_input_tokens_seen": 72492460,
      "step": 3355,
      "time_per_iteration": 2.6243205070495605
    },
    {
      "auxiliary_loss_clip": 0.01154284,
      "auxiliary_loss_mlp": 0.01027423,
      "balance_loss_clip": 1.05685687,
      "balance_loss_mlp": 1.01915026,
      "epoch": 0.40353514098478926,
      "flos": 22017838055040.0,
      "grad_norm": 1.6865313690719599,
      "language_loss": 0.6512751,
      "learning_rate": 2.7065197451137114e-06,
      "loss": 0.67309225,
      "num_input_tokens_seen": 72513840,
      "step": 3356,
      "time_per_iteration": 2.6342687606811523
    },
    {
      "auxiliary_loss_clip": 0.01160386,
      "auxiliary_loss_mlp": 0.01037791,
      "balance_loss_clip": 1.05631387,
      "balance_loss_mlp": 1.02973199,
      "epoch": 0.4036553838754284,
      "flos": 14246446089600.0,
      "grad_norm": 1.8832571306180066,
      "language_loss": 0.67507762,
      "learning_rate": 2.7057909439106894e-06,
      "loss": 0.69705939,
      "num_input_tokens_seen": 72531695,
      "step": 3357,
      "time_per_iteration": 2.573958158493042
    },
    {
      "auxiliary_loss_clip": 0.01161151,
      "auxiliary_loss_mlp": 0.00761934,
      "balance_loss_clip": 1.05203533,
      "balance_loss_mlp": 1.00032747,
      "epoch": 0.40377562676606743,
      "flos": 24790644443520.0,
      "grad_norm": 2.6948036368492088,
      "language_loss": 0.78594089,
      "learning_rate": 2.7050620356397417e-06,
      "loss": 0.80517173,
      "num_input_tokens_seen": 72550645,
      "step": 3358,
      "time_per_iteration": 2.581328868865967
    },
    {
      "auxiliary_loss_clip": 0.01185296,
      "auxiliary_loss_mlp": 0.01033901,
      "balance_loss_clip": 1.05949938,
      "balance_loss_mlp": 1.02621472,
      "epoch": 0.40389586965670654,
      "flos": 24061226958720.0,
      "grad_norm": 1.6418431255661425,
      "language_loss": 0.72153109,
      "learning_rate": 2.7043330204114437e-06,
      "loss": 0.74372303,
      "num_input_tokens_seen": 72569355,
      "step": 3359,
      "time_per_iteration": 2.517836570739746
    },
    {
      "auxiliary_loss_clip": 0.01179626,
      "auxiliary_loss_mlp": 0.01022579,
      "balance_loss_clip": 1.05387998,
      "balance_loss_mlp": 1.01471734,
      "epoch": 0.40401611254734565,
      "flos": 16399613934720.0,
      "grad_norm": 2.2536660953669942,
      "language_loss": 0.8557322,
      "learning_rate": 2.7036038983363862e-06,
      "loss": 0.87775427,
      "num_input_tokens_seen": 72585960,
      "step": 3360,
      "time_per_iteration": 2.479858160018921
    },
    {
      "auxiliary_loss_clip": 0.01165493,
      "auxiliary_loss_mlp": 0.01028238,
      "balance_loss_clip": 1.05440688,
      "balance_loss_mlp": 1.02068579,
      "epoch": 0.4041363554379847,
      "flos": 23988220565760.0,
      "grad_norm": 1.8266279295811543,
      "language_loss": 0.84515607,
      "learning_rate": 2.702874669525177e-06,
      "loss": 0.86709332,
      "num_input_tokens_seen": 72604440,
      "step": 3361,
      "time_per_iteration": 2.5679666996002197
    },
    {
      "auxiliary_loss_clip": 0.01140642,
      "auxiliary_loss_mlp": 0.01026162,
      "balance_loss_clip": 1.05202353,
      "balance_loss_mlp": 1.01885188,
      "epoch": 0.4042565983286238,
      "flos": 28401964899840.0,
      "grad_norm": 1.8339099395275695,
      "language_loss": 0.69689,
      "learning_rate": 2.7021453340884394e-06,
      "loss": 0.71855801,
      "num_input_tokens_seen": 72622165,
      "step": 3362,
      "time_per_iteration": 2.6638176441192627
    },
    {
      "auxiliary_loss_clip": 0.01144957,
      "auxiliary_loss_mlp": 0.00762084,
      "balance_loss_clip": 1.05172431,
      "balance_loss_mlp": 1.00031936,
      "epoch": 0.40437684121926293,
      "flos": 17710963660800.0,
      "grad_norm": 2.0523532767642374,
      "language_loss": 0.72985411,
      "learning_rate": 2.7014158921368125e-06,
      "loss": 0.74892449,
      "num_input_tokens_seen": 72640490,
      "step": 3363,
      "time_per_iteration": 3.30672550201416
    },
    {
      "auxiliary_loss_clip": 0.01183736,
      "auxiliary_loss_mlp": 0.01033236,
      "balance_loss_clip": 1.05586767,
      "balance_loss_mlp": 1.02553499,
      "epoch": 0.404497084109902,
      "flos": 24018959629440.0,
      "grad_norm": 1.7684909724831732,
      "language_loss": 0.85333055,
      "learning_rate": 2.700686343780953e-06,
      "loss": 0.87550026,
      "num_input_tokens_seen": 72660360,
      "step": 3364,
      "time_per_iteration": 2.5398807525634766
    },
    {
      "auxiliary_loss_clip": 0.01155766,
      "auxiliary_loss_mlp": 0.01029685,
      "balance_loss_clip": 1.05119228,
      "balance_loss_mlp": 1.0219934,
      "epoch": 0.4046173270005411,
      "flos": 22929861306240.0,
      "grad_norm": 2.090622306732791,
      "language_loss": 0.88286197,
      "learning_rate": 2.699956689131532e-06,
      "loss": 0.90471649,
      "num_input_tokens_seen": 72680345,
      "step": 3365,
      "time_per_iteration": 2.5826849937438965
    },
    {
      "auxiliary_loss_clip": 0.01156121,
      "auxiliary_loss_mlp": 0.01030659,
      "balance_loss_clip": 1.05171084,
      "balance_loss_mlp": 1.02319098,
      "epoch": 0.4047375698911802,
      "flos": 20668135582080.0,
      "grad_norm": 2.481856520410991,
      "language_loss": 0.84837413,
      "learning_rate": 2.699226928299238e-06,
      "loss": 0.870242,
      "num_input_tokens_seen": 72698365,
      "step": 3366,
      "time_per_iteration": 2.5970826148986816
    },
    {
      "auxiliary_loss_clip": 0.01168874,
      "auxiliary_loss_mlp": 0.01023436,
      "balance_loss_clip": 1.05396962,
      "balance_loss_mlp": 1.01584864,
      "epoch": 0.40485781278181926,
      "flos": 28912865996160.0,
      "grad_norm": 2.1316078681279684,
      "language_loss": 0.78825426,
      "learning_rate": 2.698497061394774e-06,
      "loss": 0.81017733,
      "num_input_tokens_seen": 72716850,
      "step": 3367,
      "time_per_iteration": 2.594989538192749
    },
    {
      "auxiliary_loss_clip": 0.01147468,
      "auxiliary_loss_mlp": 0.00761659,
      "balance_loss_clip": 1.05406976,
      "balance_loss_mlp": 1.00029325,
      "epoch": 0.40497805567245837,
      "flos": 23148377694720.0,
      "grad_norm": 1.6192981142479435,
      "language_loss": 0.80655795,
      "learning_rate": 2.6977670885288627e-06,
      "loss": 0.82564926,
      "num_input_tokens_seen": 72738250,
      "step": 3368,
      "time_per_iteration": 3.446361780166626
    },
    {
      "auxiliary_loss_clip": 0.01144108,
      "auxiliary_loss_mlp": 0.01030713,
      "balance_loss_clip": 1.04925954,
      "balance_loss_mlp": 1.02263618,
      "epoch": 0.4050982985630975,
      "flos": 16289404030080.0,
      "grad_norm": 2.290976709221643,
      "language_loss": 0.75061226,
      "learning_rate": 2.6970370098122378e-06,
      "loss": 0.77236044,
      "num_input_tokens_seen": 72755235,
      "step": 3369,
      "time_per_iteration": 3.3680992126464844
    },
    {
      "auxiliary_loss_clip": 0.01182233,
      "auxiliary_loss_mlp": 0.0102394,
      "balance_loss_clip": 1.05521059,
      "balance_loss_mlp": 1.01633418,
      "epoch": 0.40521854145373654,
      "flos": 34459484353920.0,
      "grad_norm": 1.4719927728825715,
      "language_loss": 0.86352313,
      "learning_rate": 2.6963068253556535e-06,
      "loss": 0.88558483,
      "num_input_tokens_seen": 72776620,
      "step": 3370,
      "time_per_iteration": 2.6271915435791016
    },
    {
      "auxiliary_loss_clip": 0.0117408,
      "auxiliary_loss_mlp": 0.01028726,
      "balance_loss_clip": 1.0530355,
      "balance_loss_mlp": 1.02022314,
      "epoch": 0.40533878434437565,
      "flos": 25331099454720.0,
      "grad_norm": 1.9416559640622753,
      "language_loss": 0.85777557,
      "learning_rate": 2.6955765352698763e-06,
      "loss": 0.8798036,
      "num_input_tokens_seen": 72796765,
      "step": 3371,
      "time_per_iteration": 2.6010568141937256
    },
    {
      "auxiliary_loss_clip": 0.01183495,
      "auxiliary_loss_mlp": 0.01032936,
      "balance_loss_clip": 1.05358016,
      "balance_loss_mlp": 1.02455008,
      "epoch": 0.40545902723501476,
      "flos": 15012061505280.0,
      "grad_norm": 10.242946906854163,
      "language_loss": 0.73306239,
      "learning_rate": 2.6948461396656923e-06,
      "loss": 0.75522673,
      "num_input_tokens_seen": 72814175,
      "step": 3372,
      "time_per_iteration": 2.5100531578063965
    },
    {
      "auxiliary_loss_clip": 0.01171231,
      "auxiliary_loss_mlp": 0.01031813,
      "balance_loss_clip": 1.05301082,
      "balance_loss_mlp": 1.02395439,
      "epoch": 0.4055792701256538,
      "flos": 25521103422720.0,
      "grad_norm": 2.4588589273752044,
      "language_loss": 0.74463207,
      "learning_rate": 2.6941156386539013e-06,
      "loss": 0.76666248,
      "num_input_tokens_seen": 72834125,
      "step": 3373,
      "time_per_iteration": 2.6117379665374756
    },
    {
      "auxiliary_loss_clip": 0.01154913,
      "auxiliary_loss_mlp": 0.01036591,
      "balance_loss_clip": 1.0575881,
      "balance_loss_mlp": 1.02832997,
      "epoch": 0.4056995130162929,
      "flos": 19574583972480.0,
      "grad_norm": 7.214921525087993,
      "language_loss": 0.81089365,
      "learning_rate": 2.6933850323453203e-06,
      "loss": 0.83280867,
      "num_input_tokens_seen": 72852570,
      "step": 3374,
      "time_per_iteration": 2.5762343406677246
    },
    {
      "auxiliary_loss_clip": 0.01184664,
      "auxiliary_loss_mlp": 0.01029892,
      "balance_loss_clip": 1.05804932,
      "balance_loss_mlp": 1.02194679,
      "epoch": 0.405819755906932,
      "flos": 15413794191360.0,
      "grad_norm": 1.8213181567669736,
      "language_loss": 0.74542809,
      "learning_rate": 2.6926543208507806e-06,
      "loss": 0.76757365,
      "num_input_tokens_seen": 72871250,
      "step": 3375,
      "time_per_iteration": 2.527643918991089
    },
    {
      "auxiliary_loss_clip": 0.01171067,
      "auxiliary_loss_mlp": 0.01026178,
      "balance_loss_clip": 1.05571556,
      "balance_loss_mlp": 1.01831627,
      "epoch": 0.4059399987975711,
      "flos": 21433930565760.0,
      "grad_norm": 1.9899567321470668,
      "language_loss": 0.79851091,
      "learning_rate": 2.6919235042811316e-06,
      "loss": 0.82048339,
      "num_input_tokens_seen": 72890035,
      "step": 3376,
      "time_per_iteration": 2.526902675628662
    },
    {
      "auxiliary_loss_clip": 0.01135962,
      "auxiliary_loss_mlp": 0.0103283,
      "balance_loss_clip": 1.04816639,
      "balance_loss_mlp": 1.02508712,
      "epoch": 0.4060602416882102,
      "flos": 25556942217600.0,
      "grad_norm": 2.017413014360003,
      "language_loss": 0.76488125,
      "learning_rate": 2.691192582747237e-06,
      "loss": 0.78656918,
      "num_input_tokens_seen": 72909665,
      "step": 3377,
      "time_per_iteration": 2.615504264831543
    },
    {
      "auxiliary_loss_clip": 0.01181551,
      "auxiliary_loss_mlp": 0.01028946,
      "balance_loss_clip": 1.05399084,
      "balance_loss_mlp": 1.0206964,
      "epoch": 0.40618048457884925,
      "flos": 23766759262080.0,
      "grad_norm": 1.7520978756261272,
      "language_loss": 0.73949784,
      "learning_rate": 2.6904615563599765e-06,
      "loss": 0.76160282,
      "num_input_tokens_seen": 72929465,
      "step": 3378,
      "time_per_iteration": 2.5301167964935303
    },
    {
      "auxiliary_loss_clip": 0.01135039,
      "auxiliary_loss_mlp": 0.01026069,
      "balance_loss_clip": 1.04868615,
      "balance_loss_mlp": 1.01782,
      "epoch": 0.40630072746948837,
      "flos": 17639681120640.0,
      "grad_norm": 1.7204841176003094,
      "language_loss": 0.83438599,
      "learning_rate": 2.6897304252302477e-06,
      "loss": 0.85599709,
      "num_input_tokens_seen": 72946785,
      "step": 3379,
      "time_per_iteration": 2.5903613567352295
    },
    {
      "auxiliary_loss_clip": 0.01038265,
      "auxiliary_loss_mlp": 0.01005489,
      "balance_loss_clip": 1.01484084,
      "balance_loss_mlp": 1.00416553,
      "epoch": 0.4064209703601275,
      "flos": 60836053063680.0,
      "grad_norm": 0.785068124290811,
      "language_loss": 0.54838824,
      "learning_rate": 2.688999189468962e-06,
      "loss": 0.56882572,
      "num_input_tokens_seen": 73003215,
      "step": 3380,
      "time_per_iteration": 3.03806471824646
    },
    {
      "auxiliary_loss_clip": 0.01165651,
      "auxiliary_loss_mlp": 0.01026091,
      "balance_loss_clip": 1.05370843,
      "balance_loss_mlp": 1.01821089,
      "epoch": 0.40654121325076653,
      "flos": 24024346669440.0,
      "grad_norm": 2.9981274587368683,
      "language_loss": 0.75665265,
      "learning_rate": 2.6882678491870464e-06,
      "loss": 0.77857006,
      "num_input_tokens_seen": 73023650,
      "step": 3381,
      "time_per_iteration": 2.551246166229248
    },
    {
      "auxiliary_loss_clip": 0.01171999,
      "auxiliary_loss_mlp": 0.01027512,
      "balance_loss_clip": 1.05359077,
      "balance_loss_mlp": 1.0190835,
      "epoch": 0.40666145614140564,
      "flos": 27344252085120.0,
      "grad_norm": 1.868055768090866,
      "language_loss": 0.71176028,
      "learning_rate": 2.6875364044954453e-06,
      "loss": 0.73375535,
      "num_input_tokens_seen": 73043880,
      "step": 3382,
      "time_per_iteration": 2.5642755031585693
    },
    {
      "auxiliary_loss_clip": 0.01149895,
      "auxiliary_loss_mlp": 0.0102966,
      "balance_loss_clip": 1.04704177,
      "balance_loss_mlp": 1.02249002,
      "epoch": 0.40678169903204475,
      "flos": 26176724415360.0,
      "grad_norm": 1.604990727285972,
      "language_loss": 0.82226598,
      "learning_rate": 2.6868048555051185e-06,
      "loss": 0.84406149,
      "num_input_tokens_seen": 73065410,
      "step": 3383,
      "time_per_iteration": 2.623302698135376
    },
    {
      "auxiliary_loss_clip": 0.01159318,
      "auxiliary_loss_mlp": 0.01029995,
      "balance_loss_clip": 1.04959917,
      "balance_loss_mlp": 1.02158451,
      "epoch": 0.4069019419226838,
      "flos": 28622420622720.0,
      "grad_norm": 2.882999968564185,
      "language_loss": 0.85455835,
      "learning_rate": 2.686073202327041e-06,
      "loss": 0.87645149,
      "num_input_tokens_seen": 73084410,
      "step": 3384,
      "time_per_iteration": 2.6350769996643066
    },
    {
      "auxiliary_loss_clip": 0.01140327,
      "auxiliary_loss_mlp": 0.01025861,
      "balance_loss_clip": 1.04533112,
      "balance_loss_mlp": 1.01810622,
      "epoch": 0.4070221848133229,
      "flos": 25229006023680.0,
      "grad_norm": 2.1537236730235514,
      "language_loss": 0.73322797,
      "learning_rate": 2.6853414450722043e-06,
      "loss": 0.75488985,
      "num_input_tokens_seen": 73104075,
      "step": 3385,
      "time_per_iteration": 2.6010942459106445
    },
    {
      "auxiliary_loss_clip": 0.01165518,
      "auxiliary_loss_mlp": 0.01028665,
      "balance_loss_clip": 1.05228591,
      "balance_loss_mlp": 1.02088356,
      "epoch": 0.40714242770396203,
      "flos": 18405224709120.0,
      "grad_norm": 1.7702142282644202,
      "language_loss": 0.8534984,
      "learning_rate": 2.684609583851616e-06,
      "loss": 0.87544024,
      "num_input_tokens_seen": 73122250,
      "step": 3386,
      "time_per_iteration": 2.518399953842163
    },
    {
      "auxiliary_loss_clip": 0.01125395,
      "auxiliary_loss_mlp": 0.01033015,
      "balance_loss_clip": 1.04804468,
      "balance_loss_mlp": 1.02492714,
      "epoch": 0.4072626705946011,
      "flos": 30228920403840.0,
      "grad_norm": 1.6014395355827715,
      "language_loss": 0.80749857,
      "learning_rate": 2.683877618776297e-06,
      "loss": 0.82908267,
      "num_input_tokens_seen": 73144505,
      "step": 3387,
      "time_per_iteration": 2.7148594856262207
    },
    {
      "auxiliary_loss_clip": 0.01147772,
      "auxiliary_loss_mlp": 0.01038862,
      "balance_loss_clip": 1.04895973,
      "balance_loss_mlp": 1.03089917,
      "epoch": 0.4073829134852402,
      "flos": 21834549930240.0,
      "grad_norm": 1.9840971678146528,
      "language_loss": 0.73917007,
      "learning_rate": 2.6831455499572876e-06,
      "loss": 0.76103652,
      "num_input_tokens_seen": 73162440,
      "step": 3388,
      "time_per_iteration": 2.6191625595092773
    },
    {
      "auxiliary_loss_clip": 0.01180261,
      "auxiliary_loss_mlp": 0.01031141,
      "balance_loss_clip": 1.05270314,
      "balance_loss_mlp": 1.02314782,
      "epoch": 0.40750315637587925,
      "flos": 25260211964160.0,
      "grad_norm": 1.8940229924493592,
      "language_loss": 0.77698088,
      "learning_rate": 2.682413377505641e-06,
      "loss": 0.79909492,
      "num_input_tokens_seen": 73181245,
      "step": 3389,
      "time_per_iteration": 3.3619837760925293
    },
    {
      "auxiliary_loss_clip": 0.01168284,
      "auxiliary_loss_mlp": 0.01029586,
      "balance_loss_clip": 1.05195904,
      "balance_loss_mlp": 1.021909,
      "epoch": 0.40762339926651836,
      "flos": 19712767593600.0,
      "grad_norm": 1.767825487476804,
      "language_loss": 0.76737332,
      "learning_rate": 2.6816811015324284e-06,
      "loss": 0.789352,
      "num_input_tokens_seen": 73199295,
      "step": 3390,
      "time_per_iteration": 2.5624876022338867
    },
    {
      "auxiliary_loss_clip": 0.01073336,
      "auxiliary_loss_mlp": 0.01002426,
      "balance_loss_clip": 1.01714039,
      "balance_loss_mlp": 1.00111437,
      "epoch": 0.40774364215715747,
      "flos": 71449307314560.0,
      "grad_norm": 0.7242210274620713,
      "language_loss": 0.56799078,
      "learning_rate": 2.6809487221487343e-06,
      "loss": 0.58874846,
      "num_input_tokens_seen": 73258780,
      "step": 3391,
      "time_per_iteration": 3.0162415504455566
    },
    {
      "auxiliary_loss_clip": 0.01159245,
      "auxiliary_loss_mlp": 0.01028958,
      "balance_loss_clip": 1.05013514,
      "balance_loss_mlp": 1.02137637,
      "epoch": 0.4078638850477965,
      "flos": 15084134144640.0,
      "grad_norm": 2.3985611486461673,
      "language_loss": 0.81879199,
      "learning_rate": 2.6802162394656605e-06,
      "loss": 0.84067404,
      "num_input_tokens_seen": 73275490,
      "step": 3392,
      "time_per_iteration": 2.5482261180877686
    },
    {
      "auxiliary_loss_clip": 0.0114651,
      "auxiliary_loss_mlp": 0.01032548,
      "balance_loss_clip": 1.04619575,
      "balance_loss_mlp": 1.02471554,
      "epoch": 0.40798412793843564,
      "flos": 23842890138240.0,
      "grad_norm": 1.6658248518561067,
      "language_loss": 0.71785086,
      "learning_rate": 2.679483653594324e-06,
      "loss": 0.73964143,
      "num_input_tokens_seen": 73297260,
      "step": 3393,
      "time_per_iteration": 2.5991175174713135
    },
    {
      "auxiliary_loss_clip": 0.01169868,
      "auxiliary_loss_mlp": 0.01025578,
      "balance_loss_clip": 1.05210996,
      "balance_loss_mlp": 1.01804113,
      "epoch": 0.40810437082907475,
      "flos": 21065774117760.0,
      "grad_norm": 3.4762855647919446,
      "language_loss": 0.76505554,
      "learning_rate": 2.678750964645857e-06,
      "loss": 0.78701001,
      "num_input_tokens_seen": 73316340,
      "step": 3394,
      "time_per_iteration": 3.3758387565612793
    },
    {
      "auxiliary_loss_clip": 0.01168864,
      "auxiliary_loss_mlp": 0.01035714,
      "balance_loss_clip": 1.05544674,
      "balance_loss_mlp": 1.02759564,
      "epoch": 0.4082246137197138,
      "flos": 11321377948800.0,
      "grad_norm": 2.4430279004594273,
      "language_loss": 0.83579648,
      "learning_rate": 2.6780181727314094e-06,
      "loss": 0.85784227,
      "num_input_tokens_seen": 73331245,
      "step": 3395,
      "time_per_iteration": 3.260948657989502
    },
    {
      "auxiliary_loss_clip": 0.011448,
      "auxiliary_loss_mlp": 0.00761816,
      "balance_loss_clip": 1.05045485,
      "balance_loss_mlp": 1.00033236,
      "epoch": 0.4083448566103529,
      "flos": 19062569554560.0,
      "grad_norm": 1.7432751154832158,
      "language_loss": 0.78004408,
      "learning_rate": 2.6772852779621435e-06,
      "loss": 0.79911023,
      "num_input_tokens_seen": 73349105,
      "step": 3396,
      "time_per_iteration": 2.5779008865356445
    },
    {
      "auxiliary_loss_clip": 0.01161512,
      "auxiliary_loss_mlp": 0.00761111,
      "balance_loss_clip": 1.05363882,
      "balance_loss_mlp": 1.00033188,
      "epoch": 0.408465099500992,
      "flos": 23550254035200.0,
      "grad_norm": 1.968634378698742,
      "language_loss": 0.85942686,
      "learning_rate": 2.676552280449239e-06,
      "loss": 0.87865317,
      "num_input_tokens_seen": 73368990,
      "step": 3397,
      "time_per_iteration": 2.5815048217773438
    },
    {
      "auxiliary_loss_clip": 0.01158546,
      "auxiliary_loss_mlp": 0.01031427,
      "balance_loss_clip": 1.05049431,
      "balance_loss_mlp": 1.0236789,
      "epoch": 0.4085853423916311,
      "flos": 12750012558720.0,
      "grad_norm": 3.3297550646975935,
      "language_loss": 0.75370276,
      "learning_rate": 2.6758191803038917e-06,
      "loss": 0.77560246,
      "num_input_tokens_seen": 73387485,
      "step": 3398,
      "time_per_iteration": 2.5032663345336914
    },
    {
      "auxiliary_loss_clip": 0.01105249,
      "auxiliary_loss_mlp": 0.01033071,
      "balance_loss_clip": 1.05013204,
      "balance_loss_mlp": 1.02508998,
      "epoch": 0.4087055852822702,
      "flos": 24353072962560.0,
      "grad_norm": 1.7441885885953403,
      "language_loss": 0.82493645,
      "learning_rate": 2.6750859776373125e-06,
      "loss": 0.84631968,
      "num_input_tokens_seen": 73406940,
      "step": 3399,
      "time_per_iteration": 2.6794519424438477
    },
    {
      "auxiliary_loss_clip": 0.01025883,
      "auxiliary_loss_mlp": 0.01001615,
      "balance_loss_clip": 1.0183444,
      "balance_loss_mlp": 1.00028598,
      "epoch": 0.4088258281729093,
      "flos": 66387950720640.0,
      "grad_norm": 0.7683290533082459,
      "language_loss": 0.60432839,
      "learning_rate": 2.674352672560727e-06,
      "loss": 0.62460339,
      "num_input_tokens_seen": 73468385,
      "step": 3400,
      "time_per_iteration": 3.2302324771881104
    },
    {
      "auxiliary_loss_clip": 0.01137777,
      "auxiliary_loss_mlp": 0.01032986,
      "balance_loss_clip": 1.04827857,
      "balance_loss_mlp": 1.0249033,
      "epoch": 0.40894607106354836,
      "flos": 20449260057600.0,
      "grad_norm": 1.6351414428473174,
      "language_loss": 0.77110922,
      "learning_rate": 2.673619265185377e-06,
      "loss": 0.79281682,
      "num_input_tokens_seen": 73488225,
      "step": 3401,
      "time_per_iteration": 2.6056385040283203
    },
    {
      "auxiliary_loss_clip": 0.01168916,
      "auxiliary_loss_mlp": 0.01033712,
      "balance_loss_clip": 1.0521996,
      "balance_loss_mlp": 1.02506959,
      "epoch": 0.40906631395418747,
      "flos": 27053627143680.0,
      "grad_norm": 2.235404804621795,
      "language_loss": 0.77915442,
      "learning_rate": 2.672885755622521e-06,
      "loss": 0.80118072,
      "num_input_tokens_seen": 73510640,
      "step": 3402,
      "time_per_iteration": 2.6092770099639893
    },
    {
      "auxiliary_loss_clip": 0.01125264,
      "auxiliary_loss_mlp": 0.01029257,
      "balance_loss_clip": 1.04774094,
      "balance_loss_mlp": 1.0218122,
      "epoch": 0.4091865568448266,
      "flos": 25484151306240.0,
      "grad_norm": 2.0628766122999025,
      "language_loss": 0.70550513,
      "learning_rate": 2.67215214398343e-06,
      "loss": 0.7270503,
      "num_input_tokens_seen": 73530655,
      "step": 3403,
      "time_per_iteration": 2.646634101867676
    },
    {
      "auxiliary_loss_clip": 0.01129305,
      "auxiliary_loss_mlp": 0.01033645,
      "balance_loss_clip": 1.04636335,
      "balance_loss_mlp": 1.02543211,
      "epoch": 0.40930679973546563,
      "flos": 28657864368000.0,
      "grad_norm": 2.651354143430776,
      "language_loss": 0.78181088,
      "learning_rate": 2.671418430379393e-06,
      "loss": 0.80344045,
      "num_input_tokens_seen": 73549340,
      "step": 3404,
      "time_per_iteration": 2.695010185241699
    },
    {
      "auxiliary_loss_clip": 0.01181754,
      "auxiliary_loss_mlp": 0.010284,
      "balance_loss_clip": 1.05332351,
      "balance_loss_mlp": 1.02088976,
      "epoch": 0.40942704262610474,
      "flos": 20886292834560.0,
      "grad_norm": 1.85514036255539,
      "language_loss": 0.83606291,
      "learning_rate": 2.670684614921715e-06,
      "loss": 0.85816443,
      "num_input_tokens_seen": 73568315,
      "step": 3405,
      "time_per_iteration": 2.4935150146484375
    },
    {
      "auxiliary_loss_clip": 0.01155089,
      "auxiliary_loss_mlp": 0.01028763,
      "balance_loss_clip": 1.04949236,
      "balance_loss_mlp": 1.02073383,
      "epoch": 0.4095472855167438,
      "flos": 21618080616960.0,
      "grad_norm": 2.166322051332972,
      "language_loss": 0.695768,
      "learning_rate": 2.6699506977217128e-06,
      "loss": 0.71760648,
      "num_input_tokens_seen": 73588490,
      "step": 3406,
      "time_per_iteration": 2.5986838340759277
    },
    {
      "auxiliary_loss_clip": 0.01168335,
      "auxiliary_loss_mlp": 0.01028656,
      "balance_loss_clip": 1.05620766,
      "balance_loss_mlp": 1.02059197,
      "epoch": 0.4096675284073829,
      "flos": 27926112499200.0,
      "grad_norm": 2.1728493726966875,
      "language_loss": 0.70028335,
      "learning_rate": 2.6692166788907233e-06,
      "loss": 0.72225332,
      "num_input_tokens_seen": 73608685,
      "step": 3407,
      "time_per_iteration": 2.5935885906219482
    },
    {
      "auxiliary_loss_clip": 0.0115437,
      "auxiliary_loss_mlp": 0.01028896,
      "balance_loss_clip": 1.05020642,
      "balance_loss_mlp": 1.02062893,
      "epoch": 0.409787771298022,
      "flos": 19206607092480.0,
      "grad_norm": 2.292325951293512,
      "language_loss": 0.76727957,
      "learning_rate": 2.6684825585400957e-06,
      "loss": 0.78911221,
      "num_input_tokens_seen": 73627630,
      "step": 3408,
      "time_per_iteration": 2.578110694885254
    },
    {
      "auxiliary_loss_clip": 0.01049469,
      "auxiliary_loss_mlp": 0.01001344,
      "balance_loss_clip": 1.01717615,
      "balance_loss_mlp": 1.00000846,
      "epoch": 0.4099080141886611,
      "flos": 59269234832640.0,
      "grad_norm": 0.813098794828117,
      "language_loss": 0.6513896,
      "learning_rate": 2.6677483367811947e-06,
      "loss": 0.67189777,
      "num_input_tokens_seen": 73687670,
      "step": 3409,
      "time_per_iteration": 3.2548367977142334
    },
    {
      "auxiliary_loss_clip": 0.0116922,
      "auxiliary_loss_mlp": 0.01025279,
      "balance_loss_clip": 1.05035162,
      "balance_loss_mlp": 1.01785874,
      "epoch": 0.4100282570793002,
      "flos": 21906443001600.0,
      "grad_norm": 1.6520280283916442,
      "language_loss": 0.75545847,
      "learning_rate": 2.6670140137254028e-06,
      "loss": 0.77740347,
      "num_input_tokens_seen": 73707145,
      "step": 3410,
      "time_per_iteration": 2.548207998275757
    },
    {
      "auxiliary_loss_clip": 0.0112179,
      "auxiliary_loss_mlp": 0.01032324,
      "balance_loss_clip": 1.04684377,
      "balance_loss_mlp": 1.02424157,
      "epoch": 0.4101484999699393,
      "flos": 18551596631040.0,
      "grad_norm": 2.2265577402915806,
      "language_loss": 0.89837545,
      "learning_rate": 2.666279589484115e-06,
      "loss": 0.91991663,
      "num_input_tokens_seen": 73725045,
      "step": 3411,
      "time_per_iteration": 2.6030380725860596
    },
    {
      "auxiliary_loss_clip": 0.01125636,
      "auxiliary_loss_mlp": 0.01030207,
      "balance_loss_clip": 1.04774678,
      "balance_loss_mlp": 1.02251852,
      "epoch": 0.41026874286057835,
      "flos": 19094529680640.0,
      "grad_norm": 1.8960066646899574,
      "language_loss": 0.8108139,
      "learning_rate": 2.6655450641687435e-06,
      "loss": 0.83237237,
      "num_input_tokens_seen": 73742610,
      "step": 3412,
      "time_per_iteration": 2.6661558151245117
    },
    {
      "auxiliary_loss_clip": 0.01180761,
      "auxiliary_loss_mlp": 0.01032541,
      "balance_loss_clip": 1.0547055,
      "balance_loss_mlp": 1.02477777,
      "epoch": 0.41038898575121746,
      "flos": 31209568588800.0,
      "grad_norm": 1.7179658132574744,
      "language_loss": 0.69332796,
      "learning_rate": 2.664810437890715e-06,
      "loss": 0.71546102,
      "num_input_tokens_seen": 73764280,
      "step": 3413,
      "time_per_iteration": 2.5865578651428223
    },
    {
      "auxiliary_loss_clip": 0.01105703,
      "auxiliary_loss_mlp": 0.01027005,
      "balance_loss_clip": 1.05055022,
      "balance_loss_mlp": 1.01895809,
      "epoch": 0.41050922864185657,
      "flos": 14355865895040.0,
      "grad_norm": 2.8751312955359434,
      "language_loss": 0.79619133,
      "learning_rate": 2.6640757107614714e-06,
      "loss": 0.81751835,
      "num_input_tokens_seen": 73782375,
      "step": 3414,
      "time_per_iteration": 3.4741129875183105
    },
    {
      "auxiliary_loss_clip": 0.01135976,
      "auxiliary_loss_mlp": 0.01027421,
      "balance_loss_clip": 1.05100405,
      "balance_loss_mlp": 1.01917768,
      "epoch": 0.4106294715324956,
      "flos": 30956290813440.0,
      "grad_norm": 2.427575559369949,
      "language_loss": 0.69315577,
      "learning_rate": 2.6633408828924697e-06,
      "loss": 0.71478975,
      "num_input_tokens_seen": 73801240,
      "step": 3415,
      "time_per_iteration": 2.6957147121429443
    },
    {
      "auxiliary_loss_clip": 0.01146431,
      "auxiliary_loss_mlp": 0.01029008,
      "balance_loss_clip": 1.05039692,
      "balance_loss_mlp": 1.02143216,
      "epoch": 0.41074971442313474,
      "flos": 24457321209600.0,
      "grad_norm": 1.6456039108239797,
      "language_loss": 0.69944888,
      "learning_rate": 2.662605954395185e-06,
      "loss": 0.72120333,
      "num_input_tokens_seen": 73821200,
      "step": 3416,
      "time_per_iteration": 2.6323070526123047
    },
    {
      "auxiliary_loss_clip": 0.01170873,
      "auxiliary_loss_mlp": 0.01027331,
      "balance_loss_clip": 1.05234885,
      "balance_loss_mlp": 1.01956999,
      "epoch": 0.41086995731377385,
      "flos": 21542991235200.0,
      "grad_norm": 1.7408522727418236,
      "language_loss": 0.83634198,
      "learning_rate": 2.6618709253811027e-06,
      "loss": 0.85832399,
      "num_input_tokens_seen": 73840655,
      "step": 3417,
      "time_per_iteration": 2.6192917823791504
    },
    {
      "auxiliary_loss_clip": 0.01178783,
      "auxiliary_loss_mlp": 0.01031507,
      "balance_loss_clip": 1.05532014,
      "balance_loss_mlp": 1.02449441,
      "epoch": 0.4109902002044129,
      "flos": 20702753314560.0,
      "grad_norm": 1.6071080935468598,
      "language_loss": 0.8743031,
      "learning_rate": 2.6611357959617277e-06,
      "loss": 0.89640599,
      "num_input_tokens_seen": 73860275,
      "step": 3418,
      "time_per_iteration": 2.498661756515503
    },
    {
      "auxiliary_loss_clip": 0.01135622,
      "auxiliary_loss_mlp": 0.0103289,
      "balance_loss_clip": 1.04992771,
      "balance_loss_mlp": 1.02506387,
      "epoch": 0.411110443095052,
      "flos": 18179992477440.0,
      "grad_norm": 1.8122977002664584,
      "language_loss": 0.90906507,
      "learning_rate": 2.660400566248578e-06,
      "loss": 0.93075019,
      "num_input_tokens_seen": 73878400,
      "step": 3419,
      "time_per_iteration": 2.58943510055542
    },
    {
      "auxiliary_loss_clip": 0.01141517,
      "auxiliary_loss_mlp": 0.01032264,
      "balance_loss_clip": 1.05065358,
      "balance_loss_mlp": 1.02350235,
      "epoch": 0.41123068598569107,
      "flos": 14575244209920.0,
      "grad_norm": 2.368406350496862,
      "language_loss": 0.66832113,
      "learning_rate": 2.6596652363531876e-06,
      "loss": 0.69005889,
      "num_input_tokens_seen": 73894275,
      "step": 3420,
      "time_per_iteration": 4.046716213226318
    },
    {
      "auxiliary_loss_clip": 0.01182424,
      "auxiliary_loss_mlp": 0.01033781,
      "balance_loss_clip": 1.05522251,
      "balance_loss_mlp": 1.02579403,
      "epoch": 0.4113509288763302,
      "flos": 21177995184000.0,
      "grad_norm": 1.480694397912182,
      "language_loss": 0.78417319,
      "learning_rate": 2.6589298063871055e-06,
      "loss": 0.80633533,
      "num_input_tokens_seen": 73914450,
      "step": 3421,
      "time_per_iteration": 3.371730327606201
    },
    {
      "auxiliary_loss_clip": 0.0118281,
      "auxiliary_loss_mlp": 0.01032287,
      "balance_loss_clip": 1.05589414,
      "balance_loss_mlp": 1.0248487,
      "epoch": 0.4114711717669693,
      "flos": 18442212739200.0,
      "grad_norm": 1.8936669088921665,
      "language_loss": 0.69675469,
      "learning_rate": 2.658194276461895e-06,
      "loss": 0.71890569,
      "num_input_tokens_seen": 73932375,
      "step": 3422,
      "time_per_iteration": 2.47825026512146
    },
    {
      "auxiliary_loss_clip": 0.01152002,
      "auxiliary_loss_mlp": 0.01037104,
      "balance_loss_clip": 1.04730034,
      "balance_loss_mlp": 1.02859879,
      "epoch": 0.41159141465760835,
      "flos": 27233395735680.0,
      "grad_norm": 2.1392674876623525,
      "language_loss": 0.66915703,
      "learning_rate": 2.6574586466891368e-06,
      "loss": 0.69104809,
      "num_input_tokens_seen": 73952850,
      "step": 3423,
      "time_per_iteration": 2.6013925075531006
    },
    {
      "auxiliary_loss_clip": 0.01155362,
      "auxiliary_loss_mlp": 0.00761125,
      "balance_loss_clip": 1.05198097,
      "balance_loss_mlp": 1.00028467,
      "epoch": 0.41171165754824746,
      "flos": 20006876154240.0,
      "grad_norm": 2.4563766766560162,
      "language_loss": 0.64842403,
      "learning_rate": 2.6567229171804247e-06,
      "loss": 0.66758895,
      "num_input_tokens_seen": 73970735,
      "step": 3424,
      "time_per_iteration": 2.636162757873535
    },
    {
      "auxiliary_loss_clip": 0.01146535,
      "auxiliary_loss_mlp": 0.01026377,
      "balance_loss_clip": 1.04787183,
      "balance_loss_mlp": 1.01833034,
      "epoch": 0.41183190043888657,
      "flos": 18004318035840.0,
      "grad_norm": 2.7046510782254227,
      "language_loss": 0.87705612,
      "learning_rate": 2.655987088047368e-06,
      "loss": 0.89878523,
      "num_input_tokens_seen": 73989080,
      "step": 3425,
      "time_per_iteration": 2.5636308193206787
    },
    {
      "auxiliary_loss_clip": 0.01151322,
      "auxiliary_loss_mlp": 0.01036911,
      "balance_loss_clip": 1.0504638,
      "balance_loss_mlp": 1.02848268,
      "epoch": 0.4119521433295256,
      "flos": 27163370171520.0,
      "grad_norm": 1.8845944029501807,
      "language_loss": 0.78399289,
      "learning_rate": 2.6552511594015912e-06,
      "loss": 0.80587524,
      "num_input_tokens_seen": 74009470,
      "step": 3426,
      "time_per_iteration": 2.6142683029174805
    },
    {
      "auxiliary_loss_clip": 0.01151017,
      "auxiliary_loss_mlp": 0.01032576,
      "balance_loss_clip": 1.04751325,
      "balance_loss_mlp": 1.02407634,
      "epoch": 0.41207238622016473,
      "flos": 15122020014720.0,
      "grad_norm": 1.8713548410316012,
      "language_loss": 0.85217512,
      "learning_rate": 2.654515131354735e-06,
      "loss": 0.87401104,
      "num_input_tokens_seen": 74027735,
      "step": 3427,
      "time_per_iteration": 2.568700075149536
    },
    {
      "auxiliary_loss_clip": 0.01141485,
      "auxiliary_loss_mlp": 0.01031974,
      "balance_loss_clip": 1.05004728,
      "balance_loss_mlp": 1.02427292,
      "epoch": 0.41219262911080384,
      "flos": 27052872958080.0,
      "grad_norm": 1.9298165493875126,
      "language_loss": 0.84699249,
      "learning_rate": 2.653779004018453e-06,
      "loss": 0.86872709,
      "num_input_tokens_seen": 74048300,
      "step": 3428,
      "time_per_iteration": 2.653779983520508
    },
    {
      "auxiliary_loss_clip": 0.01144699,
      "auxiliary_loss_mlp": 0.01026591,
      "balance_loss_clip": 1.04883552,
      "balance_loss_mlp": 1.01894331,
      "epoch": 0.4123128720014429,
      "flos": 24686360282880.0,
      "grad_norm": 1.8256237136121538,
      "language_loss": 0.82256877,
      "learning_rate": 2.653042777504417e-06,
      "loss": 0.84428167,
      "num_input_tokens_seen": 74070890,
      "step": 3429,
      "time_per_iteration": 2.623439311981201
    },
    {
      "auxiliary_loss_clip": 0.01158631,
      "auxiliary_loss_mlp": 0.01027105,
      "balance_loss_clip": 1.05021822,
      "balance_loss_mlp": 1.0186944,
      "epoch": 0.412433114892082,
      "flos": 26244774731520.0,
      "grad_norm": 1.8267891625611843,
      "language_loss": 0.80079323,
      "learning_rate": 2.6523064519243105e-06,
      "loss": 0.82265055,
      "num_input_tokens_seen": 74090460,
      "step": 3430,
      "time_per_iteration": 2.6495444774627686
    },
    {
      "auxiliary_loss_clip": 0.0117005,
      "auxiliary_loss_mlp": 0.01034109,
      "balance_loss_clip": 1.05569482,
      "balance_loss_mlp": 1.02588964,
      "epoch": 0.4125533577827211,
      "flos": 21361031913600.0,
      "grad_norm": 2.261948954856196,
      "language_loss": 0.78979701,
      "learning_rate": 2.6515700273898333e-06,
      "loss": 0.81183863,
      "num_input_tokens_seen": 74108335,
      "step": 3431,
      "time_per_iteration": 2.5409739017486572
    },
    {
      "auxiliary_loss_clip": 0.01142244,
      "auxiliary_loss_mlp": 0.01029514,
      "balance_loss_clip": 1.05063224,
      "balance_loss_mlp": 1.02145541,
      "epoch": 0.4126736006733602,
      "flos": 26067556005120.0,
      "grad_norm": 2.0623814743776467,
      "language_loss": 0.69585121,
      "learning_rate": 2.6508335040127018e-06,
      "loss": 0.71756876,
      "num_input_tokens_seen": 74128030,
      "step": 3432,
      "time_per_iteration": 2.581613063812256
    },
    {
      "auxiliary_loss_clip": 0.01171463,
      "auxiliary_loss_mlp": 0.01032832,
      "balance_loss_clip": 1.05363977,
      "balance_loss_mlp": 1.02460635,
      "epoch": 0.4127938435639993,
      "flos": 25666146541440.0,
      "grad_norm": 1.4727386581581858,
      "language_loss": 0.77011347,
      "learning_rate": 2.6500968819046446e-06,
      "loss": 0.7921564,
      "num_input_tokens_seen": 74148330,
      "step": 3433,
      "time_per_iteration": 2.569923162460327
    },
    {
      "auxiliary_loss_clip": 0.01128548,
      "auxiliary_loss_mlp": 0.01028175,
      "balance_loss_clip": 1.0458951,
      "balance_loss_mlp": 1.02039683,
      "epoch": 0.4129140864546384,
      "flos": 17995914253440.0,
      "grad_norm": 2.504284595247858,
      "language_loss": 0.59606034,
      "learning_rate": 2.649360161177408e-06,
      "loss": 0.6176275,
      "num_input_tokens_seen": 74163390,
      "step": 3434,
      "time_per_iteration": 2.561326265335083
    },
    {
      "auxiliary_loss_clip": 0.01173763,
      "auxiliary_loss_mlp": 0.01032917,
      "balance_loss_clip": 1.05361009,
      "balance_loss_mlp": 1.02422094,
      "epoch": 0.41303432934527745,
      "flos": 23732895715200.0,
      "grad_norm": 1.8528976360958294,
      "language_loss": 0.73316234,
      "learning_rate": 2.6486233419427504e-06,
      "loss": 0.75522923,
      "num_input_tokens_seen": 74183205,
      "step": 3435,
      "time_per_iteration": 2.553699493408203
    },
    {
      "auxiliary_loss_clip": 0.01133936,
      "auxiliary_loss_mlp": 0.01035101,
      "balance_loss_clip": 1.05167341,
      "balance_loss_mlp": 1.02637482,
      "epoch": 0.41315457223591656,
      "flos": 19755286318080.0,
      "grad_norm": 3.144277523459835,
      "language_loss": 0.75374877,
      "learning_rate": 2.6478864243124484e-06,
      "loss": 0.77543914,
      "num_input_tokens_seen": 74202870,
      "step": 3436,
      "time_per_iteration": 2.6695733070373535
    },
    {
      "auxiliary_loss_clip": 0.01165683,
      "auxiliary_loss_mlp": 0.01027814,
      "balance_loss_clip": 1.04940426,
      "balance_loss_mlp": 1.01985729,
      "epoch": 0.4132748151265556,
      "flos": 20923316778240.0,
      "grad_norm": 1.8250696474316512,
      "language_loss": 0.85167813,
      "learning_rate": 2.6471494083982903e-06,
      "loss": 0.87361312,
      "num_input_tokens_seen": 74222255,
      "step": 3437,
      "time_per_iteration": 2.545546770095825
    },
    {
      "auxiliary_loss_clip": 0.01141532,
      "auxiliary_loss_mlp": 0.0103275,
      "balance_loss_clip": 1.04746354,
      "balance_loss_mlp": 1.02453613,
      "epoch": 0.4133950580171947,
      "flos": 32232520016640.0,
      "grad_norm": 1.6652216992311457,
      "language_loss": 0.74851817,
      "learning_rate": 2.6464122943120818e-06,
      "loss": 0.77026105,
      "num_input_tokens_seen": 74242480,
      "step": 3438,
      "time_per_iteration": 2.678117275238037
    },
    {
      "auxiliary_loss_clip": 0.01139521,
      "auxiliary_loss_mlp": 0.01031753,
      "balance_loss_clip": 1.05001605,
      "balance_loss_mlp": 1.02358139,
      "epoch": 0.41351530090783384,
      "flos": 23292487059840.0,
      "grad_norm": 2.9304441451396706,
      "language_loss": 0.82671869,
      "learning_rate": 2.645675082165642e-06,
      "loss": 0.84843147,
      "num_input_tokens_seen": 74258690,
      "step": 3439,
      "time_per_iteration": 2.606657028198242
    },
    {
      "auxiliary_loss_clip": 0.01156303,
      "auxiliary_loss_mlp": 0.01033522,
      "balance_loss_clip": 1.05303693,
      "balance_loss_mlp": 1.02465868,
      "epoch": 0.4136355437984729,
      "flos": 25593571111680.0,
      "grad_norm": 2.2101939143279252,
      "language_loss": 0.76053929,
      "learning_rate": 2.644937772070806e-06,
      "loss": 0.7824375,
      "num_input_tokens_seen": 74277135,
      "step": 3440,
      "time_per_iteration": 2.6155617237091064
    },
    {
      "auxiliary_loss_clip": 0.01185192,
      "auxiliary_loss_mlp": 0.01027851,
      "balance_loss_clip": 1.05727005,
      "balance_loss_mlp": 1.01993537,
      "epoch": 0.413755786689112,
      "flos": 19828615933440.0,
      "grad_norm": 2.4516476655570907,
      "language_loss": 0.83773589,
      "learning_rate": 2.6442003641394225e-06,
      "loss": 0.85986638,
      "num_input_tokens_seen": 74294730,
      "step": 3441,
      "time_per_iteration": 3.339801788330078
    },
    {
      "auxiliary_loss_clip": 0.01151633,
      "auxiliary_loss_mlp": 0.01028226,
      "balance_loss_clip": 1.04970551,
      "balance_loss_mlp": 1.01990783,
      "epoch": 0.4138760295797511,
      "flos": 26870446759680.0,
      "grad_norm": 1.4325786358270425,
      "language_loss": 0.83850074,
      "learning_rate": 2.643462858483356e-06,
      "loss": 0.86029941,
      "num_input_tokens_seen": 74315015,
      "step": 3442,
      "time_per_iteration": 2.6005334854125977
    },
    {
      "auxiliary_loss_clip": 0.01118911,
      "auxiliary_loss_mlp": 0.0103159,
      "balance_loss_clip": 1.04484963,
      "balance_loss_mlp": 1.02343619,
      "epoch": 0.41399627247039017,
      "flos": 16399254798720.0,
      "grad_norm": 1.9826697309395884,
      "language_loss": 0.72459543,
      "learning_rate": 2.6427252552144856e-06,
      "loss": 0.74610043,
      "num_input_tokens_seen": 74333665,
      "step": 3443,
      "time_per_iteration": 2.6547365188598633
    },
    {
      "auxiliary_loss_clip": 0.01181277,
      "auxiliary_loss_mlp": 0.01031916,
      "balance_loss_clip": 1.05440044,
      "balance_loss_mlp": 1.02377415,
      "epoch": 0.4141165153610293,
      "flos": 22930220442240.0,
      "grad_norm": 1.7617620966457528,
      "language_loss": 0.74850881,
      "learning_rate": 2.6419875544447044e-06,
      "loss": 0.77064079,
      "num_input_tokens_seen": 74355065,
      "step": 3444,
      "time_per_iteration": 2.526935577392578
    },
    {
      "auxiliary_loss_clip": 0.01182776,
      "auxiliary_loss_mlp": 0.01030256,
      "balance_loss_clip": 1.05367172,
      "balance_loss_mlp": 1.02220941,
      "epoch": 0.4142367582516684,
      "flos": 25192556697600.0,
      "grad_norm": 1.6132265488720483,
      "language_loss": 0.71497381,
      "learning_rate": 2.6412497562859218e-06,
      "loss": 0.73710418,
      "num_input_tokens_seen": 74376345,
      "step": 3445,
      "time_per_iteration": 2.556248903274536
    },
    {
      "auxiliary_loss_clip": 0.01171414,
      "auxiliary_loss_mlp": 0.0102513,
      "balance_loss_clip": 1.05193639,
      "balance_loss_mlp": 1.01655889,
      "epoch": 0.41435700114230745,
      "flos": 21690476478720.0,
      "grad_norm": 2.3736010997847616,
      "language_loss": 0.76423168,
      "learning_rate": 2.6405118608500617e-06,
      "loss": 0.78619707,
      "num_input_tokens_seen": 74395170,
      "step": 3446,
      "time_per_iteration": 4.064232110977173
    },
    {
      "auxiliary_loss_clip": 0.01138995,
      "auxiliary_loss_mlp": 0.01028879,
      "balance_loss_clip": 1.05521131,
      "balance_loss_mlp": 1.02144599,
      "epoch": 0.41447724403294656,
      "flos": 25995160143360.0,
      "grad_norm": 1.6611064703982372,
      "language_loss": 0.8133899,
      "learning_rate": 2.6397738682490613e-06,
      "loss": 0.83506858,
      "num_input_tokens_seen": 74416070,
      "step": 3447,
      "time_per_iteration": 3.335458755493164
    },
    {
      "auxiliary_loss_clip": 0.01184088,
      "auxiliary_loss_mlp": 0.01030723,
      "balance_loss_clip": 1.05507481,
      "balance_loss_mlp": 1.02244091,
      "epoch": 0.41459748692358567,
      "flos": 18259678800000.0,
      "grad_norm": 1.6480215596762928,
      "language_loss": 0.75400931,
      "learning_rate": 2.6390357785948734e-06,
      "loss": 0.77615744,
      "num_input_tokens_seen": 74433185,
      "step": 3448,
      "time_per_iteration": 2.4961304664611816
    },
    {
      "auxiliary_loss_clip": 0.01168321,
      "auxiliary_loss_mlp": 0.0103384,
      "balance_loss_clip": 1.0540216,
      "balance_loss_mlp": 1.02569795,
      "epoch": 0.4147177298142247,
      "flos": 24168456034560.0,
      "grad_norm": 1.7699780354864054,
      "language_loss": 0.80116099,
      "learning_rate": 2.6382975919994667e-06,
      "loss": 0.82318264,
      "num_input_tokens_seen": 74453760,
      "step": 3449,
      "time_per_iteration": 2.575606107711792
    },
    {
      "auxiliary_loss_clip": 0.01154662,
      "auxiliary_loss_mlp": 0.01026599,
      "balance_loss_clip": 1.05027628,
      "balance_loss_mlp": 1.01942277,
      "epoch": 0.41483797270486383,
      "flos": 20084659056000.0,
      "grad_norm": 1.70684902275972,
      "language_loss": 0.72987962,
      "learning_rate": 2.637559308574822e-06,
      "loss": 0.7516923,
      "num_input_tokens_seen": 74473505,
      "step": 3450,
      "time_per_iteration": 2.610459804534912
    },
    {
      "auxiliary_loss_clip": 0.01181729,
      "auxiliary_loss_mlp": 0.01030643,
      "balance_loss_clip": 1.05361068,
      "balance_loss_mlp": 1.02242935,
      "epoch": 0.4149582155955029,
      "flos": 30081040110720.0,
      "grad_norm": 1.9380554810582133,
      "language_loss": 0.71125299,
      "learning_rate": 2.6368209284329376e-06,
      "loss": 0.73337674,
      "num_input_tokens_seen": 74494135,
      "step": 3451,
      "time_per_iteration": 2.5619311332702637
    },
    {
      "auxiliary_loss_clip": 0.01166845,
      "auxiliary_loss_mlp": 0.01030221,
      "balance_loss_clip": 1.05098128,
      "balance_loss_mlp": 1.02197766,
      "epoch": 0.415078458486142,
      "flos": 16764394504320.0,
      "grad_norm": 2.1583686447481236,
      "language_loss": 0.75415826,
      "learning_rate": 2.636082451685825e-06,
      "loss": 0.77612889,
      "num_input_tokens_seen": 74512335,
      "step": 3452,
      "time_per_iteration": 2.566957712173462
    },
    {
      "auxiliary_loss_clip": 0.01160149,
      "auxiliary_loss_mlp": 0.01027672,
      "balance_loss_clip": 1.05565333,
      "balance_loss_mlp": 1.01941681,
      "epoch": 0.4151987013767811,
      "flos": 26033692458240.0,
      "grad_norm": 1.7363858552840825,
      "language_loss": 0.86402082,
      "learning_rate": 2.6353438784455094e-06,
      "loss": 0.88589901,
      "num_input_tokens_seen": 74535620,
      "step": 3453,
      "time_per_iteration": 2.6165599822998047
    },
    {
      "auxiliary_loss_clip": 0.01154627,
      "auxiliary_loss_mlp": 0.01037755,
      "balance_loss_clip": 1.05469012,
      "balance_loss_mlp": 1.02868366,
      "epoch": 0.41531894426742016,
      "flos": 24608002763520.0,
      "grad_norm": 2.2414319328868397,
      "language_loss": 0.71548033,
      "learning_rate": 2.6346052088240326e-06,
      "loss": 0.73740411,
      "num_input_tokens_seen": 74555140,
      "step": 3454,
      "time_per_iteration": 2.6240410804748535
    },
    {
      "auxiliary_loss_clip": 0.01159684,
      "auxiliary_loss_mlp": 0.01030248,
      "balance_loss_clip": 1.05425763,
      "balance_loss_mlp": 1.02207673,
      "epoch": 0.4154391871580593,
      "flos": 14975791747200.0,
      "grad_norm": 1.9504896745296996,
      "language_loss": 0.76847076,
      "learning_rate": 2.63386644293345e-06,
      "loss": 0.79037011,
      "num_input_tokens_seen": 74571485,
      "step": 3455,
      "time_per_iteration": 2.537118911743164
    },
    {
      "auxiliary_loss_clip": 0.01135179,
      "auxiliary_loss_mlp": 0.01025847,
      "balance_loss_clip": 1.04720569,
      "balance_loss_mlp": 1.01832438,
      "epoch": 0.4155594300486984,
      "flos": 14647173194880.0,
      "grad_norm": 2.900292987531332,
      "language_loss": 0.83382159,
      "learning_rate": 2.633127580885833e-06,
      "loss": 0.8554318,
      "num_input_tokens_seen": 74585985,
      "step": 3456,
      "time_per_iteration": 2.5883262157440186
    },
    {
      "auxiliary_loss_clip": 0.01185709,
      "auxiliary_loss_mlp": 0.01030305,
      "balance_loss_clip": 1.05877113,
      "balance_loss_mlp": 1.02221644,
      "epoch": 0.41567967293933744,
      "flos": 29497276275840.0,
      "grad_norm": 2.2703279373711283,
      "language_loss": 0.64985311,
      "learning_rate": 2.632388622793265e-06,
      "loss": 0.67201316,
      "num_input_tokens_seen": 74605140,
      "step": 3457,
      "time_per_iteration": 2.5698139667510986
    },
    {
      "auxiliary_loss_clip": 0.01170143,
      "auxiliary_loss_mlp": 0.01034424,
      "balance_loss_clip": 1.0548532,
      "balance_loss_mlp": 1.0264132,
      "epoch": 0.41579991582997655,
      "flos": 19238387650560.0,
      "grad_norm": 1.654294166425432,
      "language_loss": 0.67915273,
      "learning_rate": 2.6316495687678457e-06,
      "loss": 0.7011984,
      "num_input_tokens_seen": 74623790,
      "step": 3458,
      "time_per_iteration": 2.540938138961792
    },
    {
      "auxiliary_loss_clip": 0.01121298,
      "auxiliary_loss_mlp": 0.0102994,
      "balance_loss_clip": 1.04912543,
      "balance_loss_mlp": 1.02199483,
      "epoch": 0.41592015872061566,
      "flos": 24462061804800.0,
      "grad_norm": 2.623476452610566,
      "language_loss": 0.76948071,
      "learning_rate": 2.6309104189216887e-06,
      "loss": 0.79099309,
      "num_input_tokens_seen": 74641355,
      "step": 3459,
      "time_per_iteration": 2.634948253631592
    },
    {
      "auxiliary_loss_clip": 0.01128404,
      "auxiliary_loss_mlp": 0.00762211,
      "balance_loss_clip": 1.04791546,
      "balance_loss_mlp": 1.00025272,
      "epoch": 0.4160404016112547,
      "flos": 20775651966720.0,
      "grad_norm": 2.0711615878356304,
      "language_loss": 0.74809933,
      "learning_rate": 2.630171173366923e-06,
      "loss": 0.7670055,
      "num_input_tokens_seen": 74657155,
      "step": 3460,
      "time_per_iteration": 2.617377519607544
    },
    {
      "auxiliary_loss_clip": 0.01127003,
      "auxiliary_loss_mlp": 0.01033806,
      "balance_loss_clip": 1.04962957,
      "balance_loss_mlp": 1.02555966,
      "epoch": 0.41616064450189383,
      "flos": 13916462820480.0,
      "grad_norm": 2.4789310500593724,
      "language_loss": 0.74332392,
      "learning_rate": 2.629431832215691e-06,
      "loss": 0.76493204,
      "num_input_tokens_seen": 74671960,
      "step": 3461,
      "time_per_iteration": 2.6102709770202637
    },
    {
      "auxiliary_loss_clip": 0.01153952,
      "auxiliary_loss_mlp": 0.01032804,
      "balance_loss_clip": 1.05398345,
      "balance_loss_mlp": 1.02434635,
      "epoch": 0.41628088739253294,
      "flos": 20010826650240.0,
      "grad_norm": 1.6431718563824094,
      "language_loss": 0.87276089,
      "learning_rate": 2.628692395580151e-06,
      "loss": 0.89462852,
      "num_input_tokens_seen": 74692050,
      "step": 3462,
      "time_per_iteration": 2.604914665222168
    },
    {
      "auxiliary_loss_clip": 0.01095035,
      "auxiliary_loss_mlp": 0.01033576,
      "balance_loss_clip": 1.04540634,
      "balance_loss_mlp": 1.02517748,
      "epoch": 0.416401130283172,
      "flos": 29168801377920.0,
      "grad_norm": 1.591792127577921,
      "language_loss": 0.79156715,
      "learning_rate": 2.6279528635724747e-06,
      "loss": 0.81285328,
      "num_input_tokens_seen": 74712205,
      "step": 3463,
      "time_per_iteration": 2.7355947494506836
    },
    {
      "auxiliary_loss_clip": 0.01169871,
      "auxiliary_loss_mlp": 0.01032666,
      "balance_loss_clip": 1.05279434,
      "balance_loss_mlp": 1.02390468,
      "epoch": 0.4165213731738111,
      "flos": 16246813478400.0,
      "grad_norm": 2.656959849512158,
      "language_loss": 0.78371286,
      "learning_rate": 2.627213236304848e-06,
      "loss": 0.80573821,
      "num_input_tokens_seen": 74729005,
      "step": 3464,
      "time_per_iteration": 2.51753568649292
    },
    {
      "auxiliary_loss_clip": 0.01175876,
      "auxiliary_loss_mlp": 0.01027782,
      "balance_loss_clip": 1.0562427,
      "balance_loss_mlp": 1.02003884,
      "epoch": 0.4166416160644502,
      "flos": 33765438787200.0,
      "grad_norm": 1.873316136810797,
      "language_loss": 0.70301902,
      "learning_rate": 2.626473513889472e-06,
      "loss": 0.72505558,
      "num_input_tokens_seen": 74751385,
      "step": 3465,
      "time_per_iteration": 2.6766517162323
    },
    {
      "auxiliary_loss_clip": 0.01162778,
      "auxiliary_loss_mlp": 0.01031229,
      "balance_loss_clip": 1.05353117,
      "balance_loss_mlp": 1.02315903,
      "epoch": 0.41676185895508927,
      "flos": 20917498775040.0,
      "grad_norm": 1.7865354090889305,
      "language_loss": 0.83087444,
      "learning_rate": 2.625733696438562e-06,
      "loss": 0.85281456,
      "num_input_tokens_seen": 74768890,
      "step": 3466,
      "time_per_iteration": 2.5560686588287354
    },
    {
      "auxiliary_loss_clip": 0.01155361,
      "auxiliary_loss_mlp": 0.01035075,
      "balance_loss_clip": 1.05329633,
      "balance_loss_mlp": 1.02683794,
      "epoch": 0.4168821018457284,
      "flos": 18406122549120.0,
      "grad_norm": 1.7444870414047702,
      "language_loss": 0.75122786,
      "learning_rate": 2.6249937840643476e-06,
      "loss": 0.77313226,
      "num_input_tokens_seen": 74787195,
      "step": 3467,
      "time_per_iteration": 3.3607144355773926
    },
    {
      "auxiliary_loss_clip": 0.01186527,
      "auxiliary_loss_mlp": 0.00761736,
      "balance_loss_clip": 1.05982018,
      "balance_loss_mlp": 1.00022542,
      "epoch": 0.41700234473636744,
      "flos": 18698399516160.0,
      "grad_norm": 1.6130800200153865,
      "language_loss": 0.66452098,
      "learning_rate": 2.6242537768790733e-06,
      "loss": 0.68400359,
      "num_input_tokens_seen": 74806350,
      "step": 3468,
      "time_per_iteration": 2.513181447982788
    },
    {
      "auxiliary_loss_clip": 0.01172843,
      "auxiliary_loss_mlp": 0.01026193,
      "balance_loss_clip": 1.05788684,
      "balance_loss_mlp": 1.0179143,
      "epoch": 0.41712258762700655,
      "flos": 31033283616000.0,
      "grad_norm": 4.733326937474909,
      "language_loss": 0.68618578,
      "learning_rate": 2.6235136749949975e-06,
      "loss": 0.70817614,
      "num_input_tokens_seen": 74829800,
      "step": 3469,
      "time_per_iteration": 2.650731325149536
    },
    {
      "auxiliary_loss_clip": 0.01183728,
      "auxiliary_loss_mlp": 0.01030156,
      "balance_loss_clip": 1.05621696,
      "balance_loss_mlp": 1.02180791,
      "epoch": 0.41724283051764566,
      "flos": 35914763877120.0,
      "grad_norm": 2.0540906831097465,
      "language_loss": 0.61274153,
      "learning_rate": 2.6227734785243924e-06,
      "loss": 0.63488036,
      "num_input_tokens_seen": 74849760,
      "step": 3470,
      "time_per_iteration": 2.624473810195923
    },
    {
      "auxiliary_loss_clip": 0.01105833,
      "auxiliary_loss_mlp": 0.01031594,
      "balance_loss_clip": 1.04472494,
      "balance_loss_mlp": 1.02329087,
      "epoch": 0.4173630734082847,
      "flos": 25333649320320.0,
      "grad_norm": 1.9665957482200438,
      "language_loss": 0.78995383,
      "learning_rate": 2.6220331875795466e-06,
      "loss": 0.81132805,
      "num_input_tokens_seen": 74869110,
      "step": 3471,
      "time_per_iteration": 2.701052188873291
    },
    {
      "auxiliary_loss_clip": 0.01167029,
      "auxiliary_loss_mlp": 0.01032461,
      "balance_loss_clip": 1.05584526,
      "balance_loss_mlp": 1.02390742,
      "epoch": 0.4174833162989238,
      "flos": 26685398868480.0,
      "grad_norm": 1.594105456538369,
      "language_loss": 0.75399929,
      "learning_rate": 2.62129280227276e-06,
      "loss": 0.77599418,
      "num_input_tokens_seen": 74889110,
      "step": 3472,
      "time_per_iteration": 4.912471771240234
    },
    {
      "auxiliary_loss_clip": 0.01178099,
      "auxiliary_loss_mlp": 0.01035498,
      "balance_loss_clip": 1.05865717,
      "balance_loss_mlp": 1.02732658,
      "epoch": 0.41760355918956293,
      "flos": 74739584010240.0,
      "grad_norm": 2.2130894016469345,
      "language_loss": 0.68342423,
      "learning_rate": 2.62055232271635e-06,
      "loss": 0.70556021,
      "num_input_tokens_seen": 74916260,
      "step": 3473,
      "time_per_iteration": 2.9343502521514893
    },
    {
      "auxiliary_loss_clip": 0.01132628,
      "auxiliary_loss_mlp": 0.01031925,
      "balance_loss_clip": 1.04887128,
      "balance_loss_mlp": 1.02358627,
      "epoch": 0.417723802080202,
      "flos": 14317513148160.0,
      "grad_norm": 2.182897822566128,
      "language_loss": 0.88092399,
      "learning_rate": 2.619811749022646e-06,
      "loss": 0.90256941,
      "num_input_tokens_seen": 74931570,
      "step": 3474,
      "time_per_iteration": 2.5923590660095215
    },
    {
      "auxiliary_loss_clip": 0.01175466,
      "auxiliary_loss_mlp": 0.01033905,
      "balance_loss_clip": 1.058864,
      "balance_loss_mlp": 1.02514935,
      "epoch": 0.4178440449708411,
      "flos": 14643797316480.0,
      "grad_norm": 2.2572486798842406,
      "language_loss": 0.71336657,
      "learning_rate": 2.6190710813039917e-06,
      "loss": 0.73546028,
      "num_input_tokens_seen": 74944695,
      "step": 3475,
      "time_per_iteration": 2.491145133972168
    },
    {
      "auxiliary_loss_clip": 0.01119264,
      "auxiliary_loss_mlp": 0.00762643,
      "balance_loss_clip": 1.04452884,
      "balance_loss_mlp": 1.00023532,
      "epoch": 0.4179642878614802,
      "flos": 21507296094720.0,
      "grad_norm": 2.988858590721069,
      "language_loss": 0.83980376,
      "learning_rate": 2.618330319672747e-06,
      "loss": 0.85862279,
      "num_input_tokens_seen": 74964115,
      "step": 3476,
      "time_per_iteration": 2.6449849605560303
    },
    {
      "auxiliary_loss_clip": 0.01188608,
      "auxiliary_loss_mlp": 0.01031737,
      "balance_loss_clip": 1.05953002,
      "balance_loss_mlp": 1.02382731,
      "epoch": 0.41808453075211927,
      "flos": 18441997257600.0,
      "grad_norm": 2.1851191262485927,
      "language_loss": 0.92041111,
      "learning_rate": 2.617589464241284e-06,
      "loss": 0.94261456,
      "num_input_tokens_seen": 74978515,
      "step": 3477,
      "time_per_iteration": 2.487933397293091
    },
    {
      "auxiliary_loss_clip": 0.01149139,
      "auxiliary_loss_mlp": 0.01032851,
      "balance_loss_clip": 1.05523038,
      "balance_loss_mlp": 1.02556157,
      "epoch": 0.4182047736427584,
      "flos": 20301020628480.0,
      "grad_norm": 1.9018786591088341,
      "language_loss": 0.74397212,
      "learning_rate": 2.6168485151219914e-06,
      "loss": 0.76579201,
      "num_input_tokens_seen": 74998135,
      "step": 3478,
      "time_per_iteration": 2.643768072128296
    },
    {
      "auxiliary_loss_clip": 0.01173965,
      "auxiliary_loss_mlp": 0.01031619,
      "balance_loss_clip": 1.05908835,
      "balance_loss_mlp": 1.02348852,
      "epoch": 0.4183250165333975,
      "flos": 18876623823360.0,
      "grad_norm": 2.119472793802139,
      "language_loss": 0.71363485,
      "learning_rate": 2.616107472427269e-06,
      "loss": 0.73569071,
      "num_input_tokens_seen": 75012830,
      "step": 3479,
      "time_per_iteration": 2.514937400817871
    },
    {
      "auxiliary_loss_clip": 0.01174268,
      "auxiliary_loss_mlp": 0.01029563,
      "balance_loss_clip": 1.05443716,
      "balance_loss_mlp": 1.02110553,
      "epoch": 0.41844525942403654,
      "flos": 17740050698880.0,
      "grad_norm": 2.541750093137917,
      "language_loss": 0.76056945,
      "learning_rate": 2.615366336269533e-06,
      "loss": 0.78260779,
      "num_input_tokens_seen": 75026495,
      "step": 3480,
      "time_per_iteration": 2.5297300815582275
    },
    {
      "auxiliary_loss_clip": 0.01189484,
      "auxiliary_loss_mlp": 0.01034398,
      "balance_loss_clip": 1.05830002,
      "balance_loss_mlp": 1.02562976,
      "epoch": 0.41856550231467565,
      "flos": 18361377181440.0,
      "grad_norm": 2.4254761477534448,
      "language_loss": 0.80722034,
      "learning_rate": 2.6146251067612126e-06,
      "loss": 0.82945919,
      "num_input_tokens_seen": 75041970,
      "step": 3481,
      "time_per_iteration": 2.480426788330078
    },
    {
      "auxiliary_loss_clip": 0.01170415,
      "auxiliary_loss_mlp": 0.01032132,
      "balance_loss_clip": 1.05801404,
      "balance_loss_mlp": 1.02410364,
      "epoch": 0.41868574520531476,
      "flos": 22781801445120.0,
      "grad_norm": 1.8846427411514612,
      "language_loss": 0.82556105,
      "learning_rate": 2.6138837840147525e-06,
      "loss": 0.84758651,
      "num_input_tokens_seen": 75061005,
      "step": 3482,
      "time_per_iteration": 2.537447690963745
    },
    {
      "auxiliary_loss_clip": 0.01140685,
      "auxiliary_loss_mlp": 0.01032867,
      "balance_loss_clip": 1.05119205,
      "balance_loss_mlp": 1.02380729,
      "epoch": 0.4188059880959538,
      "flos": 13699167494400.0,
      "grad_norm": 2.174100577432015,
      "language_loss": 0.76577389,
      "learning_rate": 2.6131423681426103e-06,
      "loss": 0.78750932,
      "num_input_tokens_seen": 75076920,
      "step": 3483,
      "time_per_iteration": 2.5713999271392822
    },
    {
      "auxiliary_loss_clip": 0.01184262,
      "auxiliary_loss_mlp": 0.01034233,
      "balance_loss_clip": 1.05742574,
      "balance_loss_mlp": 1.0263114,
      "epoch": 0.41892623098659293,
      "flos": 37818281220480.0,
      "grad_norm": 1.7667775902007472,
      "language_loss": 0.72807598,
      "learning_rate": 2.6124008592572587e-06,
      "loss": 0.75026095,
      "num_input_tokens_seen": 75100905,
      "step": 3484,
      "time_per_iteration": 2.682039499282837
    },
    {
      "auxiliary_loss_clip": 0.0118785,
      "auxiliary_loss_mlp": 0.01030989,
      "balance_loss_clip": 1.05694234,
      "balance_loss_mlp": 1.02223945,
      "epoch": 0.419046473877232,
      "flos": 23258874908160.0,
      "grad_norm": 2.4692633686802505,
      "language_loss": 0.81359011,
      "learning_rate": 2.6116592574711835e-06,
      "loss": 0.83577847,
      "num_input_tokens_seen": 75119205,
      "step": 3485,
      "time_per_iteration": 2.5075910091400146
    },
    {
      "auxiliary_loss_clip": 0.01187857,
      "auxiliary_loss_mlp": 0.01034667,
      "balance_loss_clip": 1.0577904,
      "balance_loss_mlp": 1.0260365,
      "epoch": 0.4191667167678711,
      "flos": 20741034234240.0,
      "grad_norm": 1.8064727324105714,
      "language_loss": 0.83826619,
      "learning_rate": 2.6109175628968853e-06,
      "loss": 0.86049151,
      "num_input_tokens_seen": 75138970,
      "step": 3486,
      "time_per_iteration": 2.5484867095947266
    },
    {
      "auxiliary_loss_clip": 0.01159941,
      "auxiliary_loss_mlp": 0.0102829,
      "balance_loss_clip": 1.05244231,
      "balance_loss_mlp": 1.02010012,
      "epoch": 0.4192869596585102,
      "flos": 23586416052480.0,
      "grad_norm": 1.90220184249877,
      "language_loss": 0.82511795,
      "learning_rate": 2.610175775646878e-06,
      "loss": 0.84700024,
      "num_input_tokens_seen": 75157550,
      "step": 3487,
      "time_per_iteration": 2.5615739822387695
    },
    {
      "auxiliary_loss_clip": 0.01155452,
      "auxiliary_loss_mlp": 0.01035546,
      "balance_loss_clip": 1.05297065,
      "balance_loss_mlp": 1.02715969,
      "epoch": 0.41940720254914926,
      "flos": 25081269384960.0,
      "grad_norm": 1.885030405023015,
      "language_loss": 0.73065513,
      "learning_rate": 2.6094338958336907e-06,
      "loss": 0.75256515,
      "num_input_tokens_seen": 75176220,
      "step": 3488,
      "time_per_iteration": 2.6210906505584717
    },
    {
      "auxiliary_loss_clip": 0.01158059,
      "auxiliary_loss_mlp": 0.01030089,
      "balance_loss_clip": 1.05613291,
      "balance_loss_mlp": 1.02273703,
      "epoch": 0.41952744543978837,
      "flos": 15554132628480.0,
      "grad_norm": 1.9715129278974122,
      "language_loss": 0.81925255,
      "learning_rate": 2.608691923569867e-06,
      "loss": 0.84113407,
      "num_input_tokens_seen": 75193095,
      "step": 3489,
      "time_per_iteration": 2.5470285415649414
    },
    {
      "auxiliary_loss_clip": 0.01176651,
      "auxiliary_loss_mlp": 0.01037817,
      "balance_loss_clip": 1.0582782,
      "balance_loss_mlp": 1.02935886,
      "epoch": 0.4196476883304275,
      "flos": 24644775312000.0,
      "grad_norm": 1.664511613313216,
      "language_loss": 0.7576685,
      "learning_rate": 2.6079498589679616e-06,
      "loss": 0.77981317,
      "num_input_tokens_seen": 75214185,
      "step": 3490,
      "time_per_iteration": 2.5672545433044434
    },
    {
      "auxiliary_loss_clip": 0.01111322,
      "auxiliary_loss_mlp": 0.0103357,
      "balance_loss_clip": 1.04607856,
      "balance_loss_mlp": 1.02499902,
      "epoch": 0.41976793122106654,
      "flos": 24531333183360.0,
      "grad_norm": 1.953089129942015,
      "language_loss": 0.75910807,
      "learning_rate": 2.6072077021405465e-06,
      "loss": 0.78055704,
      "num_input_tokens_seen": 75233020,
      "step": 3491,
      "time_per_iteration": 2.7050507068634033
    },
    {
      "auxiliary_loss_clip": 0.0114834,
      "auxiliary_loss_mlp": 0.01030982,
      "balance_loss_clip": 1.05014682,
      "balance_loss_mlp": 1.02346635,
      "epoch": 0.41988817411170565,
      "flos": 21175301664000.0,
      "grad_norm": 1.8424427335420928,
      "language_loss": 0.69392455,
      "learning_rate": 2.6064654532002054e-06,
      "loss": 0.71571773,
      "num_input_tokens_seen": 75252030,
      "step": 3492,
      "time_per_iteration": 2.614002227783203
    },
    {
      "auxiliary_loss_clip": 0.01187712,
      "auxiliary_loss_mlp": 0.01033794,
      "balance_loss_clip": 1.0589813,
      "balance_loss_mlp": 1.02616167,
      "epoch": 0.42000841700234476,
      "flos": 31649402626560.0,
      "grad_norm": 3.262968948774268,
      "language_loss": 0.75843024,
      "learning_rate": 2.6057231122595375e-06,
      "loss": 0.78064537,
      "num_input_tokens_seen": 75273340,
      "step": 3493,
      "time_per_iteration": 3.445108413696289
    },
    {
      "auxiliary_loss_clip": 0.01158407,
      "auxiliary_loss_mlp": 0.01037653,
      "balance_loss_clip": 1.05245996,
      "balance_loss_mlp": 1.0295769,
      "epoch": 0.4201286598929838,
      "flos": 21281525159040.0,
      "grad_norm": 1.6243008911880297,
      "language_loss": 0.72834587,
      "learning_rate": 2.604980679431154e-06,
      "loss": 0.75030649,
      "num_input_tokens_seen": 75291580,
      "step": 3494,
      "time_per_iteration": 2.5698959827423096
    },
    {
      "auxiliary_loss_clip": 0.01172373,
      "auxiliary_loss_mlp": 0.01027361,
      "balance_loss_clip": 1.05228698,
      "balance_loss_mlp": 1.01926684,
      "epoch": 0.4202489027836229,
      "flos": 18546532813440.0,
      "grad_norm": 2.440048805860495,
      "language_loss": 0.74466449,
      "learning_rate": 2.604238154827684e-06,
      "loss": 0.76666188,
      "num_input_tokens_seen": 75308205,
      "step": 3495,
      "time_per_iteration": 2.5141279697418213
    },
    {
      "auxiliary_loss_clip": 0.01174765,
      "auxiliary_loss_mlp": 0.01026363,
      "balance_loss_clip": 1.05624902,
      "balance_loss_mlp": 1.0192225,
      "epoch": 0.42036914567426203,
      "flos": 19317643009920.0,
      "grad_norm": 1.8336332340323391,
      "language_loss": 0.72966468,
      "learning_rate": 2.6034955385617656e-06,
      "loss": 0.75167596,
      "num_input_tokens_seen": 75326535,
      "step": 3496,
      "time_per_iteration": 2.5247223377227783
    },
    {
      "auxiliary_loss_clip": 0.01047883,
      "auxiliary_loss_mlp": 0.01005175,
      "balance_loss_clip": 1.02145767,
      "balance_loss_mlp": 1.00387537,
      "epoch": 0.4204893885649011,
      "flos": 67842942935040.0,
      "grad_norm": 0.7282328108378927,
      "language_loss": 0.61719263,
      "learning_rate": 2.6027528307460544e-06,
      "loss": 0.63772321,
      "num_input_tokens_seen": 75390540,
      "step": 3497,
      "time_per_iteration": 3.25300669670105
    },
    {
      "auxiliary_loss_clip": 0.01187341,
      "auxiliary_loss_mlp": 0.01025224,
      "balance_loss_clip": 1.057621,
      "balance_loss_mlp": 1.01717734,
      "epoch": 0.4206096314555402,
      "flos": 21908777385600.0,
      "grad_norm": 1.943516556399264,
      "language_loss": 0.86929882,
      "learning_rate": 2.602010031493217e-06,
      "loss": 0.89142448,
      "num_input_tokens_seen": 75408770,
      "step": 3498,
      "time_per_iteration": 4.772298336029053
    },
    {
      "auxiliary_loss_clip": 0.0113765,
      "auxiliary_loss_mlp": 0.01028492,
      "balance_loss_clip": 1.04993105,
      "balance_loss_mlp": 1.0204339,
      "epoch": 0.42072987434617926,
      "flos": 29278185269760.0,
      "grad_norm": 1.8907431993242858,
      "language_loss": 0.87278438,
      "learning_rate": 2.6012671409159367e-06,
      "loss": 0.89444578,
      "num_input_tokens_seen": 75430105,
      "step": 3499,
      "time_per_iteration": 2.670034646987915
    },
    {
      "auxiliary_loss_clip": 0.01152741,
      "auxiliary_loss_mlp": 0.01024682,
      "balance_loss_clip": 1.05315411,
      "balance_loss_mlp": 1.01617694,
      "epoch": 0.42085011723681837,
      "flos": 27600726170880.0,
      "grad_norm": 1.5983131032612379,
      "language_loss": 0.81877744,
      "learning_rate": 2.6005241591269097e-06,
      "loss": 0.84055167,
      "num_input_tokens_seen": 75449475,
      "step": 3500,
      "time_per_iteration": 2.626570463180542
    },
    {
      "auxiliary_loss_clip": 0.01137967,
      "auxiliary_loss_mlp": 0.01032174,
      "balance_loss_clip": 1.05382907,
      "balance_loss_mlp": 1.02468491,
      "epoch": 0.4209703601274575,
      "flos": 27818632028160.0,
      "grad_norm": 1.8057243406085994,
      "language_loss": 0.79689336,
      "learning_rate": 2.5997810862388454e-06,
      "loss": 0.81859475,
      "num_input_tokens_seen": 75469315,
      "step": 3501,
      "time_per_iteration": 2.6550354957580566
    },
    {
      "auxiliary_loss_clip": 0.01154609,
      "auxiliary_loss_mlp": 0.01031868,
      "balance_loss_clip": 1.04993463,
      "balance_loss_mlp": 1.02337408,
      "epoch": 0.42109060301809653,
      "flos": 27525529048320.0,
      "grad_norm": 2.559728587930794,
      "language_loss": 0.75832236,
      "learning_rate": 2.599037922364467e-06,
      "loss": 0.78018713,
      "num_input_tokens_seen": 75488215,
      "step": 3502,
      "time_per_iteration": 2.62187123298645
    },
    {
      "auxiliary_loss_clip": 0.01135832,
      "auxiliary_loss_mlp": 0.01028823,
      "balance_loss_clip": 1.05138969,
      "balance_loss_mlp": 1.02121198,
      "epoch": 0.42121084590873564,
      "flos": 29314275459840.0,
      "grad_norm": 2.506327299717238,
      "language_loss": 0.75491107,
      "learning_rate": 2.5982946676165112e-06,
      "loss": 0.77655768,
      "num_input_tokens_seen": 75507985,
      "step": 3503,
      "time_per_iteration": 2.662526845932007
    },
    {
      "auxiliary_loss_clip": 0.01041473,
      "auxiliary_loss_mlp": 0.01005379,
      "balance_loss_clip": 1.02087712,
      "balance_loss_mlp": 1.00402021,
      "epoch": 0.42133108879937475,
      "flos": 67398835178880.0,
      "grad_norm": 0.7275168824376893,
      "language_loss": 0.57686812,
      "learning_rate": 2.5975513221077313e-06,
      "loss": 0.59733665,
      "num_input_tokens_seen": 75571955,
      "step": 3504,
      "time_per_iteration": 3.246126174926758
    },
    {
      "auxiliary_loss_clip": 0.01148034,
      "auxiliary_loss_mlp": 0.01032902,
      "balance_loss_clip": 1.05052674,
      "balance_loss_mlp": 1.02480757,
      "epoch": 0.4214513316900138,
      "flos": 23106038538240.0,
      "grad_norm": 2.230404719719764,
      "language_loss": 0.88608241,
      "learning_rate": 2.5968078859508897e-06,
      "loss": 0.90789175,
      "num_input_tokens_seen": 75589155,
      "step": 3505,
      "time_per_iteration": 2.576901435852051
    },
    {
      "auxiliary_loss_clip": 0.01171157,
      "auxiliary_loss_mlp": 0.01027064,
      "balance_loss_clip": 1.05472362,
      "balance_loss_mlp": 1.01921988,
      "epoch": 0.4215715745806529,
      "flos": 15336190857600.0,
      "grad_norm": 1.8264308526132738,
      "language_loss": 0.79725635,
      "learning_rate": 2.5960643592587673e-06,
      "loss": 0.8192386,
      "num_input_tokens_seen": 75606565,
      "step": 3506,
      "time_per_iteration": 2.5058271884918213
    },
    {
      "auxiliary_loss_clip": 0.01142186,
      "auxiliary_loss_mlp": 0.010333,
      "balance_loss_clip": 1.04925752,
      "balance_loss_mlp": 1.02517009,
      "epoch": 0.42169181747129203,
      "flos": 22127257860480.0,
      "grad_norm": 1.8339397834681952,
      "language_loss": 0.81494701,
      "learning_rate": 2.5953207421441553e-06,
      "loss": 0.83670187,
      "num_input_tokens_seen": 75625165,
      "step": 3507,
      "time_per_iteration": 2.602940559387207
    },
    {
      "auxiliary_loss_clip": 0.0114339,
      "auxiliary_loss_mlp": 0.01028844,
      "balance_loss_clip": 1.0502212,
      "balance_loss_mlp": 1.02098179,
      "epoch": 0.4218120603619311,
      "flos": 22630724841600.0,
      "grad_norm": 2.241868976252056,
      "language_loss": 0.75179935,
      "learning_rate": 2.5945770347198603e-06,
      "loss": 0.77352172,
      "num_input_tokens_seen": 75643320,
      "step": 3508,
      "time_per_iteration": 2.594249725341797
    },
    {
      "auxiliary_loss_clip": 0.01147995,
      "auxiliary_loss_mlp": 0.01029242,
      "balance_loss_clip": 1.04844332,
      "balance_loss_mlp": 1.02160621,
      "epoch": 0.4219323032525702,
      "flos": 19682818629120.0,
      "grad_norm": 1.7627605661178114,
      "language_loss": 0.81732285,
      "learning_rate": 2.593833237098701e-06,
      "loss": 0.83909523,
      "num_input_tokens_seen": 75660920,
      "step": 3509,
      "time_per_iteration": 2.566551446914673
    },
    {
      "auxiliary_loss_clip": 0.01167089,
      "auxiliary_loss_mlp": 0.0103271,
      "balance_loss_clip": 1.0495038,
      "balance_loss_mlp": 1.02370369,
      "epoch": 0.4220525461432093,
      "flos": 30190747224960.0,
      "grad_norm": 2.451371246360345,
      "language_loss": 0.62634528,
      "learning_rate": 2.593089349393512e-06,
      "loss": 0.64834332,
      "num_input_tokens_seen": 75681410,
      "step": 3510,
      "time_per_iteration": 2.5852394104003906
    },
    {
      "auxiliary_loss_clip": 0.01167049,
      "auxiliary_loss_mlp": 0.01037352,
      "balance_loss_clip": 1.05325079,
      "balance_loss_mlp": 1.02946639,
      "epoch": 0.42217278903384836,
      "flos": 24315941278080.0,
      "grad_norm": 2.4401091612846306,
      "language_loss": 0.83643866,
      "learning_rate": 2.592345371717141e-06,
      "loss": 0.85848272,
      "num_input_tokens_seen": 75700940,
      "step": 3511,
      "time_per_iteration": 2.542020559310913
    },
    {
      "auxiliary_loss_clip": 0.01172118,
      "auxiliary_loss_mlp": 0.01033573,
      "balance_loss_clip": 1.05724168,
      "balance_loss_mlp": 1.02489424,
      "epoch": 0.42229303192448747,
      "flos": 17092474352640.0,
      "grad_norm": 2.021320555818237,
      "language_loss": 0.72221094,
      "learning_rate": 2.591601304182448e-06,
      "loss": 0.74426794,
      "num_input_tokens_seen": 75718910,
      "step": 3512,
      "time_per_iteration": 2.49338960647583
    },
    {
      "auxiliary_loss_clip": 0.01157131,
      "auxiliary_loss_mlp": 0.01028901,
      "balance_loss_clip": 1.05381536,
      "balance_loss_mlp": 1.0211463,
      "epoch": 0.4224132748151266,
      "flos": 22784530878720.0,
      "grad_norm": 1.9811853750962924,
      "language_loss": 0.79165322,
      "learning_rate": 2.5908571469023067e-06,
      "loss": 0.81351352,
      "num_input_tokens_seen": 75738395,
      "step": 3513,
      "time_per_iteration": 2.5865509510040283
    },
    {
      "auxiliary_loss_clip": 0.01183031,
      "auxiliary_loss_mlp": 0.01032491,
      "balance_loss_clip": 1.0547049,
      "balance_loss_mlp": 1.02394414,
      "epoch": 0.42253351770576564,
      "flos": 17819090576640.0,
      "grad_norm": 2.454645099029267,
      "language_loss": 0.75859642,
      "learning_rate": 2.5901128999896067e-06,
      "loss": 0.78075165,
      "num_input_tokens_seen": 75753825,
      "step": 3514,
      "time_per_iteration": 2.444157838821411
    },
    {
      "auxiliary_loss_clip": 0.01168148,
      "auxiliary_loss_mlp": 0.01029988,
      "balance_loss_clip": 1.05263448,
      "balance_loss_mlp": 1.02181602,
      "epoch": 0.42265376059640475,
      "flos": 28512390286080.0,
      "grad_norm": 1.5885827255514073,
      "language_loss": 0.68141115,
      "learning_rate": 2.5893685635572487e-06,
      "loss": 0.70339251,
      "num_input_tokens_seen": 75774675,
      "step": 3515,
      "time_per_iteration": 2.5777931213378906
    },
    {
      "auxiliary_loss_clip": 0.01154641,
      "auxiliary_loss_mlp": 0.01034004,
      "balance_loss_clip": 1.05227923,
      "balance_loss_mlp": 1.0253433,
      "epoch": 0.4227740034870438,
      "flos": 16253349753600.0,
      "grad_norm": 2.421810760210724,
      "language_loss": 0.69649732,
      "learning_rate": 2.5886241377181483e-06,
      "loss": 0.71838379,
      "num_input_tokens_seen": 75793545,
      "step": 3516,
      "time_per_iteration": 2.544175863265991
    },
    {
      "auxiliary_loss_clip": 0.01171595,
      "auxiliary_loss_mlp": 0.01035153,
      "balance_loss_clip": 1.0533402,
      "balance_loss_mlp": 1.02606916,
      "epoch": 0.4228942463776829,
      "flos": 25295691623040.0,
      "grad_norm": 2.072457822094871,
      "language_loss": 0.81299049,
      "learning_rate": 2.587879622585234e-06,
      "loss": 0.83505791,
      "num_input_tokens_seen": 75812145,
      "step": 3517,
      "time_per_iteration": 2.563154935836792
    },
    {
      "auxiliary_loss_clip": 0.01170818,
      "auxiliary_loss_mlp": 0.01030964,
      "balance_loss_clip": 1.05642498,
      "balance_loss_mlp": 1.02346539,
      "epoch": 0.423014489268322,
      "flos": 26395779507840.0,
      "grad_norm": 1.9548951563634942,
      "language_loss": 0.76455414,
      "learning_rate": 2.5871350182714486e-06,
      "loss": 0.78657192,
      "num_input_tokens_seen": 75833025,
      "step": 3518,
      "time_per_iteration": 2.5727250576019287
    },
    {
      "auxiliary_loss_clip": 0.01181328,
      "auxiliary_loss_mlp": 0.01036689,
      "balance_loss_clip": 1.05291891,
      "balance_loss_mlp": 1.02860689,
      "epoch": 0.4231347321589611,
      "flos": 17274002711040.0,
      "grad_norm": 1.9751636009400335,
      "language_loss": 0.80377674,
      "learning_rate": 2.586390324889748e-06,
      "loss": 0.82595694,
      "num_input_tokens_seen": 75848925,
      "step": 3519,
      "time_per_iteration": 3.279618263244629
    },
    {
      "auxiliary_loss_clip": 0.01162624,
      "auxiliary_loss_mlp": 0.01032535,
      "balance_loss_clip": 1.0500623,
      "balance_loss_mlp": 1.02451229,
      "epoch": 0.4232549750496002,
      "flos": 22999635475200.0,
      "grad_norm": 1.7920387712202162,
      "language_loss": 0.67368472,
      "learning_rate": 2.5856455425531003e-06,
      "loss": 0.69563627,
      "num_input_tokens_seen": 75870400,
      "step": 3520,
      "time_per_iteration": 2.535332679748535
    },
    {
      "auxiliary_loss_clip": 0.01168216,
      "auxiliary_loss_mlp": 0.01032542,
      "balance_loss_clip": 1.05334187,
      "balance_loss_mlp": 1.02470422,
      "epoch": 0.4233752179402393,
      "flos": 21248343970560.0,
      "grad_norm": 2.4736974637994256,
      "language_loss": 0.80687439,
      "learning_rate": 2.5849006713744902e-06,
      "loss": 0.82888186,
      "num_input_tokens_seen": 75889195,
      "step": 3521,
      "time_per_iteration": 2.5192248821258545
    },
    {
      "auxiliary_loss_clip": 0.01149078,
      "auxiliary_loss_mlp": 0.01027338,
      "balance_loss_clip": 1.05004549,
      "balance_loss_mlp": 1.01880825,
      "epoch": 0.42349546083087836,
      "flos": 20704297599360.0,
      "grad_norm": 3.0480304614226235,
      "language_loss": 0.72776365,
      "learning_rate": 2.5841557114669135e-06,
      "loss": 0.74952781,
      "num_input_tokens_seen": 75906055,
      "step": 3522,
      "time_per_iteration": 2.551253318786621
    },
    {
      "auxiliary_loss_clip": 0.0118554,
      "auxiliary_loss_mlp": 0.01032693,
      "balance_loss_clip": 1.05399561,
      "balance_loss_mlp": 1.02409792,
      "epoch": 0.42361570372151747,
      "flos": 18585065128320.0,
      "grad_norm": 3.084171921132483,
      "language_loss": 0.67372882,
      "learning_rate": 2.58341066294338e-06,
      "loss": 0.69591117,
      "num_input_tokens_seen": 75922720,
      "step": 3523,
      "time_per_iteration": 2.446063995361328
    },
    {
      "auxiliary_loss_clip": 0.01129357,
      "auxiliary_loss_mlp": 0.00762126,
      "balance_loss_clip": 1.04828715,
      "balance_loss_mlp": 1.00026202,
      "epoch": 0.4237359466121566,
      "flos": 20959478795520.0,
      "grad_norm": 3.675846409416609,
      "language_loss": 0.85560393,
      "learning_rate": 2.5826655259169124e-06,
      "loss": 0.87451881,
      "num_input_tokens_seen": 75941375,
      "step": 3524,
      "time_per_iteration": 4.19012975692749
    },
    {
      "auxiliary_loss_clip": 0.01186861,
      "auxiliary_loss_mlp": 0.01029791,
      "balance_loss_clip": 1.05769765,
      "balance_loss_mlp": 1.02187538,
      "epoch": 0.42385618950279563,
      "flos": 18038181582720.0,
      "grad_norm": 1.8096748581740825,
      "language_loss": 0.90427226,
      "learning_rate": 2.5819203005005475e-06,
      "loss": 0.92643869,
      "num_input_tokens_seen": 75958710,
      "step": 3525,
      "time_per_iteration": 3.3526864051818848
    },
    {
      "auxiliary_loss_clip": 0.0114837,
      "auxiliary_loss_mlp": 0.01027161,
      "balance_loss_clip": 1.05152822,
      "balance_loss_mlp": 1.01975775,
      "epoch": 0.42397643239343474,
      "flos": 23769129559680.0,
      "grad_norm": 1.6790216191579026,
      "language_loss": 0.78910536,
      "learning_rate": 2.581174986807336e-06,
      "loss": 0.81086063,
      "num_input_tokens_seen": 75978945,
      "step": 3526,
      "time_per_iteration": 2.556105852127075
    },
    {
      "auxiliary_loss_clip": 0.01160695,
      "auxiliary_loss_mlp": 0.00762434,
      "balance_loss_clip": 1.05174851,
      "balance_loss_mlp": 1.0002265,
      "epoch": 0.42409667528407385,
      "flos": 16545088016640.0,
      "grad_norm": 2.080561334251366,
      "language_loss": 0.91387498,
      "learning_rate": 2.580429584950341e-06,
      "loss": 0.9331063,
      "num_input_tokens_seen": 75994695,
      "step": 3527,
      "time_per_iteration": 2.520254611968994
    },
    {
      "auxiliary_loss_clip": 0.01144012,
      "auxiliary_loss_mlp": 0.01031,
      "balance_loss_clip": 1.05041981,
      "balance_loss_mlp": 1.02239311,
      "epoch": 0.4242169181747129,
      "flos": 16034186920320.0,
      "grad_norm": 2.117491170732389,
      "language_loss": 0.66232002,
      "learning_rate": 2.5796840950426397e-06,
      "loss": 0.68407017,
      "num_input_tokens_seen": 76011780,
      "step": 3528,
      "time_per_iteration": 2.5469422340393066
    },
    {
      "auxiliary_loss_clip": 0.01159186,
      "auxiliary_loss_mlp": 0.01025044,
      "balance_loss_clip": 1.05034685,
      "balance_loss_mlp": 1.01699185,
      "epoch": 0.424337161065352,
      "flos": 20084012611200.0,
      "grad_norm": 1.6774667491020243,
      "language_loss": 0.65460372,
      "learning_rate": 2.578938517197322e-06,
      "loss": 0.67644608,
      "num_input_tokens_seen": 76029875,
      "step": 3529,
      "time_per_iteration": 2.5285439491271973
    },
    {
      "auxiliary_loss_clip": 0.0114601,
      "auxiliary_loss_mlp": 0.0103564,
      "balance_loss_clip": 1.05062509,
      "balance_loss_mlp": 1.0276649,
      "epoch": 0.4244574039559911,
      "flos": 23878369797120.0,
      "grad_norm": 2.312528248818574,
      "language_loss": 0.6215831,
      "learning_rate": 2.5781928515274916e-06,
      "loss": 0.6433996,
      "num_input_tokens_seen": 76048595,
      "step": 3530,
      "time_per_iteration": 2.5553436279296875
    },
    {
      "auxiliary_loss_clip": 0.01173968,
      "auxiliary_loss_mlp": 0.01030239,
      "balance_loss_clip": 1.05647421,
      "balance_loss_mlp": 1.02170324,
      "epoch": 0.4245776468466302,
      "flos": 17565920542080.0,
      "grad_norm": 2.066287835927777,
      "language_loss": 0.67784119,
      "learning_rate": 2.577447098146265e-06,
      "loss": 0.69988328,
      "num_input_tokens_seen": 76065770,
      "step": 3531,
      "time_per_iteration": 2.5095884799957275
    },
    {
      "auxiliary_loss_clip": 0.01144525,
      "auxiliary_loss_mlp": 0.01033597,
      "balance_loss_clip": 1.05100989,
      "balance_loss_mlp": 1.0257293,
      "epoch": 0.4246978897372693,
      "flos": 27776256958080.0,
      "grad_norm": 1.6135778046324214,
      "language_loss": 0.79102302,
      "learning_rate": 2.5767012571667724e-06,
      "loss": 0.81280422,
      "num_input_tokens_seen": 76085250,
      "step": 3532,
      "time_per_iteration": 2.6287782192230225
    },
    {
      "auxiliary_loss_clip": 0.01168454,
      "auxiliary_loss_mlp": 0.01028374,
      "balance_loss_clip": 1.04976439,
      "balance_loss_mlp": 1.01955259,
      "epoch": 0.42481813262790835,
      "flos": 15596615439360.0,
      "grad_norm": 1.8427547775192907,
      "language_loss": 0.68474454,
      "learning_rate": 2.5759553287021587e-06,
      "loss": 0.70671278,
      "num_input_tokens_seen": 76103580,
      "step": 3533,
      "time_per_iteration": 2.512155294418335
    },
    {
      "auxiliary_loss_clip": 0.01152838,
      "auxiliary_loss_mlp": 0.01033142,
      "balance_loss_clip": 1.05119658,
      "balance_loss_mlp": 1.02472031,
      "epoch": 0.42493837551854746,
      "flos": 23951088881280.0,
      "grad_norm": 2.042815085442589,
      "language_loss": 0.77572459,
      "learning_rate": 2.5752093128655786e-06,
      "loss": 0.79758435,
      "num_input_tokens_seen": 76121825,
      "step": 3534,
      "time_per_iteration": 2.5657310485839844
    },
    {
      "auxiliary_loss_clip": 0.01146404,
      "auxiliary_loss_mlp": 0.01035371,
      "balance_loss_clip": 1.04886365,
      "balance_loss_mlp": 1.02679992,
      "epoch": 0.4250586184091866,
      "flos": 20813466009600.0,
      "grad_norm": 1.6545393258571086,
      "language_loss": 0.7355082,
      "learning_rate": 2.574463209770204e-06,
      "loss": 0.75732595,
      "num_input_tokens_seen": 76141140,
      "step": 3535,
      "time_per_iteration": 2.563304901123047
    },
    {
      "auxiliary_loss_clip": 0.01134553,
      "auxiliary_loss_mlp": 0.01033312,
      "balance_loss_clip": 1.04462647,
      "balance_loss_mlp": 1.02459157,
      "epoch": 0.42517886129982563,
      "flos": 30371018607360.0,
      "grad_norm": 1.790897842677319,
      "language_loss": 0.79517722,
      "learning_rate": 2.5737170195292165e-06,
      "loss": 0.81685585,
      "num_input_tokens_seen": 76164475,
      "step": 3536,
      "time_per_iteration": 2.6618492603302
    },
    {
      "auxiliary_loss_clip": 0.01142198,
      "auxiliary_loss_mlp": 0.01030455,
      "balance_loss_clip": 1.04954755,
      "balance_loss_mlp": 1.02264118,
      "epoch": 0.42529910419046474,
      "flos": 20080636732800.0,
      "grad_norm": 1.7539232116611918,
      "language_loss": 0.77965492,
      "learning_rate": 2.572970742255814e-06,
      "loss": 0.80138147,
      "num_input_tokens_seen": 76182965,
      "step": 3537,
      "time_per_iteration": 2.6171765327453613
    },
    {
      "auxiliary_loss_clip": 0.01169664,
      "auxiliary_loss_mlp": 0.01032677,
      "balance_loss_clip": 1.05544984,
      "balance_loss_mlp": 1.02507186,
      "epoch": 0.42541934708110385,
      "flos": 22632448694400.0,
      "grad_norm": 1.6098970344324968,
      "language_loss": 0.81619358,
      "learning_rate": 2.5722243780632046e-06,
      "loss": 0.83821702,
      "num_input_tokens_seen": 76201230,
      "step": 3538,
      "time_per_iteration": 2.5432956218719482
    },
    {
      "auxiliary_loss_clip": 0.01036457,
      "auxiliary_loss_mlp": 0.01001983,
      "balance_loss_clip": 1.0186044,
      "balance_loss_mlp": 1.00064743,
      "epoch": 0.4255395899717429,
      "flos": 66200676186240.0,
      "grad_norm": 0.7460297209605776,
      "language_loss": 0.60482728,
      "learning_rate": 2.5714779270646125e-06,
      "loss": 0.62521172,
      "num_input_tokens_seen": 76262000,
      "step": 3539,
      "time_per_iteration": 3.1595652103424072
    },
    {
      "auxiliary_loss_clip": 0.01159473,
      "auxiliary_loss_mlp": 0.00761874,
      "balance_loss_clip": 1.05507588,
      "balance_loss_mlp": 1.00023866,
      "epoch": 0.425659832862382,
      "flos": 17931814433280.0,
      "grad_norm": 2.3529548464870906,
      "language_loss": 0.77055436,
      "learning_rate": 2.5707313893732735e-06,
      "loss": 0.78976786,
      "num_input_tokens_seen": 76280540,
      "step": 3540,
      "time_per_iteration": 2.5637471675872803
    },
    {
      "auxiliary_loss_clip": 0.0109095,
      "auxiliary_loss_mlp": 0.010262,
      "balance_loss_clip": 1.04064798,
      "balance_loss_mlp": 1.01845145,
      "epoch": 0.4257800757530211,
      "flos": 24022550989440.0,
      "grad_norm": 1.715210662677564,
      "language_loss": 0.76539767,
      "learning_rate": 2.5699847651024364e-06,
      "loss": 0.78656912,
      "num_input_tokens_seen": 76301180,
      "step": 3541,
      "time_per_iteration": 2.7320802211761475
    },
    {
      "auxiliary_loss_clip": 0.01169087,
      "auxiliary_loss_mlp": 0.01028247,
      "balance_loss_clip": 1.05576897,
      "balance_loss_mlp": 1.02048612,
      "epoch": 0.4259003186436602,
      "flos": 23696015425920.0,
      "grad_norm": 2.0296911777047595,
      "language_loss": 0.77134466,
      "learning_rate": 2.5692380543653627e-06,
      "loss": 0.79331803,
      "num_input_tokens_seen": 76319335,
      "step": 3542,
      "time_per_iteration": 2.580747604370117
    },
    {
      "auxiliary_loss_clip": 0.01173677,
      "auxiliary_loss_mlp": 0.00761962,
      "balance_loss_clip": 1.05470514,
      "balance_loss_mlp": 1.00024724,
      "epoch": 0.4260205615342993,
      "flos": 15259772672640.0,
      "grad_norm": 1.9255014320585002,
      "language_loss": 0.69535279,
      "learning_rate": 2.5684912572753293e-06,
      "loss": 0.71470916,
      "num_input_tokens_seen": 76335010,
      "step": 3543,
      "time_per_iteration": 2.491575002670288
    },
    {
      "auxiliary_loss_clip": 0.01179138,
      "auxiliary_loss_mlp": 0.01030458,
      "balance_loss_clip": 1.05357289,
      "balance_loss_mlp": 1.02338314,
      "epoch": 0.4261408044249384,
      "flos": 30665306736000.0,
      "grad_norm": 1.8082419896929556,
      "language_loss": 0.83797491,
      "learning_rate": 2.5677443739456245e-06,
      "loss": 0.86007088,
      "num_input_tokens_seen": 76356670,
      "step": 3544,
      "time_per_iteration": 2.570080280303955
    },
    {
      "auxiliary_loss_clip": 0.01154523,
      "auxiliary_loss_mlp": 0.01026979,
      "balance_loss_clip": 1.05202174,
      "balance_loss_mlp": 1.01952505,
      "epoch": 0.42626104731557746,
      "flos": 23257905240960.0,
      "grad_norm": 2.7780525164864804,
      "language_loss": 0.79281634,
      "learning_rate": 2.5669974044895495e-06,
      "loss": 0.81463134,
      "num_input_tokens_seen": 76373065,
      "step": 3545,
      "time_per_iteration": 3.3444738388061523
    },
    {
      "auxiliary_loss_clip": 0.0114776,
      "auxiliary_loss_mlp": 0.01026418,
      "balance_loss_clip": 1.05002415,
      "balance_loss_mlp": 1.0195874,
      "epoch": 0.42638129020621657,
      "flos": 25884770670720.0,
      "grad_norm": 1.721572569615772,
      "language_loss": 0.79546058,
      "learning_rate": 2.5662503490204187e-06,
      "loss": 0.81720233,
      "num_input_tokens_seen": 76393230,
      "step": 3546,
      "time_per_iteration": 2.659017562866211
    },
    {
      "auxiliary_loss_clip": 0.01151841,
      "auxiliary_loss_mlp": 0.01025179,
      "balance_loss_clip": 1.04820275,
      "balance_loss_mlp": 1.01747239,
      "epoch": 0.4265015330968556,
      "flos": 26502362138880.0,
      "grad_norm": 1.6924505897823745,
      "language_loss": 0.76018107,
      "learning_rate": 2.5655032076515603e-06,
      "loss": 0.78195119,
      "num_input_tokens_seen": 76412555,
      "step": 3547,
      "time_per_iteration": 2.596353530883789
    },
    {
      "auxiliary_loss_clip": 0.01155073,
      "auxiliary_loss_mlp": 0.01030461,
      "balance_loss_clip": 1.05291629,
      "balance_loss_mlp": 1.02263474,
      "epoch": 0.42662177598749473,
      "flos": 24389522288640.0,
      "grad_norm": 1.9785522239464388,
      "language_loss": 0.82257265,
      "learning_rate": 2.5647559804963155e-06,
      "loss": 0.844428,
      "num_input_tokens_seen": 76432485,
      "step": 3548,
      "time_per_iteration": 2.5991241931915283
    },
    {
      "auxiliary_loss_clip": 0.01134055,
      "auxiliary_loss_mlp": 0.01028829,
      "balance_loss_clip": 1.05037582,
      "balance_loss_mlp": 1.02105045,
      "epoch": 0.42674201887813384,
      "flos": 23148629089920.0,
      "grad_norm": 2.0302918210737135,
      "language_loss": 0.79135823,
      "learning_rate": 2.5640086676680364e-06,
      "loss": 0.81298703,
      "num_input_tokens_seen": 76453980,
      "step": 3549,
      "time_per_iteration": 2.64599609375
    },
    {
      "auxiliary_loss_clip": 0.01167609,
      "auxiliary_loss_mlp": 0.01034126,
      "balance_loss_clip": 1.05155659,
      "balance_loss_mlp": 1.02545953,
      "epoch": 0.4268622617687729,
      "flos": 21689614552320.0,
      "grad_norm": 2.3509104120552795,
      "language_loss": 0.80175078,
      "learning_rate": 2.5632612692800923e-06,
      "loss": 0.82376808,
      "num_input_tokens_seen": 76473045,
      "step": 3550,
      "time_per_iteration": 4.062756299972534
    },
    {
      "auxiliary_loss_clip": 0.01144022,
      "auxiliary_loss_mlp": 0.01029329,
      "balance_loss_clip": 1.05178022,
      "balance_loss_mlp": 1.02091837,
      "epoch": 0.426982504659412,
      "flos": 23440151871360.0,
      "grad_norm": 2.055721493537037,
      "language_loss": 0.75207937,
      "learning_rate": 2.5625137854458603e-06,
      "loss": 0.77381283,
      "num_input_tokens_seen": 76492060,
      "step": 3551,
      "time_per_iteration": 2.63824200630188
    },
    {
      "auxiliary_loss_clip": 0.01158515,
      "auxiliary_loss_mlp": 0.010228,
      "balance_loss_clip": 1.05308771,
      "balance_loss_mlp": 1.01540899,
      "epoch": 0.4271027475500511,
      "flos": 18916556768640.0,
      "grad_norm": 1.9502123933766422,
      "language_loss": 0.802019,
      "learning_rate": 2.561766216278735e-06,
      "loss": 0.82383215,
      "num_input_tokens_seen": 76509655,
      "step": 3552,
      "time_per_iteration": 2.5512309074401855
    },
    {
      "auxiliary_loss_clip": 0.01123839,
      "auxiliary_loss_mlp": 0.0103535,
      "balance_loss_clip": 1.0474546,
      "balance_loss_mlp": 1.02701092,
      "epoch": 0.4272229904406902,
      "flos": 26870554500480.0,
      "grad_norm": 2.36861838428947,
      "language_loss": 0.81483746,
      "learning_rate": 2.561018561892121e-06,
      "loss": 0.83642936,
      "num_input_tokens_seen": 76528795,
      "step": 3553,
      "time_per_iteration": 2.710111379623413
    },
    {
      "auxiliary_loss_clip": 0.01150858,
      "auxiliary_loss_mlp": 0.01030347,
      "balance_loss_clip": 1.04971576,
      "balance_loss_mlp": 1.02256227,
      "epoch": 0.4273432333313293,
      "flos": 23951376190080.0,
      "grad_norm": 1.5264465562472036,
      "language_loss": 0.76468539,
      "learning_rate": 2.5602708223994363e-06,
      "loss": 0.78649735,
      "num_input_tokens_seen": 76550660,
      "step": 3554,
      "time_per_iteration": 2.5845320224761963
    },
    {
      "auxiliary_loss_clip": 0.01141492,
      "auxiliary_loss_mlp": 0.01032281,
      "balance_loss_clip": 1.04617262,
      "balance_loss_mlp": 1.02391791,
      "epoch": 0.4274634762219684,
      "flos": 29570354496000.0,
      "grad_norm": 4.246482863044678,
      "language_loss": 0.67731261,
      "learning_rate": 2.559522997914115e-06,
      "loss": 0.69905031,
      "num_input_tokens_seen": 76570240,
      "step": 3555,
      "time_per_iteration": 2.652238607406616
    },
    {
      "auxiliary_loss_clip": 0.01181905,
      "auxiliary_loss_mlp": 0.01029393,
      "balance_loss_clip": 1.05590987,
      "balance_loss_mlp": 1.02181101,
      "epoch": 0.42758371911260745,
      "flos": 21434146047360.0,
      "grad_norm": 2.5309860042308645,
      "language_loss": 0.84552777,
      "learning_rate": 2.558775088549599e-06,
      "loss": 0.86764073,
      "num_input_tokens_seen": 76589820,
      "step": 3556,
      "time_per_iteration": 2.4821133613586426
    },
    {
      "auxiliary_loss_clip": 0.01171928,
      "auxiliary_loss_mlp": 0.01035808,
      "balance_loss_clip": 1.05196667,
      "balance_loss_mlp": 1.02696323,
      "epoch": 0.42770396200324656,
      "flos": 14752822072320.0,
      "grad_norm": 2.841339589706848,
      "language_loss": 0.6683538,
      "learning_rate": 2.5580270944193467e-06,
      "loss": 0.69043118,
      "num_input_tokens_seen": 76606640,
      "step": 3557,
      "time_per_iteration": 2.4975662231445312
    },
    {
      "auxiliary_loss_clip": 0.01079801,
      "auxiliary_loss_mlp": 0.01003127,
      "balance_loss_clip": 1.02344513,
      "balance_loss_mlp": 1.00173807,
      "epoch": 0.4278242048938857,
      "flos": 70654712601600.0,
      "grad_norm": 0.746915425445617,
      "language_loss": 0.5549078,
      "learning_rate": 2.557279015636827e-06,
      "loss": 0.57573706,
      "num_input_tokens_seen": 76667050,
      "step": 3558,
      "time_per_iteration": 3.076401472091675
    },
    {
      "auxiliary_loss_clip": 0.01063261,
      "auxiliary_loss_mlp": 0.01002362,
      "balance_loss_clip": 1.01901424,
      "balance_loss_mlp": 1.00104463,
      "epoch": 0.42794444778452473,
      "flos": 69366165033600.0,
      "grad_norm": 0.7704086459669501,
      "language_loss": 0.61241019,
      "learning_rate": 2.5565308523155245e-06,
      "loss": 0.63306642,
      "num_input_tokens_seen": 76726650,
      "step": 3559,
      "time_per_iteration": 3.0400943756103516
    },
    {
      "auxiliary_loss_clip": 0.01124021,
      "auxiliary_loss_mlp": 0.01026229,
      "balance_loss_clip": 1.05110693,
      "balance_loss_mlp": 1.01843262,
      "epoch": 0.42806469067516384,
      "flos": 18215328481920.0,
      "grad_norm": 2.2350273676517443,
      "language_loss": 0.82064974,
      "learning_rate": 2.5557826045689336e-06,
      "loss": 0.84215224,
      "num_input_tokens_seen": 76742890,
      "step": 3560,
      "time_per_iteration": 2.602969169616699
    },
    {
      "auxiliary_loss_clip": 0.01040104,
      "auxiliary_loss_mlp": 0.01008519,
      "balance_loss_clip": 1.01976562,
      "balance_loss_mlp": 1.00715411,
      "epoch": 0.4281849335658029,
      "flos": 54535814432640.0,
      "grad_norm": 0.8228894270035056,
      "language_loss": 0.58820242,
      "learning_rate": 2.5550342725105643e-06,
      "loss": 0.60868865,
      "num_input_tokens_seen": 76801055,
      "step": 3561,
      "time_per_iteration": 3.1535661220550537
    },
    {
      "auxiliary_loss_clip": 0.01171641,
      "auxiliary_loss_mlp": 0.01028789,
      "balance_loss_clip": 1.05600858,
      "balance_loss_mlp": 1.0208497,
      "epoch": 0.428305176456442,
      "flos": 17274828723840.0,
      "grad_norm": 1.771844185203597,
      "language_loss": 0.80860913,
      "learning_rate": 2.554285856253937e-06,
      "loss": 0.83061337,
      "num_input_tokens_seen": 76819890,
      "step": 3562,
      "time_per_iteration": 2.547060251235962
    },
    {
      "auxiliary_loss_clip": 0.01154747,
      "auxiliary_loss_mlp": 0.01032753,
      "balance_loss_clip": 1.05499101,
      "balance_loss_mlp": 1.02462888,
      "epoch": 0.4284254193470811,
      "flos": 26359509749760.0,
      "grad_norm": 1.7018321167782147,
      "language_loss": 0.77253103,
      "learning_rate": 2.5535373559125855e-06,
      "loss": 0.79440606,
      "num_input_tokens_seen": 76840255,
      "step": 3563,
      "time_per_iteration": 2.6081290245056152
    },
    {
      "auxiliary_loss_clip": 0.01099273,
      "auxiliary_loss_mlp": 0.01033802,
      "balance_loss_clip": 1.04551113,
      "balance_loss_mlp": 1.02549314,
      "epoch": 0.42854566223772017,
      "flos": 29714248379520.0,
      "grad_norm": 1.5543949688232546,
      "language_loss": 0.82085967,
      "learning_rate": 2.552788771600057e-06,
      "loss": 0.84219038,
      "num_input_tokens_seen": 76860565,
      "step": 3564,
      "time_per_iteration": 2.7671759128570557
    },
    {
      "auxiliary_loss_clip": 0.0114564,
      "auxiliary_loss_mlp": 0.01028406,
      "balance_loss_clip": 1.05305481,
      "balance_loss_mlp": 1.01999009,
      "epoch": 0.4286659051283593,
      "flos": 22018161277440.0,
      "grad_norm": 1.8908039603351283,
      "language_loss": 0.8197329,
      "learning_rate": 2.5520401034299118e-06,
      "loss": 0.8414734,
      "num_input_tokens_seen": 76878325,
      "step": 3565,
      "time_per_iteration": 2.5884132385253906
    },
    {
      "auxiliary_loss_clip": 0.01169523,
      "auxiliary_loss_mlp": 0.01030344,
      "balance_loss_clip": 1.051718,
      "balance_loss_mlp": 1.02167726,
      "epoch": 0.4287861480189984,
      "flos": 13334422838400.0,
      "grad_norm": 1.8221299009625975,
      "language_loss": 0.87862378,
      "learning_rate": 2.551291351515722e-06,
      "loss": 0.90062249,
      "num_input_tokens_seen": 76895340,
      "step": 3566,
      "time_per_iteration": 2.531500816345215
    },
    {
      "auxiliary_loss_clip": 0.01135825,
      "auxiliary_loss_mlp": 0.00761798,
      "balance_loss_clip": 1.04648161,
      "balance_loss_mlp": 1.0002017,
      "epoch": 0.42890639090963745,
      "flos": 26651535321600.0,
      "grad_norm": 1.6950210506206846,
      "language_loss": 0.85685831,
      "learning_rate": 2.5505425159710726e-06,
      "loss": 0.87583458,
      "num_input_tokens_seen": 76915150,
      "step": 3567,
      "time_per_iteration": 2.643615245819092
    },
    {
      "auxiliary_loss_clip": 0.01160972,
      "auxiliary_loss_mlp": 0.00761588,
      "balance_loss_clip": 1.05129385,
      "balance_loss_mlp": 1.00017905,
      "epoch": 0.42902663380027656,
      "flos": 24055768091520.0,
      "grad_norm": 1.7762381468990922,
      "language_loss": 0.82559282,
      "learning_rate": 2.549793596909561e-06,
      "loss": 0.84481847,
      "num_input_tokens_seen": 76933770,
      "step": 3568,
      "time_per_iteration": 2.5990307331085205
    },
    {
      "auxiliary_loss_clip": 0.01150202,
      "auxiliary_loss_mlp": 0.0102549,
      "balance_loss_clip": 1.05101418,
      "balance_loss_mlp": 1.01787901,
      "epoch": 0.42914687669091567,
      "flos": 15632561975040.0,
      "grad_norm": 1.9548974937729346,
      "language_loss": 0.65694249,
      "learning_rate": 2.5490445944447976e-06,
      "loss": 0.67869943,
      "num_input_tokens_seen": 76952265,
      "step": 3569,
      "time_per_iteration": 2.5296788215637207
    },
    {
      "auxiliary_loss_clip": 0.01169766,
      "auxiliary_loss_mlp": 0.01031722,
      "balance_loss_clip": 1.0541482,
      "balance_loss_mlp": 1.02402687,
      "epoch": 0.4292671195815547,
      "flos": 31467802440960.0,
      "grad_norm": 1.9940041934317976,
      "language_loss": 0.65264922,
      "learning_rate": 2.548295508690406e-06,
      "loss": 0.67466414,
      "num_input_tokens_seen": 76973560,
      "step": 3570,
      "time_per_iteration": 3.3866593837738037
    },
    {
      "auxiliary_loss_clip": 0.01171313,
      "auxiliary_loss_mlp": 0.01031306,
      "balance_loss_clip": 1.05283761,
      "balance_loss_mlp": 1.02295542,
      "epoch": 0.42938736247219383,
      "flos": 30257756046720.0,
      "grad_norm": 3.0222863447335544,
      "language_loss": 0.76473081,
      "learning_rate": 2.5475463397600217e-06,
      "loss": 0.78675699,
      "num_input_tokens_seen": 76993640,
      "step": 3571,
      "time_per_iteration": 2.612635612487793
    },
    {
      "auxiliary_loss_clip": 0.01189098,
      "auxiliary_loss_mlp": 0.01031115,
      "balance_loss_clip": 1.05813837,
      "balance_loss_mlp": 1.02238286,
      "epoch": 0.42950760536283294,
      "flos": 29349683291520.0,
      "grad_norm": 2.2183247363756102,
      "language_loss": 0.77527893,
      "learning_rate": 2.546797087767293e-06,
      "loss": 0.79748106,
      "num_input_tokens_seen": 77013765,
      "step": 3572,
      "time_per_iteration": 2.534398317337036
    },
    {
      "auxiliary_loss_clip": 0.01125453,
      "auxiliary_loss_mlp": 0.0103065,
      "balance_loss_clip": 1.05023336,
      "balance_loss_mlp": 1.02196574,
      "epoch": 0.429627848253472,
      "flos": 26869943969280.0,
      "grad_norm": 1.727589318268984,
      "language_loss": 0.8742559,
      "learning_rate": 2.546047752825881e-06,
      "loss": 0.89581692,
      "num_input_tokens_seen": 77034370,
      "step": 3573,
      "time_per_iteration": 2.6650795936584473
    },
    {
      "auxiliary_loss_clip": 0.0113314,
      "auxiliary_loss_mlp": 0.01026882,
      "balance_loss_clip": 1.04841328,
      "balance_loss_mlp": 1.01896644,
      "epoch": 0.4297480911441111,
      "flos": 13881270470400.0,
      "grad_norm": 1.9764190050206698,
      "language_loss": 0.93142402,
      "learning_rate": 2.5452983350494595e-06,
      "loss": 0.95302415,
      "num_input_tokens_seen": 77049925,
      "step": 3574,
      "time_per_iteration": 2.5811736583709717
    },
    {
      "auxiliary_loss_clip": 0.01172147,
      "auxiliary_loss_mlp": 0.00761925,
      "balance_loss_clip": 1.05435908,
      "balance_loss_mlp": 1.00017154,
      "epoch": 0.4298683340347502,
      "flos": 20741141975040.0,
      "grad_norm": 11.889245555598064,
      "language_loss": 0.65001309,
      "learning_rate": 2.544548834551713e-06,
      "loss": 0.66935384,
      "num_input_tokens_seen": 77068930,
      "step": 3575,
      "time_per_iteration": 2.540356159210205
    },
    {
      "auxiliary_loss_clip": 0.01137615,
      "auxiliary_loss_mlp": 0.00761924,
      "balance_loss_clip": 1.05086493,
      "balance_loss_mlp": 1.00019598,
      "epoch": 0.4299885769253893,
      "flos": 20882126856960.0,
      "grad_norm": 2.342104251670206,
      "language_loss": 0.94706917,
      "learning_rate": 2.5437992514463424e-06,
      "loss": 0.96606463,
      "num_input_tokens_seen": 77082255,
      "step": 3576,
      "time_per_iteration": 3.4396071434020996
    },
    {
      "auxiliary_loss_clip": 0.0116622,
      "auxiliary_loss_mlp": 0.01032093,
      "balance_loss_clip": 1.05347705,
      "balance_loss_mlp": 1.0238378,
      "epoch": 0.4301088198160284,
      "flos": 25484618183040.0,
      "grad_norm": 2.3370439997608825,
      "language_loss": 0.8795765,
      "learning_rate": 2.5430495858470565e-06,
      "loss": 0.90155965,
      "num_input_tokens_seen": 77101725,
      "step": 3577,
      "time_per_iteration": 2.588655710220337
    },
    {
      "auxiliary_loss_clip": 0.0116361,
      "auxiliary_loss_mlp": 0.01029515,
      "balance_loss_clip": 1.05150926,
      "balance_loss_mlp": 1.02165294,
      "epoch": 0.43022906270666744,
      "flos": 18259427404800.0,
      "grad_norm": 2.4764458153396713,
      "language_loss": 0.77410656,
      "learning_rate": 2.54229983786758e-06,
      "loss": 0.79603785,
      "num_input_tokens_seen": 77119670,
      "step": 3578,
      "time_per_iteration": 2.483489990234375
    },
    {
      "auxiliary_loss_clip": 0.0115351,
      "auxiliary_loss_mlp": 0.0103103,
      "balance_loss_clip": 1.04918385,
      "balance_loss_mlp": 1.02285779,
      "epoch": 0.43034930559730655,
      "flos": 23399536567680.0,
      "grad_norm": 1.8131931564504504,
      "language_loss": 0.85152471,
      "learning_rate": 2.541550007621651e-06,
      "loss": 0.87337005,
      "num_input_tokens_seen": 77138160,
      "step": 3579,
      "time_per_iteration": 2.5920259952545166
    },
    {
      "auxiliary_loss_clip": 0.01166389,
      "auxiliary_loss_mlp": 0.01029652,
      "balance_loss_clip": 1.05363154,
      "balance_loss_mlp": 1.02141166,
      "epoch": 0.43046954848794566,
      "flos": 28184382264960.0,
      "grad_norm": 1.875515146419343,
      "language_loss": 0.79969609,
      "learning_rate": 2.5408000952230156e-06,
      "loss": 0.82165653,
      "num_input_tokens_seen": 77156950,
      "step": 3580,
      "time_per_iteration": 2.5608794689178467
    },
    {
      "auxiliary_loss_clip": 0.01151928,
      "auxiliary_loss_mlp": 0.01027627,
      "balance_loss_clip": 1.0510962,
      "balance_loss_mlp": 1.01937735,
      "epoch": 0.4305897913785847,
      "flos": 28580476515840.0,
      "grad_norm": 1.9618394321626313,
      "language_loss": 0.90385222,
      "learning_rate": 2.5400501007854357e-06,
      "loss": 0.92564774,
      "num_input_tokens_seen": 77176395,
      "step": 3581,
      "time_per_iteration": 2.6603164672851562
    },
    {
      "auxiliary_loss_clip": 0.0112615,
      "auxiliary_loss_mlp": 0.0102984,
      "balance_loss_clip": 1.04699039,
      "balance_loss_mlp": 1.02211475,
      "epoch": 0.43071003426922383,
      "flos": 20448721353600.0,
      "grad_norm": 1.7843197964945579,
      "language_loss": 0.7529065,
      "learning_rate": 2.539300024422685e-06,
      "loss": 0.77446634,
      "num_input_tokens_seen": 77194340,
      "step": 3582,
      "time_per_iteration": 2.617633819580078
    },
    {
      "auxiliary_loss_clip": 0.01041657,
      "auxiliary_loss_mlp": 0.01004346,
      "balance_loss_clip": 1.01801682,
      "balance_loss_mlp": 1.00312376,
      "epoch": 0.43083027715986294,
      "flos": 51997969883520.0,
      "grad_norm": 0.7892402288317762,
      "language_loss": 0.60958135,
      "learning_rate": 2.538549866248549e-06,
      "loss": 0.63004136,
      "num_input_tokens_seen": 77249320,
      "step": 3583,
      "time_per_iteration": 3.0012712478637695
    },
    {
      "auxiliary_loss_clip": 0.01172655,
      "auxiliary_loss_mlp": 0.01033354,
      "balance_loss_clip": 1.05470753,
      "balance_loss_mlp": 1.02503347,
      "epoch": 0.430950520050502,
      "flos": 16690885320960.0,
      "grad_norm": 1.812193596638398,
      "language_loss": 0.81083238,
      "learning_rate": 2.5377996263768274e-06,
      "loss": 0.83289242,
      "num_input_tokens_seen": 77267400,
      "step": 3584,
      "time_per_iteration": 2.515700101852417
    },
    {
      "auxiliary_loss_clip": 0.01167175,
      "auxiliary_loss_mlp": 0.010349,
      "balance_loss_clip": 1.05244279,
      "balance_loss_mlp": 1.0268476,
      "epoch": 0.4310707629411411,
      "flos": 24608433726720.0,
      "grad_norm": 1.8304490656569599,
      "language_loss": 0.68311709,
      "learning_rate": 2.5370493049213293e-06,
      "loss": 0.70513785,
      "num_input_tokens_seen": 77287045,
      "step": 3585,
      "time_per_iteration": 2.562526226043701
    },
    {
      "auxiliary_loss_clip": 0.0107879,
      "auxiliary_loss_mlp": 0.01030949,
      "balance_loss_clip": 1.04462981,
      "balance_loss_mlp": 1.02195466,
      "epoch": 0.4311910058317802,
      "flos": 26432983019520.0,
      "grad_norm": 1.8756317518122574,
      "language_loss": 0.79639399,
      "learning_rate": 2.536298901995878e-06,
      "loss": 0.81749135,
      "num_input_tokens_seen": 77306255,
      "step": 3586,
      "time_per_iteration": 2.8930907249450684
    },
    {
      "auxiliary_loss_clip": 0.01157198,
      "auxiliary_loss_mlp": 0.01032473,
      "balance_loss_clip": 1.05341399,
      "balance_loss_mlp": 1.02437901,
      "epoch": 0.43131124872241927,
      "flos": 25155891889920.0,
      "grad_norm": 1.8679843746911116,
      "language_loss": 0.80173123,
      "learning_rate": 2.535548417714311e-06,
      "loss": 0.82362795,
      "num_input_tokens_seen": 77325555,
      "step": 3587,
      "time_per_iteration": 2.6780292987823486
    },
    {
      "auxiliary_loss_clip": 0.01171361,
      "auxiliary_loss_mlp": 0.01031251,
      "balance_loss_clip": 1.05193388,
      "balance_loss_mlp": 1.02312064,
      "epoch": 0.4314314916130584,
      "flos": 21614812479360.0,
      "grad_norm": 1.4810496776651372,
      "language_loss": 0.87379265,
      "learning_rate": 2.534797852190474e-06,
      "loss": 0.89581877,
      "num_input_tokens_seen": 77345735,
      "step": 3588,
      "time_per_iteration": 2.5390706062316895
    },
    {
      "auxiliary_loss_clip": 0.01166681,
      "auxiliary_loss_mlp": 0.01034842,
      "balance_loss_clip": 1.0520916,
      "balance_loss_mlp": 1.02681923,
      "epoch": 0.4315517345036975,
      "flos": 19275016544640.0,
      "grad_norm": 1.9612873316895616,
      "language_loss": 0.81836873,
      "learning_rate": 2.5340472055382283e-06,
      "loss": 0.84038389,
      "num_input_tokens_seen": 77361765,
      "step": 3589,
      "time_per_iteration": 2.492030382156372
    },
    {
      "auxiliary_loss_clip": 0.01140162,
      "auxiliary_loss_mlp": 0.0102693,
      "balance_loss_clip": 1.04852247,
      "balance_loss_mlp": 1.01925898,
      "epoch": 0.43167197739433655,
      "flos": 24273853516800.0,
      "grad_norm": 1.9627728519318932,
      "language_loss": 0.81093127,
      "learning_rate": 2.5332964778714468e-06,
      "loss": 0.8326022,
      "num_input_tokens_seen": 77378950,
      "step": 3590,
      "time_per_iteration": 2.6189703941345215
    },
    {
      "auxiliary_loss_clip": 0.01141884,
      "auxiliary_loss_mlp": 0.01029632,
      "balance_loss_clip": 1.05269396,
      "balance_loss_mlp": 1.02178824,
      "epoch": 0.43179222028497566,
      "flos": 16867816738560.0,
      "grad_norm": 1.525306450822741,
      "language_loss": 0.65973306,
      "learning_rate": 2.5325456693040123e-06,
      "loss": 0.68144822,
      "num_input_tokens_seen": 77396145,
      "step": 3591,
      "time_per_iteration": 2.5647053718566895
    },
    {
      "auxiliary_loss_clip": 0.01173229,
      "auxiliary_loss_mlp": 0.01035829,
      "balance_loss_clip": 1.05179477,
      "balance_loss_mlp": 1.02696586,
      "epoch": 0.43191246317561477,
      "flos": 17639214243840.0,
      "grad_norm": 2.0520741408021044,
      "language_loss": 0.74564111,
      "learning_rate": 2.531794779949824e-06,
      "loss": 0.76773167,
      "num_input_tokens_seen": 77414045,
      "step": 3592,
      "time_per_iteration": 2.5061001777648926
    },
    {
      "auxiliary_loss_clip": 0.01135387,
      "auxiliary_loss_mlp": 0.01027681,
      "balance_loss_clip": 1.04943776,
      "balance_loss_mlp": 1.02038002,
      "epoch": 0.4320327060662538,
      "flos": 23878800760320.0,
      "grad_norm": 1.7032242848205428,
      "language_loss": 0.88127005,
      "learning_rate": 2.5310438099227903e-06,
      "loss": 0.90290082,
      "num_input_tokens_seen": 77431310,
      "step": 3593,
      "time_per_iteration": 2.605987310409546
    },
    {
      "auxiliary_loss_clip": 0.01064514,
      "auxiliary_loss_mlp": 0.0100115,
      "balance_loss_clip": 1.018435,
      "balance_loss_mlp": 0.99995786,
      "epoch": 0.43215294895689293,
      "flos": 66394917959040.0,
      "grad_norm": 0.8049650495500036,
      "language_loss": 0.53329861,
      "learning_rate": 2.530292759336833e-06,
      "loss": 0.5539552,
      "num_input_tokens_seen": 77492045,
      "step": 3594,
      "time_per_iteration": 3.178866147994995
    },
    {
      "auxiliary_loss_clip": 0.01151926,
      "auxiliary_loss_mlp": 0.01030004,
      "balance_loss_clip": 1.05294347,
      "balance_loss_mlp": 1.02136707,
      "epoch": 0.432273191847532,
      "flos": 20594267262720.0,
      "grad_norm": 2.1885519149514434,
      "language_loss": 0.69326806,
      "learning_rate": 2.5295416283058855e-06,
      "loss": 0.71508729,
      "num_input_tokens_seen": 77510910,
      "step": 3595,
      "time_per_iteration": 2.5556068420410156
    },
    {
      "auxiliary_loss_clip": 0.01151253,
      "auxiliary_loss_mlp": 0.00761783,
      "balance_loss_clip": 1.05060267,
      "balance_loss_mlp": 1.00018835,
      "epoch": 0.4323934347381711,
      "flos": 19282127437440.0,
      "grad_norm": 1.9226800306517742,
      "language_loss": 0.66127521,
      "learning_rate": 2.5287904169438943e-06,
      "loss": 0.68040562,
      "num_input_tokens_seen": 77530115,
      "step": 3596,
      "time_per_iteration": 3.549539089202881
    },
    {
      "auxiliary_loss_clip": 0.01107738,
      "auxiliary_loss_mlp": 0.01037109,
      "balance_loss_clip": 1.04779124,
      "balance_loss_mlp": 1.02835965,
      "epoch": 0.4325136776288102,
      "flos": 21726315273600.0,
      "grad_norm": 2.6074660072986,
      "language_loss": 0.63858062,
      "learning_rate": 2.528039125364817e-06,
      "loss": 0.66002911,
      "num_input_tokens_seen": 77548920,
      "step": 3597,
      "time_per_iteration": 2.7257888317108154
    },
    {
      "auxiliary_loss_clip": 0.01147032,
      "auxiliary_loss_mlp": 0.01032704,
      "balance_loss_clip": 1.05263627,
      "balance_loss_mlp": 1.02438879,
      "epoch": 0.43263392051944927,
      "flos": 22340746344960.0,
      "grad_norm": 1.9103278348464414,
      "language_loss": 0.75973475,
      "learning_rate": 2.5272877536826246e-06,
      "loss": 0.78153211,
      "num_input_tokens_seen": 77567715,
      "step": 3598,
      "time_per_iteration": 2.59759259223938
    },
    {
      "auxiliary_loss_clip": 0.01126038,
      "auxiliary_loss_mlp": 0.01034591,
      "balance_loss_clip": 1.04352677,
      "balance_loss_mlp": 1.02624011,
      "epoch": 0.4327541634100884,
      "flos": 29168406328320.0,
      "grad_norm": 2.349886209214647,
      "language_loss": 0.70451754,
      "learning_rate": 2.5265363020112986e-06,
      "loss": 0.72612387,
      "num_input_tokens_seen": 77588035,
      "step": 3599,
      "time_per_iteration": 2.701719284057617
    },
    {
      "auxiliary_loss_clip": 0.01169244,
      "auxiliary_loss_mlp": 0.01032303,
      "balance_loss_clip": 1.05394244,
      "balance_loss_mlp": 1.02419066,
      "epoch": 0.4328744063007275,
      "flos": 26067448264320.0,
      "grad_norm": 1.994671950463894,
      "language_loss": 0.8436203,
      "learning_rate": 2.5257847704648344e-06,
      "loss": 0.86563575,
      "num_input_tokens_seen": 77609265,
      "step": 3600,
      "time_per_iteration": 2.57527232170105
    },
    {
      "auxiliary_loss_clip": 0.01183263,
      "auxiliary_loss_mlp": 0.01035906,
      "balance_loss_clip": 1.05555034,
      "balance_loss_mlp": 1.02815115,
      "epoch": 0.43299464919136654,
      "flos": 16581357774720.0,
      "grad_norm": 1.806419277832725,
      "language_loss": 0.75297475,
      "learning_rate": 2.525033159157239e-06,
      "loss": 0.77516645,
      "num_input_tokens_seen": 77625580,
      "step": 3601,
      "time_per_iteration": 4.024463653564453
    },
    {
      "auxiliary_loss_clip": 0.01166197,
      "auxiliary_loss_mlp": 0.01030242,
      "balance_loss_clip": 1.05272865,
      "balance_loss_mlp": 1.02108049,
      "epoch": 0.43311489208200565,
      "flos": 16107265140480.0,
      "grad_norm": 1.790854493371774,
      "language_loss": 0.77050787,
      "learning_rate": 2.52428146820253e-06,
      "loss": 0.79247224,
      "num_input_tokens_seen": 77643835,
      "step": 3602,
      "time_per_iteration": 2.5137691497802734
    },
    {
      "auxiliary_loss_clip": 0.01143894,
      "auxiliary_loss_mlp": 0.01029632,
      "balance_loss_clip": 1.05102789,
      "balance_loss_mlp": 1.02060235,
      "epoch": 0.43323513497264476,
      "flos": 22930220442240.0,
      "grad_norm": 1.668592461694098,
      "language_loss": 0.81774116,
      "learning_rate": 2.52352969771474e-06,
      "loss": 0.83947641,
      "num_input_tokens_seen": 77663060,
      "step": 3603,
      "time_per_iteration": 2.6179873943328857
    },
    {
      "auxiliary_loss_clip": 0.01157422,
      "auxiliary_loss_mlp": 0.01030938,
      "balance_loss_clip": 1.05284572,
      "balance_loss_mlp": 1.02324045,
      "epoch": 0.4333553778632838,
      "flos": 25299031587840.0,
      "grad_norm": 2.1331858928940233,
      "language_loss": 0.88414061,
      "learning_rate": 2.5227778478079106e-06,
      "loss": 0.90602422,
      "num_input_tokens_seen": 77682470,
      "step": 3604,
      "time_per_iteration": 2.5929932594299316
    },
    {
      "auxiliary_loss_clip": 0.01164585,
      "auxiliary_loss_mlp": 0.01035873,
      "balance_loss_clip": 1.05008209,
      "balance_loss_mlp": 1.02765942,
      "epoch": 0.43347562075392293,
      "flos": 19387165783680.0,
      "grad_norm": 1.5820467580025375,
      "language_loss": 0.76470697,
      "learning_rate": 2.522025918596098e-06,
      "loss": 0.78671157,
      "num_input_tokens_seen": 77700770,
      "step": 3605,
      "time_per_iteration": 2.514413356781006
    },
    {
      "auxiliary_loss_clip": 0.01167822,
      "auxiliary_loss_mlp": 0.01030205,
      "balance_loss_clip": 1.05369961,
      "balance_loss_mlp": 1.02299237,
      "epoch": 0.43359586364456204,
      "flos": 26325969425280.0,
      "grad_norm": 1.929499599705031,
      "language_loss": 0.65473723,
      "learning_rate": 2.521273910193368e-06,
      "loss": 0.67671752,
      "num_input_tokens_seen": 77723950,
      "step": 3606,
      "time_per_iteration": 2.604297399520874
    },
    {
      "auxiliary_loss_clip": 0.01174541,
      "auxiliary_loss_mlp": 0.01029729,
      "balance_loss_clip": 1.05425644,
      "balance_loss_mlp": 1.02130103,
      "epoch": 0.4337161065352011,
      "flos": 15989261984640.0,
      "grad_norm": 2.0825110612524864,
      "language_loss": 0.87266529,
      "learning_rate": 2.5205218227138006e-06,
      "loss": 0.89470804,
      "num_input_tokens_seen": 77736905,
      "step": 3607,
      "time_per_iteration": 2.4767937660217285
    },
    {
      "auxiliary_loss_clip": 0.01185633,
      "auxiliary_loss_mlp": 0.01029098,
      "balance_loss_clip": 1.05561185,
      "balance_loss_mlp": 1.02096224,
      "epoch": 0.4338363494258402,
      "flos": 20224710184320.0,
      "grad_norm": 1.8489097331950264,
      "language_loss": 0.79140967,
      "learning_rate": 2.519769656271486e-06,
      "loss": 0.81355697,
      "num_input_tokens_seen": 77754325,
      "step": 3608,
      "time_per_iteration": 2.480135440826416
    },
    {
      "auxiliary_loss_clip": 0.01116881,
      "auxiliary_loss_mlp": 0.01030726,
      "balance_loss_clip": 1.04739618,
      "balance_loss_mlp": 1.02241731,
      "epoch": 0.43395659231647926,
      "flos": 20083904870400.0,
      "grad_norm": 2.306512799439487,
      "language_loss": 0.67602122,
      "learning_rate": 2.5190174109805285e-06,
      "loss": 0.69749731,
      "num_input_tokens_seen": 77774150,
      "step": 3609,
      "time_per_iteration": 2.5902862548828125
    },
    {
      "auxiliary_loss_clip": 0.01143692,
      "auxiliary_loss_mlp": 0.01027105,
      "balance_loss_clip": 1.04842615,
      "balance_loss_mlp": 1.01852834,
      "epoch": 0.43407683520711837,
      "flos": 19901801894400.0,
      "grad_norm": 2.3156158105556863,
      "language_loss": 0.64202988,
      "learning_rate": 2.518265086955042e-06,
      "loss": 0.66373789,
      "num_input_tokens_seen": 77791870,
      "step": 3610,
      "time_per_iteration": 2.5484066009521484
    },
    {
      "auxiliary_loss_clip": 0.01185414,
      "auxiliary_loss_mlp": 0.0103434,
      "balance_loss_clip": 1.05539536,
      "balance_loss_mlp": 1.02628696,
      "epoch": 0.4341970780977575,
      "flos": 23108732058240.0,
      "grad_norm": 1.7525056545646718,
      "language_loss": 0.8388952,
      "learning_rate": 2.5175126843091534e-06,
      "loss": 0.86109269,
      "num_input_tokens_seen": 77811240,
      "step": 3611,
      "time_per_iteration": 2.494366407394409
    },
    {
      "auxiliary_loss_clip": 0.01157506,
      "auxiliary_loss_mlp": 0.01038192,
      "balance_loss_clip": 1.05403209,
      "balance_loss_mlp": 1.03056228,
      "epoch": 0.43431732098839654,
      "flos": 37408288406400.0,
      "grad_norm": 2.226265724524801,
      "language_loss": 0.75613588,
      "learning_rate": 2.5167602031570034e-06,
      "loss": 0.77809286,
      "num_input_tokens_seen": 77831425,
      "step": 3612,
      "time_per_iteration": 2.684640884399414
    },
    {
      "auxiliary_loss_clip": 0.01185882,
      "auxiliary_loss_mlp": 0.01025872,
      "balance_loss_clip": 1.05549526,
      "balance_loss_mlp": 1.01779008,
      "epoch": 0.43443756387903565,
      "flos": 31868206323840.0,
      "grad_norm": 1.6303638061974248,
      "language_loss": 0.73214149,
      "learning_rate": 2.51600764361274e-06,
      "loss": 0.75425899,
      "num_input_tokens_seen": 77852950,
      "step": 3613,
      "time_per_iteration": 2.572516441345215
    },
    {
      "auxiliary_loss_clip": 0.01183424,
      "auxiliary_loss_mlp": 0.01031471,
      "balance_loss_clip": 1.05476022,
      "balance_loss_mlp": 1.02309704,
      "epoch": 0.43455780676967476,
      "flos": 23477139901440.0,
      "grad_norm": 2.1166228524089754,
      "language_loss": 0.79079098,
      "learning_rate": 2.5152550057905283e-06,
      "loss": 0.81293988,
      "num_input_tokens_seen": 77872840,
      "step": 3614,
      "time_per_iteration": 2.5285396575927734
    },
    {
      "auxiliary_loss_clip": 0.01171933,
      "auxiliary_loss_mlp": 0.00762401,
      "balance_loss_clip": 1.05482078,
      "balance_loss_mlp": 1.00016975,
      "epoch": 0.4346780496603138,
      "flos": 24207060176640.0,
      "grad_norm": 2.3049703736160856,
      "language_loss": 0.7681092,
      "learning_rate": 2.5145022898045415e-06,
      "loss": 0.78745258,
      "num_input_tokens_seen": 77892025,
      "step": 3615,
      "time_per_iteration": 2.5454652309417725
    },
    {
      "auxiliary_loss_clip": 0.0115582,
      "auxiliary_loss_mlp": 0.01028683,
      "balance_loss_clip": 1.04925311,
      "balance_loss_mlp": 1.02011788,
      "epoch": 0.4347982925509529,
      "flos": 17092366611840.0,
      "grad_norm": 1.9958052453647352,
      "language_loss": 0.89839894,
      "learning_rate": 2.5137494957689664e-06,
      "loss": 0.92024398,
      "num_input_tokens_seen": 77907635,
      "step": 3616,
      "time_per_iteration": 2.5301103591918945
    },
    {
      "auxiliary_loss_clip": 0.01051256,
      "auxiliary_loss_mlp": 0.01001832,
      "balance_loss_clip": 1.01752114,
      "balance_loss_mlp": 1.00046754,
      "epoch": 0.43491853544159204,
      "flos": 60945544696320.0,
      "grad_norm": 0.7668876585836026,
      "language_loss": 0.57349288,
      "learning_rate": 2.5129966237980016e-06,
      "loss": 0.59402382,
      "num_input_tokens_seen": 77970630,
      "step": 3617,
      "time_per_iteration": 3.1194331645965576
    },
    {
      "auxiliary_loss_clip": 0.01141451,
      "auxiliary_loss_mlp": 0.0103259,
      "balance_loss_clip": 1.04885209,
      "balance_loss_mlp": 1.02387547,
      "epoch": 0.4350387783322311,
      "flos": 21944652094080.0,
      "grad_norm": 1.8317883712192207,
      "language_loss": 0.78330994,
      "learning_rate": 2.512243674005857e-06,
      "loss": 0.80505037,
      "num_input_tokens_seen": 77989995,
      "step": 3618,
      "time_per_iteration": 2.6191089153289795
    },
    {
      "auxiliary_loss_clip": 0.01108589,
      "auxiliary_loss_mlp": 0.01025951,
      "balance_loss_clip": 1.04410207,
      "balance_loss_mlp": 1.01796746,
      "epoch": 0.4351590212228702,
      "flos": 25082705928960.0,
      "grad_norm": 1.7916766536241555,
      "language_loss": 0.86246312,
      "learning_rate": 2.5114906465067537e-06,
      "loss": 0.88380849,
      "num_input_tokens_seen": 78010980,
      "step": 3619,
      "time_per_iteration": 2.6998515129089355
    },
    {
      "auxiliary_loss_clip": 0.01168976,
      "auxiliary_loss_mlp": 0.01023736,
      "balance_loss_clip": 1.05112267,
      "balance_loss_mlp": 1.01599979,
      "epoch": 0.4352792641135093,
      "flos": 21506541909120.0,
      "grad_norm": 2.2375425216032756,
      "language_loss": 0.75143516,
      "learning_rate": 2.5107375414149264e-06,
      "loss": 0.77336228,
      "num_input_tokens_seen": 78030225,
      "step": 3620,
      "time_per_iteration": 2.562272787094116
    },
    {
      "auxiliary_loss_clip": 0.01116137,
      "auxiliary_loss_mlp": 0.01035214,
      "balance_loss_clip": 1.04306424,
      "balance_loss_mlp": 1.02655983,
      "epoch": 0.43539950700414837,
      "flos": 16253457494400.0,
      "grad_norm": 1.9807502836476336,
      "language_loss": 0.72257829,
      "learning_rate": 2.5099843588446197e-06,
      "loss": 0.74409175,
      "num_input_tokens_seen": 78048545,
      "step": 3621,
      "time_per_iteration": 2.586515188217163
    },
    {
      "auxiliary_loss_clip": 0.01130718,
      "auxiliary_loss_mlp": 0.01029607,
      "balance_loss_clip": 1.04946136,
      "balance_loss_mlp": 1.02126276,
      "epoch": 0.4355197498947875,
      "flos": 16691819074560.0,
      "grad_norm": 2.0325284165617346,
      "language_loss": 0.61170185,
      "learning_rate": 2.509231098910091e-06,
      "loss": 0.63330507,
      "num_input_tokens_seen": 78068415,
      "step": 3622,
      "time_per_iteration": 3.436128616333008
    },
    {
      "auxiliary_loss_clip": 0.01152348,
      "auxiliary_loss_mlp": 0.01031168,
      "balance_loss_clip": 1.05368948,
      "balance_loss_mlp": 1.02314496,
      "epoch": 0.4356399927854266,
      "flos": 16362733645440.0,
      "grad_norm": 2.4454130761593844,
      "language_loss": 0.74820101,
      "learning_rate": 2.508477761725611e-06,
      "loss": 0.77003616,
      "num_input_tokens_seen": 78086690,
      "step": 3623,
      "time_per_iteration": 2.542694568634033
    },
    {
      "auxiliary_loss_clip": 0.01171937,
      "auxiliary_loss_mlp": 0.01029426,
      "balance_loss_clip": 1.05187571,
      "balance_loss_mlp": 1.02155805,
      "epoch": 0.43576023567606564,
      "flos": 17202037812480.0,
      "grad_norm": 2.918574315264654,
      "language_loss": 0.80943024,
      "learning_rate": 2.507724347405458e-06,
      "loss": 0.83144391,
      "num_input_tokens_seen": 78104640,
      "step": 3624,
      "time_per_iteration": 2.5061440467834473
    },
    {
      "auxiliary_loss_clip": 0.01121003,
      "auxiliary_loss_mlp": 0.01028,
      "balance_loss_clip": 1.04417205,
      "balance_loss_mlp": 1.02015591,
      "epoch": 0.43588047856670475,
      "flos": 15917656222080.0,
      "grad_norm": 2.0023700866876752,
      "language_loss": 0.82144189,
      "learning_rate": 2.5069708560639243e-06,
      "loss": 0.84293193,
      "num_input_tokens_seen": 78122550,
      "step": 3625,
      "time_per_iteration": 2.6119801998138428
    },
    {
      "auxiliary_loss_clip": 0.01144345,
      "auxiliary_loss_mlp": 0.01032257,
      "balance_loss_clip": 1.05092967,
      "balance_loss_mlp": 1.02371585,
      "epoch": 0.4360007214573438,
      "flos": 23659566099840.0,
      "grad_norm": 1.8153053805938022,
      "language_loss": 0.61237109,
      "learning_rate": 2.5062172878153158e-06,
      "loss": 0.63413709,
      "num_input_tokens_seen": 78141825,
      "step": 3626,
      "time_per_iteration": 2.6256260871887207
    },
    {
      "auxiliary_loss_clip": 0.01118638,
      "auxiliary_loss_mlp": 0.01032832,
      "balance_loss_clip": 1.04738164,
      "balance_loss_mlp": 1.02430868,
      "epoch": 0.4361209643479829,
      "flos": 21978767036160.0,
      "grad_norm": 2.019967294684137,
      "language_loss": 0.87470901,
      "learning_rate": 2.505463642773947e-06,
      "loss": 0.89622366,
      "num_input_tokens_seen": 78161790,
      "step": 3627,
      "time_per_iteration": 4.288119554519653
    },
    {
      "auxiliary_loss_clip": 0.01141358,
      "auxiliary_loss_mlp": 0.0076226,
      "balance_loss_clip": 1.04898822,
      "balance_loss_mlp": 1.00013936,
      "epoch": 0.43624120723862203,
      "flos": 17420159151360.0,
      "grad_norm": 2.2194486742410326,
      "language_loss": 0.74859357,
      "learning_rate": 2.504709921054146e-06,
      "loss": 0.76762974,
      "num_input_tokens_seen": 78178605,
      "step": 3628,
      "time_per_iteration": 2.5614426136016846
    },
    {
      "auxiliary_loss_clip": 0.01133497,
      "auxiliary_loss_mlp": 0.01029387,
      "balance_loss_clip": 1.04402483,
      "balance_loss_mlp": 1.02102447,
      "epoch": 0.4363614501292611,
      "flos": 17895293280000.0,
      "grad_norm": 2.1031936585057567,
      "language_loss": 0.83388603,
      "learning_rate": 2.50395612277025e-06,
      "loss": 0.85551488,
      "num_input_tokens_seen": 78194460,
      "step": 3629,
      "time_per_iteration": 2.5531394481658936
    },
    {
      "auxiliary_loss_clip": 0.01159725,
      "auxiliary_loss_mlp": 0.01030575,
      "balance_loss_clip": 1.05144787,
      "balance_loss_mlp": 1.02270126,
      "epoch": 0.4364816930199002,
      "flos": 20302888135680.0,
      "grad_norm": 2.30116793901757,
      "language_loss": 0.72930813,
      "learning_rate": 2.503202248036612e-06,
      "loss": 0.75121111,
      "num_input_tokens_seen": 78213315,
      "step": 3630,
      "time_per_iteration": 2.550079345703125
    },
    {
      "auxiliary_loss_clip": 0.01181487,
      "auxiliary_loss_mlp": 0.01034671,
      "balance_loss_clip": 1.054497,
      "balance_loss_mlp": 1.02668989,
      "epoch": 0.4366019359105393,
      "flos": 24061334699520.0,
      "grad_norm": 1.7282447196458484,
      "language_loss": 0.73279142,
      "learning_rate": 2.5024482969675927e-06,
      "loss": 0.75495297,
      "num_input_tokens_seen": 78233270,
      "step": 3631,
      "time_per_iteration": 2.5163164138793945
    },
    {
      "auxiliary_loss_clip": 0.01128967,
      "auxiliary_loss_mlp": 0.01030096,
      "balance_loss_clip": 1.04657793,
      "balance_loss_mlp": 1.02258611,
      "epoch": 0.43672217880117836,
      "flos": 21754109422080.0,
      "grad_norm": 2.386931888186135,
      "language_loss": 0.84378922,
      "learning_rate": 2.501694269677566e-06,
      "loss": 0.86537981,
      "num_input_tokens_seen": 78251040,
      "step": 3632,
      "time_per_iteration": 2.667170763015747
    },
    {
      "auxiliary_loss_clip": 0.01171047,
      "auxiliary_loss_mlp": 0.01029438,
      "balance_loss_clip": 1.05074692,
      "balance_loss_mlp": 1.02109885,
      "epoch": 0.4368424216918175,
      "flos": 18035200753920.0,
      "grad_norm": 1.8924003049118214,
      "language_loss": 0.80559939,
      "learning_rate": 2.500940166280918e-06,
      "loss": 0.82760423,
      "num_input_tokens_seen": 78269470,
      "step": 3633,
      "time_per_iteration": 2.5299673080444336
    },
    {
      "auxiliary_loss_clip": 0.01167241,
      "auxiliary_loss_mlp": 0.01026286,
      "balance_loss_clip": 1.05265713,
      "balance_loss_mlp": 1.01863265,
      "epoch": 0.4369626645824566,
      "flos": 25447127362560.0,
      "grad_norm": 1.8993859611156458,
      "language_loss": 0.79189652,
      "learning_rate": 2.500185986892045e-06,
      "loss": 0.81383181,
      "num_input_tokens_seen": 78288955,
      "step": 3634,
      "time_per_iteration": 2.5598156452178955
    },
    {
      "auxiliary_loss_clip": 0.01162371,
      "auxiliary_loss_mlp": 0.01025076,
      "balance_loss_clip": 1.05010462,
      "balance_loss_mlp": 1.01711857,
      "epoch": 0.43708290747309564,
      "flos": 25302694775040.0,
      "grad_norm": 1.8616549633305064,
      "language_loss": 0.77005261,
      "learning_rate": 2.499431731625355e-06,
      "loss": 0.7919271,
      "num_input_tokens_seen": 78307980,
      "step": 3635,
      "time_per_iteration": 2.5477821826934814
    },
    {
      "auxiliary_loss_clip": 0.01183751,
      "auxiliary_loss_mlp": 0.01033728,
      "balance_loss_clip": 1.05366182,
      "balance_loss_mlp": 1.02496016,
      "epoch": 0.43720315036373475,
      "flos": 31575103344000.0,
      "grad_norm": 1.8146480193967212,
      "language_loss": 0.79521209,
      "learning_rate": 2.4986774005952686e-06,
      "loss": 0.81738687,
      "num_input_tokens_seen": 78330355,
      "step": 3636,
      "time_per_iteration": 2.5874812602996826
    },
    {
      "auxiliary_loss_clip": 0.01163634,
      "auxiliary_loss_mlp": 0.01027784,
      "balance_loss_clip": 1.05189586,
      "balance_loss_mlp": 1.02016091,
      "epoch": 0.43732339325437386,
      "flos": 23112000195840.0,
      "grad_norm": 2.2318340942458126,
      "language_loss": 0.84761918,
      "learning_rate": 2.4979229939162166e-06,
      "loss": 0.86953342,
      "num_input_tokens_seen": 78349135,
      "step": 3637,
      "time_per_iteration": 2.5399155616760254
    },
    {
      "auxiliary_loss_clip": 0.01167319,
      "auxiliary_loss_mlp": 0.01034106,
      "balance_loss_clip": 1.05420494,
      "balance_loss_mlp": 1.0258863,
      "epoch": 0.4374436361450129,
      "flos": 27746272080000.0,
      "grad_norm": 1.80224567131789,
      "language_loss": 0.80507398,
      "learning_rate": 2.4971685117026433e-06,
      "loss": 0.82708824,
      "num_input_tokens_seen": 78368900,
      "step": 3638,
      "time_per_iteration": 2.5968129634857178
    },
    {
      "auxiliary_loss_clip": 0.01169031,
      "auxiliary_loss_mlp": 0.01023706,
      "balance_loss_clip": 1.05221069,
      "balance_loss_mlp": 1.01613617,
      "epoch": 0.437563879035652,
      "flos": 24172370616960.0,
      "grad_norm": 1.5238071572154641,
      "language_loss": 0.76567465,
      "learning_rate": 2.4964139540690018e-06,
      "loss": 0.78760201,
      "num_input_tokens_seen": 78392235,
      "step": 3639,
      "time_per_iteration": 2.555224657058716
    },
    {
      "auxiliary_loss_clip": 0.01137893,
      "auxiliary_loss_mlp": 0.01029401,
      "balance_loss_clip": 1.04836655,
      "balance_loss_mlp": 1.02029967,
      "epoch": 0.4376841219262911,
      "flos": 23477211728640.0,
      "grad_norm": 2.351266796541749,
      "language_loss": 0.72711015,
      "learning_rate": 2.495659321129758e-06,
      "loss": 0.74878311,
      "num_input_tokens_seen": 78409980,
      "step": 3640,
      "time_per_iteration": 2.604569435119629
    },
    {
      "auxiliary_loss_clip": 0.01164136,
      "auxiliary_loss_mlp": 0.0103061,
      "balance_loss_clip": 1.05023217,
      "balance_loss_mlp": 1.02301598,
      "epoch": 0.4378043648169302,
      "flos": 25447809720960.0,
      "grad_norm": 1.7473342370396368,
      "language_loss": 0.7542429,
      "learning_rate": 2.494904612999389e-06,
      "loss": 0.77619028,
      "num_input_tokens_seen": 78428690,
      "step": 3641,
      "time_per_iteration": 2.548095941543579
    },
    {
      "auxiliary_loss_clip": 0.01062955,
      "auxiliary_loss_mlp": 0.01004581,
      "balance_loss_clip": 1.01967096,
      "balance_loss_mlp": 1.00324535,
      "epoch": 0.4379246077075693,
      "flos": 53914056986880.0,
      "grad_norm": 0.7483243929496436,
      "language_loss": 0.5658747,
      "learning_rate": 2.4941498297923843e-06,
      "loss": 0.58655,
      "num_input_tokens_seen": 78489260,
      "step": 3642,
      "time_per_iteration": 3.0750815868377686
    },
    {
      "auxiliary_loss_clip": 0.01165031,
      "auxiliary_loss_mlp": 0.01028984,
      "balance_loss_clip": 1.05206823,
      "balance_loss_mlp": 1.02067471,
      "epoch": 0.43804485059820836,
      "flos": 20588305605120.0,
      "grad_norm": 1.715225972078742,
      "language_loss": 0.69634312,
      "learning_rate": 2.4933949716232424e-06,
      "loss": 0.7182833,
      "num_input_tokens_seen": 78506785,
      "step": 3643,
      "time_per_iteration": 2.506934642791748
    },
    {
      "auxiliary_loss_clip": 0.01137145,
      "auxiliary_loss_mlp": 0.0102621,
      "balance_loss_clip": 1.05043471,
      "balance_loss_mlp": 1.01831222,
      "epoch": 0.43816509348884747,
      "flos": 23876214981120.0,
      "grad_norm": 1.9858395287895985,
      "language_loss": 0.7346769,
      "learning_rate": 2.492640038606476e-06,
      "loss": 0.75631046,
      "num_input_tokens_seen": 78525150,
      "step": 3644,
      "time_per_iteration": 2.5861377716064453
    },
    {
      "auxiliary_loss_clip": 0.01165805,
      "auxiliary_loss_mlp": 0.010327,
      "balance_loss_clip": 1.04979563,
      "balance_loss_mlp": 1.02446246,
      "epoch": 0.4382853363794866,
      "flos": 14684448533760.0,
      "grad_norm": 1.9959624198261936,
      "language_loss": 0.78891188,
      "learning_rate": 2.491885030856608e-06,
      "loss": 0.81089693,
      "num_input_tokens_seen": 78543245,
      "step": 3645,
      "time_per_iteration": 2.4959805011749268
    },
    {
      "auxiliary_loss_clip": 0.01155053,
      "auxiliary_loss_mlp": 0.01031689,
      "balance_loss_clip": 1.05147398,
      "balance_loss_mlp": 1.02397609,
      "epoch": 0.43840557927012563,
      "flos": 17165301177600.0,
      "grad_norm": 2.00398963515423,
      "language_loss": 0.82627708,
      "learning_rate": 2.4911299484881713e-06,
      "loss": 0.84814453,
      "num_input_tokens_seen": 78560775,
      "step": 3646,
      "time_per_iteration": 2.534351348876953
    },
    {
      "auxiliary_loss_clip": 0.01148707,
      "auxiliary_loss_mlp": 0.0102957,
      "balance_loss_clip": 1.04922676,
      "balance_loss_mlp": 1.02190495,
      "epoch": 0.43852582216076474,
      "flos": 19390685316480.0,
      "grad_norm": 1.6537787792669818,
      "language_loss": 0.80927181,
      "learning_rate": 2.490374791615712e-06,
      "loss": 0.83105457,
      "num_input_tokens_seen": 78580800,
      "step": 3647,
      "time_per_iteration": 2.5626070499420166
    },
    {
      "auxiliary_loss_clip": 0.01185538,
      "auxiliary_loss_mlp": 0.00762872,
      "balance_loss_clip": 1.05554664,
      "balance_loss_mlp": 1.00014544,
      "epoch": 0.43864606505140386,
      "flos": 18075133699200.0,
      "grad_norm": 2.4399299879048084,
      "language_loss": 0.77647495,
      "learning_rate": 2.4896195603537867e-06,
      "loss": 0.79595912,
      "num_input_tokens_seen": 78595410,
      "step": 3648,
      "time_per_iteration": 3.260354518890381
    },
    {
      "auxiliary_loss_clip": 0.01118852,
      "auxiliary_loss_mlp": 0.01035343,
      "balance_loss_clip": 1.05047047,
      "balance_loss_mlp": 1.02749896,
      "epoch": 0.4387663079420429,
      "flos": 19644896845440.0,
      "grad_norm": 2.236622160298854,
      "language_loss": 0.74193943,
      "learning_rate": 2.488864254816964e-06,
      "loss": 0.76348144,
      "num_input_tokens_seen": 78614100,
      "step": 3649,
      "time_per_iteration": 2.5674123764038086
    },
    {
      "auxiliary_loss_clip": 0.01171034,
      "auxiliary_loss_mlp": 0.01033729,
      "balance_loss_clip": 1.05424976,
      "balance_loss_mlp": 1.0252353,
      "epoch": 0.438886550832682,
      "flos": 19719339782400.0,
      "grad_norm": 2.6446003621770933,
      "language_loss": 0.6790961,
      "learning_rate": 2.4881088751198218e-06,
      "loss": 0.70114374,
      "num_input_tokens_seen": 78632260,
      "step": 3650,
      "time_per_iteration": 2.515728235244751
    },
    {
      "auxiliary_loss_clip": 0.01151969,
      "auxiliary_loss_mlp": 0.01030032,
      "balance_loss_clip": 1.04853034,
      "balance_loss_mlp": 1.02187204,
      "epoch": 0.43900679372332113,
      "flos": 14536675981440.0,
      "grad_norm": 2.5067159553675973,
      "language_loss": 0.64679307,
      "learning_rate": 2.4873534213769517e-06,
      "loss": 0.66861308,
      "num_input_tokens_seen": 78647490,
      "step": 3651,
      "time_per_iteration": 2.567798614501953
    },
    {
      "auxiliary_loss_clip": 0.0113675,
      "auxiliary_loss_mlp": 0.0102906,
      "balance_loss_clip": 1.05186558,
      "balance_loss_mlp": 1.02141905,
      "epoch": 0.4391270366139602,
      "flos": 24056234968320.0,
      "grad_norm": 1.8703564818113394,
      "language_loss": 0.71792251,
      "learning_rate": 2.4865978937029547e-06,
      "loss": 0.73958063,
      "num_input_tokens_seen": 78666470,
      "step": 3652,
      "time_per_iteration": 2.6023499965667725
    },
    {
      "auxiliary_loss_clip": 0.01112626,
      "auxiliary_loss_mlp": 0.01029055,
      "balance_loss_clip": 1.04566455,
      "balance_loss_mlp": 1.02150917,
      "epoch": 0.4392472795045993,
      "flos": 31538510363520.0,
      "grad_norm": 1.5552366077013986,
      "language_loss": 0.66296995,
      "learning_rate": 2.485842292212445e-06,
      "loss": 0.68438679,
      "num_input_tokens_seen": 78687685,
      "step": 3653,
      "time_per_iteration": 3.6000313758850098
    },
    {
      "auxiliary_loss_clip": 0.01184159,
      "auxiliary_loss_mlp": 0.01032092,
      "balance_loss_clip": 1.05536664,
      "balance_loss_mlp": 1.02425408,
      "epoch": 0.4393675223952384,
      "flos": 14866300114560.0,
      "grad_norm": 2.4287720431926347,
      "language_loss": 0.80511844,
      "learning_rate": 2.485086617020045e-06,
      "loss": 0.827281,
      "num_input_tokens_seen": 78706180,
      "step": 3654,
      "time_per_iteration": 3.246129035949707
    },
    {
      "auxiliary_loss_clip": 0.01145312,
      "auxiliary_loss_mlp": 0.01025921,
      "balance_loss_clip": 1.04799676,
      "balance_loss_mlp": 1.01786232,
      "epoch": 0.43948776528587746,
      "flos": 14825900292480.0,
      "grad_norm": 1.9764129928488916,
      "language_loss": 0.81934607,
      "learning_rate": 2.4843308682403903e-06,
      "loss": 0.84105837,
      "num_input_tokens_seen": 78723095,
      "step": 3655,
      "time_per_iteration": 2.5311496257781982
    },
    {
      "auxiliary_loss_clip": 0.01180921,
      "auxiliary_loss_mlp": 0.01031697,
      "balance_loss_clip": 1.05215883,
      "balance_loss_mlp": 1.02418709,
      "epoch": 0.4396080081765166,
      "flos": 13914523486080.0,
      "grad_norm": 1.597471782173635,
      "language_loss": 0.82419157,
      "learning_rate": 2.4835750459881294e-06,
      "loss": 0.84631771,
      "num_input_tokens_seen": 78739720,
      "step": 3656,
      "time_per_iteration": 2.4628803730010986
    },
    {
      "auxiliary_loss_clip": 0.01147614,
      "auxiliary_loss_mlp": 0.01035227,
      "balance_loss_clip": 1.04930472,
      "balance_loss_mlp": 1.02612531,
      "epoch": 0.43972825106715563,
      "flos": 18222978078720.0,
      "grad_norm": 1.941843068724885,
      "language_loss": 0.81867075,
      "learning_rate": 2.4828191503779177e-06,
      "loss": 0.84049916,
      "num_input_tokens_seen": 78757820,
      "step": 3657,
      "time_per_iteration": 2.5373754501342773
    },
    {
      "auxiliary_loss_clip": 0.01136334,
      "auxiliary_loss_mlp": 0.01030813,
      "balance_loss_clip": 1.04775023,
      "balance_loss_mlp": 1.02180719,
      "epoch": 0.43984849395779474,
      "flos": 16873239692160.0,
      "grad_norm": 1.8515180652248469,
      "language_loss": 0.89761603,
      "learning_rate": 2.482063181524425e-06,
      "loss": 0.91928756,
      "num_input_tokens_seen": 78773720,
      "step": 3658,
      "time_per_iteration": 2.533421754837036
    },
    {
      "auxiliary_loss_clip": 0.01185472,
      "auxiliary_loss_mlp": 0.01036014,
      "balance_loss_clip": 1.0574007,
      "balance_loss_mlp": 1.02663803,
      "epoch": 0.43996873684843385,
      "flos": 18691504104960.0,
      "grad_norm": 2.0852773996166207,
      "language_loss": 0.81583095,
      "learning_rate": 2.4813071395423307e-06,
      "loss": 0.83804578,
      "num_input_tokens_seen": 78791285,
      "step": 3659,
      "time_per_iteration": 2.4743473529815674
    },
    {
      "auxiliary_loss_clip": 0.0117028,
      "auxiliary_loss_mlp": 0.01030044,
      "balance_loss_clip": 1.0519917,
      "balance_loss_mlp": 1.02153277,
      "epoch": 0.4400889797390729,
      "flos": 23653460787840.0,
      "grad_norm": 1.7064208478686842,
      "language_loss": 0.6441437,
      "learning_rate": 2.4805510245463263e-06,
      "loss": 0.66614693,
      "num_input_tokens_seen": 78811440,
      "step": 3660,
      "time_per_iteration": 2.5392651557922363
    },
    {
      "auxiliary_loss_clip": 0.01163987,
      "auxiliary_loss_mlp": 0.01033012,
      "balance_loss_clip": 1.04937363,
      "balance_loss_mlp": 1.02454245,
      "epoch": 0.440209222629712,
      "flos": 23149203707520.0,
      "grad_norm": 2.4011307182740924,
      "language_loss": 0.6025871,
      "learning_rate": 2.4797948366511137e-06,
      "loss": 0.62455714,
      "num_input_tokens_seen": 78831150,
      "step": 3661,
      "time_per_iteration": 2.529275417327881
    },
    {
      "auxiliary_loss_clip": 0.01135531,
      "auxiliary_loss_mlp": 0.01033196,
      "balance_loss_clip": 1.04580712,
      "balance_loss_mlp": 1.02498889,
      "epoch": 0.4403294655203511,
      "flos": 24823394668800.0,
      "grad_norm": 1.7447646205832903,
      "language_loss": 0.76046097,
      "learning_rate": 2.4790385759714055e-06,
      "loss": 0.78214824,
      "num_input_tokens_seen": 78850215,
      "step": 3662,
      "time_per_iteration": 2.607314109802246
    },
    {
      "auxiliary_loss_clip": 0.01164688,
      "auxiliary_loss_mlp": 0.01028166,
      "balance_loss_clip": 1.05341768,
      "balance_loss_mlp": 1.02073956,
      "epoch": 0.4404497084109902,
      "flos": 22565080736640.0,
      "grad_norm": 1.6664371334875647,
      "language_loss": 0.71180958,
      "learning_rate": 2.478282242621926e-06,
      "loss": 0.73373806,
      "num_input_tokens_seen": 78870675,
      "step": 3663,
      "time_per_iteration": 2.550309419631958
    },
    {
      "auxiliary_loss_clip": 0.0104595,
      "auxiliary_loss_mlp": 0.01003969,
      "balance_loss_clip": 1.02023816,
      "balance_loss_mlp": 1.00262761,
      "epoch": 0.4405699513016293,
      "flos": 64967073448320.0,
      "grad_norm": 0.8436384432442813,
      "language_loss": 0.59605211,
      "learning_rate": 2.477525836717411e-06,
      "loss": 0.61655128,
      "num_input_tokens_seen": 78938440,
      "step": 3664,
      "time_per_iteration": 3.287449836730957
    },
    {
      "auxiliary_loss_clip": 0.0116681,
      "auxiliary_loss_mlp": 0.01032329,
      "balance_loss_clip": 1.05150414,
      "balance_loss_mlp": 1.02405596,
      "epoch": 0.4406901941922684,
      "flos": 35661952978560.0,
      "grad_norm": 2.381316932851427,
      "language_loss": 0.79098427,
      "learning_rate": 2.476769358372606e-06,
      "loss": 0.81297565,
      "num_input_tokens_seen": 78960090,
      "step": 3665,
      "time_per_iteration": 2.629042148590088
    },
    {
      "auxiliary_loss_clip": 0.01137725,
      "auxiliary_loss_mlp": 0.01028866,
      "balance_loss_clip": 1.05157721,
      "balance_loss_mlp": 1.02128696,
      "epoch": 0.44081043708290746,
      "flos": 18040767361920.0,
      "grad_norm": 2.769569073390376,
      "language_loss": 0.75174904,
      "learning_rate": 2.4760128077022683e-06,
      "loss": 0.77341491,
      "num_input_tokens_seen": 78978225,
      "step": 3666,
      "time_per_iteration": 2.557176113128662
    },
    {
      "auxiliary_loss_clip": 0.01111277,
      "auxiliary_loss_mlp": 0.01026427,
      "balance_loss_clip": 1.04482329,
      "balance_loss_mlp": 1.01834786,
      "epoch": 0.44093067997354657,
      "flos": 30153507799680.0,
      "grad_norm": 1.4503410054776142,
      "language_loss": 0.68460256,
      "learning_rate": 2.4752561848211672e-06,
      "loss": 0.70597959,
      "num_input_tokens_seen": 79000625,
      "step": 3667,
      "time_per_iteration": 2.661189317703247
    },
    {
      "auxiliary_loss_clip": 0.01166118,
      "auxiliary_loss_mlp": 0.01031824,
      "balance_loss_clip": 1.05454731,
      "balance_loss_mlp": 1.02439785,
      "epoch": 0.4410509228641857,
      "flos": 23255068066560.0,
      "grad_norm": 2.034989413271475,
      "language_loss": 0.71115094,
      "learning_rate": 2.4744994898440797e-06,
      "loss": 0.73313034,
      "num_input_tokens_seen": 79019415,
      "step": 3668,
      "time_per_iteration": 2.537574291229248
    },
    {
      "auxiliary_loss_clip": 0.01143853,
      "auxiliary_loss_mlp": 0.01028223,
      "balance_loss_clip": 1.04916382,
      "balance_loss_mlp": 1.01983643,
      "epoch": 0.44117116575482473,
      "flos": 19500571998720.0,
      "grad_norm": 1.9579947909509583,
      "language_loss": 0.83239084,
      "learning_rate": 2.473742722885797e-06,
      "loss": 0.85411167,
      "num_input_tokens_seen": 79038435,
      "step": 3669,
      "time_per_iteration": 2.57110333442688
    },
    {
      "auxiliary_loss_clip": 0.01167249,
      "auxiliary_loss_mlp": 0.00762111,
      "balance_loss_clip": 1.05410194,
      "balance_loss_mlp": 1.00019169,
      "epoch": 0.44129140864546385,
      "flos": 27053124353280.0,
      "grad_norm": 2.2732836795887255,
      "language_loss": 0.65231735,
      "learning_rate": 2.4729858840611197e-06,
      "loss": 0.67161095,
      "num_input_tokens_seen": 79057345,
      "step": 3670,
      "time_per_iteration": 2.5720767974853516
    },
    {
      "auxiliary_loss_clip": 0.01181796,
      "auxiliary_loss_mlp": 0.01033392,
      "balance_loss_clip": 1.05471992,
      "balance_loss_mlp": 1.02521467,
      "epoch": 0.4414116515361029,
      "flos": 26102101910400.0,
      "grad_norm": 1.8114658503077117,
      "language_loss": 0.72671103,
      "learning_rate": 2.4722289734848605e-06,
      "loss": 0.74886286,
      "num_input_tokens_seen": 79077810,
      "step": 3671,
      "time_per_iteration": 2.5232865810394287
    },
    {
      "auxiliary_loss_clip": 0.0113757,
      "auxiliary_loss_mlp": 0.01027967,
      "balance_loss_clip": 1.0527178,
      "balance_loss_mlp": 1.02030778,
      "epoch": 0.441531894426742,
      "flos": 21906083865600.0,
      "grad_norm": 1.979328064028822,
      "language_loss": 0.77698612,
      "learning_rate": 2.471471991271841e-06,
      "loss": 0.79864144,
      "num_input_tokens_seen": 79094935,
      "step": 3672,
      "time_per_iteration": 2.5861072540283203
    },
    {
      "auxiliary_loss_clip": 0.01157678,
      "auxiliary_loss_mlp": 0.01030315,
      "balance_loss_clip": 1.04863203,
      "balance_loss_mlp": 1.02239418,
      "epoch": 0.4416521373173811,
      "flos": 23437099215360.0,
      "grad_norm": 1.7858626856995332,
      "language_loss": 0.79535162,
      "learning_rate": 2.470714937536896e-06,
      "loss": 0.81723154,
      "num_input_tokens_seen": 79113660,
      "step": 3673,
      "time_per_iteration": 2.5166759490966797
    },
    {
      "auxiliary_loss_clip": 0.01119573,
      "auxiliary_loss_mlp": 0.01030909,
      "balance_loss_clip": 1.04470587,
      "balance_loss_mlp": 1.02282071,
      "epoch": 0.4417723802080202,
      "flos": 20334345471360.0,
      "grad_norm": 1.8978006927696147,
      "language_loss": 0.70746291,
      "learning_rate": 2.469957812394868e-06,
      "loss": 0.72896773,
      "num_input_tokens_seen": 79132470,
      "step": 3674,
      "time_per_iteration": 3.373581647872925
    },
    {
      "auxiliary_loss_clip": 0.0118179,
      "auxiliary_loss_mlp": 0.01030038,
      "balance_loss_clip": 1.05560493,
      "balance_loss_mlp": 1.02205729,
      "epoch": 0.4418926230986593,
      "flos": 18880682060160.0,
      "grad_norm": 1.8888998618160189,
      "language_loss": 0.76250106,
      "learning_rate": 2.4692006159606148e-06,
      "loss": 0.78461933,
      "num_input_tokens_seen": 79150000,
      "step": 3675,
      "time_per_iteration": 2.5005860328674316
    },
    {
      "auxiliary_loss_clip": 0.01181065,
      "auxiliary_loss_mlp": 0.01027873,
      "balance_loss_clip": 1.05392253,
      "balance_loss_mlp": 1.01988649,
      "epoch": 0.4420128659892984,
      "flos": 19464409981440.0,
      "grad_norm": 1.6646423916833633,
      "language_loss": 0.78793371,
      "learning_rate": 2.468443348349e-06,
      "loss": 0.81002319,
      "num_input_tokens_seen": 79167875,
      "step": 3676,
      "time_per_iteration": 2.462533950805664
    },
    {
      "auxiliary_loss_clip": 0.01121264,
      "auxiliary_loss_mlp": 0.01026836,
      "balance_loss_clip": 1.04470873,
      "balance_loss_mlp": 1.01925445,
      "epoch": 0.44213310887993745,
      "flos": 17894359526400.0,
      "grad_norm": 2.298887330239364,
      "language_loss": 0.82528985,
      "learning_rate": 2.467686009674902e-06,
      "loss": 0.84677088,
      "num_input_tokens_seen": 79182325,
      "step": 3677,
      "time_per_iteration": 2.5998852252960205
    },
    {
      "auxiliary_loss_clip": 0.01161147,
      "auxiliary_loss_mlp": 0.01027201,
      "balance_loss_clip": 1.04958391,
      "balance_loss_mlp": 1.01834941,
      "epoch": 0.44225335177057656,
      "flos": 19204667758080.0,
      "grad_norm": 1.9823622068428723,
      "language_loss": 0.85050106,
      "learning_rate": 2.466928600053209e-06,
      "loss": 0.87238455,
      "num_input_tokens_seen": 79197630,
      "step": 3678,
      "time_per_iteration": 2.4864389896392822
    },
    {
      "auxiliary_loss_clip": 0.01148075,
      "auxiliary_loss_mlp": 0.01030169,
      "balance_loss_clip": 1.04715312,
      "balance_loss_mlp": 1.02196717,
      "epoch": 0.4423735946612157,
      "flos": 23471321898240.0,
      "grad_norm": 1.7468959035035383,
      "language_loss": 0.71397638,
      "learning_rate": 2.466171119598818e-06,
      "loss": 0.73575878,
      "num_input_tokens_seen": 79217600,
      "step": 3679,
      "time_per_iteration": 4.1176323890686035
    },
    {
      "auxiliary_loss_clip": 0.01170383,
      "auxiliary_loss_mlp": 0.01028538,
      "balance_loss_clip": 1.05006528,
      "balance_loss_mlp": 1.01984131,
      "epoch": 0.44249383755185473,
      "flos": 26685398868480.0,
      "grad_norm": 1.7584698318825305,
      "language_loss": 0.77125156,
      "learning_rate": 2.465413568426639e-06,
      "loss": 0.79324073,
      "num_input_tokens_seen": 79238550,
      "step": 3680,
      "time_per_iteration": 3.2588226795196533
    },
    {
      "auxiliary_loss_clip": 0.01165047,
      "auxiliary_loss_mlp": 0.01023049,
      "balance_loss_clip": 1.05396843,
      "balance_loss_mlp": 1.01546109,
      "epoch": 0.44261408044249384,
      "flos": 23147659422720.0,
      "grad_norm": 1.5992341460296036,
      "language_loss": 0.81109458,
      "learning_rate": 2.464655946651591e-06,
      "loss": 0.83297551,
      "num_input_tokens_seen": 79257555,
      "step": 3681,
      "time_per_iteration": 2.5403196811676025
    },
    {
      "auxiliary_loss_clip": 0.01166969,
      "auxiliary_loss_mlp": 0.01035452,
      "balance_loss_clip": 1.05196035,
      "balance_loss_mlp": 1.02720857,
      "epoch": 0.44273432333313295,
      "flos": 24462564595200.0,
      "grad_norm": 1.950938520072444,
      "language_loss": 0.8086955,
      "learning_rate": 2.4638982543886065e-06,
      "loss": 0.83071971,
      "num_input_tokens_seen": 79277595,
      "step": 3682,
      "time_per_iteration": 2.547748565673828
    },
    {
      "auxiliary_loss_clip": 0.01171683,
      "auxiliary_loss_mlp": 0.01028525,
      "balance_loss_clip": 1.05376184,
      "balance_loss_mlp": 1.02011466,
      "epoch": 0.442854566223772,
      "flos": 17528932512000.0,
      "grad_norm": 5.180652532348916,
      "language_loss": 0.8735956,
      "learning_rate": 2.4631404917526254e-06,
      "loss": 0.89559758,
      "num_input_tokens_seen": 79294550,
      "step": 3683,
      "time_per_iteration": 2.4795420169830322
    },
    {
      "auxiliary_loss_clip": 0.01156708,
      "auxiliary_loss_mlp": 0.01025036,
      "balance_loss_clip": 1.04748988,
      "balance_loss_mlp": 1.01678097,
      "epoch": 0.4429748091144111,
      "flos": 24896293320960.0,
      "grad_norm": 1.6357021328017425,
      "language_loss": 0.78990614,
      "learning_rate": 2.4623826588586e-06,
      "loss": 0.81172353,
      "num_input_tokens_seen": 79314820,
      "step": 3684,
      "time_per_iteration": 2.5401480197906494
    },
    {
      "auxiliary_loss_clip": 0.01143427,
      "auxiliary_loss_mlp": 0.01025318,
      "balance_loss_clip": 1.04477668,
      "balance_loss_mlp": 1.01685452,
      "epoch": 0.4430950520050502,
      "flos": 21614704738560.0,
      "grad_norm": 1.6421963109300628,
      "language_loss": 0.82871526,
      "learning_rate": 2.461624755821492e-06,
      "loss": 0.85040271,
      "num_input_tokens_seen": 79334300,
      "step": 3685,
      "time_per_iteration": 2.5685486793518066
    },
    {
      "auxiliary_loss_clip": 0.01141008,
      "auxiliary_loss_mlp": 0.01032448,
      "balance_loss_clip": 1.05037796,
      "balance_loss_mlp": 1.02430046,
      "epoch": 0.4432152948956893,
      "flos": 24572271709440.0,
      "grad_norm": 1.9585472290685821,
      "language_loss": 0.76424009,
      "learning_rate": 2.4608667827562763e-06,
      "loss": 0.78597462,
      "num_input_tokens_seen": 79353630,
      "step": 3686,
      "time_per_iteration": 2.606872797012329
    },
    {
      "auxiliary_loss_clip": 0.01171178,
      "auxiliary_loss_mlp": 0.01028976,
      "balance_loss_clip": 1.05279326,
      "balance_loss_mlp": 1.02089345,
      "epoch": 0.4433355377863284,
      "flos": 21762261809280.0,
      "grad_norm": 1.9605163276579123,
      "language_loss": 0.90052909,
      "learning_rate": 2.460108739777936e-06,
      "loss": 0.92253065,
      "num_input_tokens_seen": 79372765,
      "step": 3687,
      "time_per_iteration": 2.5326123237609863
    },
    {
      "auxiliary_loss_clip": 0.01152105,
      "auxiliary_loss_mlp": 0.01028489,
      "balance_loss_clip": 1.05253077,
      "balance_loss_mlp": 1.02103806,
      "epoch": 0.44345578067696745,
      "flos": 20084479488000.0,
      "grad_norm": 1.460018936225211,
      "language_loss": 0.76464283,
      "learning_rate": 2.4593506270014656e-06,
      "loss": 0.78644872,
      "num_input_tokens_seen": 79391735,
      "step": 3688,
      "time_per_iteration": 2.525702714920044
    },
    {
      "auxiliary_loss_clip": 0.01152971,
      "auxiliary_loss_mlp": 0.01030079,
      "balance_loss_clip": 1.0482862,
      "balance_loss_mlp": 1.02188945,
      "epoch": 0.44357602356760656,
      "flos": 24169497528960.0,
      "grad_norm": 1.5547629625927406,
      "language_loss": 0.81860745,
      "learning_rate": 2.45859244454187e-06,
      "loss": 0.84043789,
      "num_input_tokens_seen": 79411525,
      "step": 3689,
      "time_per_iteration": 2.611504554748535
    },
    {
      "auxiliary_loss_clip": 0.011656,
      "auxiliary_loss_mlp": 0.01028378,
      "balance_loss_clip": 1.052652,
      "balance_loss_mlp": 1.02068293,
      "epoch": 0.44369626645824567,
      "flos": 22707717644160.0,
      "grad_norm": 1.6026801094566474,
      "language_loss": 0.65959865,
      "learning_rate": 2.4578341925141655e-06,
      "loss": 0.6815384,
      "num_input_tokens_seen": 79430740,
      "step": 3690,
      "time_per_iteration": 2.529740571975708
    },
    {
      "auxiliary_loss_clip": 0.01170563,
      "auxiliary_loss_mlp": 0.01032795,
      "balance_loss_clip": 1.05125046,
      "balance_loss_mlp": 1.02464151,
      "epoch": 0.4438165093488847,
      "flos": 38030225420160.0,
      "grad_norm": 2.431541790362739,
      "language_loss": 0.72273368,
      "learning_rate": 2.457075871033378e-06,
      "loss": 0.74476731,
      "num_input_tokens_seen": 79452615,
      "step": 3691,
      "time_per_iteration": 2.672551393508911
    },
    {
      "auxiliary_loss_clip": 0.01137355,
      "auxiliary_loss_mlp": 0.01027844,
      "balance_loss_clip": 1.05109489,
      "balance_loss_mlp": 1.01996446,
      "epoch": 0.44393675223952384,
      "flos": 15523213996800.0,
      "grad_norm": 1.9604060501236367,
      "language_loss": 0.88461423,
      "learning_rate": 2.4563174802145445e-06,
      "loss": 0.90626621,
      "num_input_tokens_seen": 79469865,
      "step": 3692,
      "time_per_iteration": 2.5745689868927
    },
    {
      "auxiliary_loss_clip": 0.01054237,
      "auxiliary_loss_mlp": 0.01008638,
      "balance_loss_clip": 1.01842332,
      "balance_loss_mlp": 1.00718331,
      "epoch": 0.44405699513016295,
      "flos": 64574893779840.0,
      "grad_norm": 0.6406451687479837,
      "language_loss": 0.48622996,
      "learning_rate": 2.455559020172712e-06,
      "loss": 0.50685877,
      "num_input_tokens_seen": 79537220,
      "step": 3693,
      "time_per_iteration": 3.225107431411743
    },
    {
      "auxiliary_loss_clip": 0.01129023,
      "auxiliary_loss_mlp": 0.01033024,
      "balance_loss_clip": 1.05228472,
      "balance_loss_mlp": 1.02420247,
      "epoch": 0.444177238020802,
      "flos": 23987394552960.0,
      "grad_norm": 2.2250632606652596,
      "language_loss": 0.896837,
      "learning_rate": 2.4548004910229385e-06,
      "loss": 0.91845751,
      "num_input_tokens_seen": 79554795,
      "step": 3694,
      "time_per_iteration": 2.655611276626587
    },
    {
      "auxiliary_loss_clip": 0.01170702,
      "auxiliary_loss_mlp": 0.00761911,
      "balance_loss_clip": 1.05355299,
      "balance_loss_mlp": 1.00020242,
      "epoch": 0.4442974809114411,
      "flos": 22563069575040.0,
      "grad_norm": 1.7965310753674983,
      "language_loss": 0.8687489,
      "learning_rate": 2.4540418928802913e-06,
      "loss": 0.88807505,
      "num_input_tokens_seen": 79573530,
      "step": 3695,
      "time_per_iteration": 2.5192062854766846
    },
    {
      "auxiliary_loss_clip": 0.0115067,
      "auxiliary_loss_mlp": 0.01033914,
      "balance_loss_clip": 1.04919207,
      "balance_loss_mlp": 1.02528906,
      "epoch": 0.4444177238020802,
      "flos": 17675699483520.0,
      "grad_norm": 2.3241125390670847,
      "language_loss": 0.66085792,
      "learning_rate": 2.4532832258598506e-06,
      "loss": 0.68270373,
      "num_input_tokens_seen": 79591360,
      "step": 3696,
      "time_per_iteration": 2.5364935398101807
    },
    {
      "auxiliary_loss_clip": 0.01178289,
      "auxiliary_loss_mlp": 0.01029482,
      "balance_loss_clip": 1.05276024,
      "balance_loss_mlp": 1.02198362,
      "epoch": 0.4445379666927193,
      "flos": 28621594609920.0,
      "grad_norm": 1.783029574133204,
      "language_loss": 0.80561662,
      "learning_rate": 2.4525244900767047e-06,
      "loss": 0.82769436,
      "num_input_tokens_seen": 79612175,
      "step": 3697,
      "time_per_iteration": 2.5528178215026855
    },
    {
      "auxiliary_loss_clip": 0.01065124,
      "auxiliary_loss_mlp": 0.01013828,
      "balance_loss_clip": 1.02373707,
      "balance_loss_mlp": 1.01257038,
      "epoch": 0.4446582095833584,
      "flos": 70487370115200.0,
      "grad_norm": 0.8475610869453014,
      "language_loss": 0.60530293,
      "learning_rate": 2.4517656856459536e-06,
      "loss": 0.62609243,
      "num_input_tokens_seen": 79678020,
      "step": 3698,
      "time_per_iteration": 3.1985974311828613
    },
    {
      "auxiliary_loss_clip": 0.01163865,
      "auxiliary_loss_mlp": 0.01032096,
      "balance_loss_clip": 1.04893899,
      "balance_loss_mlp": 1.02441883,
      "epoch": 0.4447784524739975,
      "flos": 26505199313280.0,
      "grad_norm": 1.6781375798734215,
      "language_loss": 0.67935461,
      "learning_rate": 2.4510068126827073e-06,
      "loss": 0.70131421,
      "num_input_tokens_seen": 79699020,
      "step": 3699,
      "time_per_iteration": 2.5809292793273926
    },
    {
      "auxiliary_loss_clip": 0.01150847,
      "auxiliary_loss_mlp": 0.01027039,
      "balance_loss_clip": 1.04884779,
      "balance_loss_mlp": 1.01921844,
      "epoch": 0.44489869536463655,
      "flos": 11656209553920.0,
      "grad_norm": 2.1862589120238622,
      "language_loss": 0.81894016,
      "learning_rate": 2.450247871302086e-06,
      "loss": 0.84071898,
      "num_input_tokens_seen": 79716795,
      "step": 3700,
      "time_per_iteration": 3.329430103302002
    },
    {
      "auxiliary_loss_clip": 0.01168277,
      "auxiliary_loss_mlp": 0.01025239,
      "balance_loss_clip": 1.05094039,
      "balance_loss_mlp": 1.0178417,
      "epoch": 0.44501893825527566,
      "flos": 20448469958400.0,
      "grad_norm": 2.2351595138707494,
      "language_loss": 0.83607012,
      "learning_rate": 2.44948886161922e-06,
      "loss": 0.85800529,
      "num_input_tokens_seen": 79735810,
      "step": 3701,
      "time_per_iteration": 2.49894642829895
    },
    {
      "auxiliary_loss_clip": 0.01170972,
      "auxiliary_loss_mlp": 0.01035231,
      "balance_loss_clip": 1.05405581,
      "balance_loss_mlp": 1.02781367,
      "epoch": 0.4451391811459148,
      "flos": 18261079430400.0,
      "grad_norm": 1.5503074821111649,
      "language_loss": 0.85090256,
      "learning_rate": 2.4487297837492524e-06,
      "loss": 0.87296462,
      "num_input_tokens_seen": 79754975,
      "step": 3702,
      "time_per_iteration": 2.525719404220581
    },
    {
      "auxiliary_loss_clip": 0.01138484,
      "auxiliary_loss_mlp": 0.01027797,
      "balance_loss_clip": 1.04967093,
      "balance_loss_mlp": 1.01979828,
      "epoch": 0.44525942403655383,
      "flos": 16910155895040.0,
      "grad_norm": 2.240812010575333,
      "language_loss": 0.62692833,
      "learning_rate": 2.4479706378073323e-06,
      "loss": 0.64859116,
      "num_input_tokens_seen": 79773515,
      "step": 3703,
      "time_per_iteration": 2.569617509841919
    },
    {
      "auxiliary_loss_clip": 0.01126677,
      "auxiliary_loss_mlp": 0.01032116,
      "balance_loss_clip": 1.04432261,
      "balance_loss_mlp": 1.02454054,
      "epoch": 0.44537966692719294,
      "flos": 23258838994560.0,
      "grad_norm": 1.5422658458408698,
      "language_loss": 0.83756363,
      "learning_rate": 2.447211423908623e-06,
      "loss": 0.8591516,
      "num_input_tokens_seen": 79793560,
      "step": 3704,
      "time_per_iteration": 2.636363983154297
    },
    {
      "auxiliary_loss_clip": 0.01169902,
      "auxiliary_loss_mlp": 0.01029388,
      "balance_loss_clip": 1.05280948,
      "balance_loss_mlp": 1.02209878,
      "epoch": 0.445499909817832,
      "flos": 21724160457600.0,
      "grad_norm": 2.019599411597836,
      "language_loss": 0.74949366,
      "learning_rate": 2.4464521421682966e-06,
      "loss": 0.77148658,
      "num_input_tokens_seen": 79811150,
      "step": 3705,
      "time_per_iteration": 4.126534938812256
    },
    {
      "auxiliary_loss_clip": 0.01162982,
      "auxiliary_loss_mlp": 0.01024951,
      "balance_loss_clip": 1.0525552,
      "balance_loss_mlp": 1.01797771,
      "epoch": 0.4456201527084711,
      "flos": 23987969170560.0,
      "grad_norm": 1.346617224504672,
      "language_loss": 0.87628913,
      "learning_rate": 2.4456927927015345e-06,
      "loss": 0.89816856,
      "num_input_tokens_seen": 79832190,
      "step": 3706,
      "time_per_iteration": 2.545527696609497
    },
    {
      "auxiliary_loss_clip": 0.01156033,
      "auxiliary_loss_mlp": 0.01029488,
      "balance_loss_clip": 1.05177712,
      "balance_loss_mlp": 1.02086902,
      "epoch": 0.4457403955991102,
      "flos": 18807065136000.0,
      "grad_norm": 2.1806278859943244,
      "language_loss": 0.76488209,
      "learning_rate": 2.4449333756235307e-06,
      "loss": 0.78673732,
      "num_input_tokens_seen": 79848905,
      "step": 3707,
      "time_per_iteration": 2.5758256912231445
    },
    {
      "auxiliary_loss_clip": 0.01168022,
      "auxiliary_loss_mlp": 0.01031765,
      "balance_loss_clip": 1.05269551,
      "balance_loss_mlp": 1.02389121,
      "epoch": 0.4458606384897493,
      "flos": 19207756327680.0,
      "grad_norm": 2.16023427299854,
      "language_loss": 0.7876147,
      "learning_rate": 2.4441738910494876e-06,
      "loss": 0.80961257,
      "num_input_tokens_seen": 79863640,
      "step": 3708,
      "time_per_iteration": 2.5344762802124023
    },
    {
      "auxiliary_loss_clip": 0.01155958,
      "auxiliary_loss_mlp": 0.01034429,
      "balance_loss_clip": 1.04800391,
      "balance_loss_mlp": 1.02628708,
      "epoch": 0.4459808813803884,
      "flos": 21361283308800.0,
      "grad_norm": 2.146947507334015,
      "language_loss": 0.81888247,
      "learning_rate": 2.4434143390946176e-06,
      "loss": 0.84078634,
      "num_input_tokens_seen": 79882450,
      "step": 3709,
      "time_per_iteration": 2.57564115524292
    },
    {
      "auxiliary_loss_clip": 0.01135503,
      "auxiliary_loss_mlp": 0.01027938,
      "balance_loss_clip": 1.04769921,
      "balance_loss_mlp": 1.02077651,
      "epoch": 0.4461011242710275,
      "flos": 23288967527040.0,
      "grad_norm": 1.9294741624848095,
      "language_loss": 0.85486013,
      "learning_rate": 2.4426547198741457e-06,
      "loss": 0.87649453,
      "num_input_tokens_seen": 79900655,
      "step": 3710,
      "time_per_iteration": 2.658766031265259
    },
    {
      "auxiliary_loss_clip": 0.0112571,
      "auxiliary_loss_mlp": 0.01026283,
      "balance_loss_clip": 1.05090213,
      "balance_loss_mlp": 1.01846337,
      "epoch": 0.44622136716166655,
      "flos": 20193001453440.0,
      "grad_norm": 1.9216181891323745,
      "language_loss": 0.74265778,
      "learning_rate": 2.441895033503305e-06,
      "loss": 0.76417768,
      "num_input_tokens_seen": 79918575,
      "step": 3711,
      "time_per_iteration": 2.60860013961792
    },
    {
      "auxiliary_loss_clip": 0.0116624,
      "auxiliary_loss_mlp": 0.01033464,
      "balance_loss_clip": 1.0521102,
      "balance_loss_mlp": 1.02501178,
      "epoch": 0.44634161005230566,
      "flos": 21283033530240.0,
      "grad_norm": 1.6854196319795625,
      "language_loss": 0.82279009,
      "learning_rate": 2.4411352800973375e-06,
      "loss": 0.84478712,
      "num_input_tokens_seen": 79937010,
      "step": 3712,
      "time_per_iteration": 2.5374560356140137
    },
    {
      "auxiliary_loss_clip": 0.01132739,
      "auxiliary_loss_mlp": 0.0102818,
      "balance_loss_clip": 1.04544508,
      "balance_loss_mlp": 1.02036572,
      "epoch": 0.44646185294294477,
      "flos": 22929358515840.0,
      "grad_norm": 2.5776990667171757,
      "language_loss": 0.75019848,
      "learning_rate": 2.4403754597715005e-06,
      "loss": 0.77180767,
      "num_input_tokens_seen": 79956455,
      "step": 3713,
      "time_per_iteration": 2.585812568664551
    },
    {
      "auxiliary_loss_clip": 0.01155317,
      "auxiliary_loss_mlp": 0.01026999,
      "balance_loss_clip": 1.04809725,
      "balance_loss_mlp": 1.01838601,
      "epoch": 0.4465820958335838,
      "flos": 22637692080000.0,
      "grad_norm": 2.2632990812346754,
      "language_loss": 0.93091381,
      "learning_rate": 2.4396155726410553e-06,
      "loss": 0.95273697,
      "num_input_tokens_seen": 79975065,
      "step": 3714,
      "time_per_iteration": 2.568756580352783
    },
    {
      "auxiliary_loss_clip": 0.01167334,
      "auxiliary_loss_mlp": 0.01028005,
      "balance_loss_clip": 1.0504899,
      "balance_loss_mlp": 1.02005935,
      "epoch": 0.44670233872422294,
      "flos": 22672525294080.0,
      "grad_norm": 2.2560874931145354,
      "language_loss": 0.90351045,
      "learning_rate": 2.438855618821278e-06,
      "loss": 0.92546391,
      "num_input_tokens_seen": 79990865,
      "step": 3715,
      "time_per_iteration": 2.5048890113830566
    },
    {
      "auxiliary_loss_clip": 0.01157022,
      "auxiliary_loss_mlp": 0.01028958,
      "balance_loss_clip": 1.04685879,
      "balance_loss_mlp": 1.02126908,
      "epoch": 0.44682258161486205,
      "flos": 23582178247680.0,
      "grad_norm": 1.6061326143039418,
      "language_loss": 0.67200565,
      "learning_rate": 2.4380955984274517e-06,
      "loss": 0.69386542,
      "num_input_tokens_seen": 80009520,
      "step": 3716,
      "time_per_iteration": 2.536433219909668
    },
    {
      "auxiliary_loss_clip": 0.01166814,
      "auxiliary_loss_mlp": 0.01030726,
      "balance_loss_clip": 1.05048037,
      "balance_loss_mlp": 1.02276921,
      "epoch": 0.4469428245055011,
      "flos": 26501356558080.0,
      "grad_norm": 1.9347400156395684,
      "language_loss": 0.76858413,
      "learning_rate": 2.4373355115748716e-06,
      "loss": 0.79055953,
      "num_input_tokens_seen": 80030350,
      "step": 3717,
      "time_per_iteration": 2.548346519470215
    },
    {
      "auxiliary_loss_clip": 0.01144682,
      "auxiliary_loss_mlp": 0.01035463,
      "balance_loss_clip": 1.04814243,
      "balance_loss_mlp": 1.02710068,
      "epoch": 0.4470630673961402,
      "flos": 21504925797120.0,
      "grad_norm": 1.7058735975554502,
      "language_loss": 0.72164947,
      "learning_rate": 2.436575358378842e-06,
      "loss": 0.74345094,
      "num_input_tokens_seen": 80049840,
      "step": 3718,
      "time_per_iteration": 2.5607166290283203
    },
    {
      "auxiliary_loss_clip": 0.01156465,
      "auxiliary_loss_mlp": 0.01037657,
      "balance_loss_clip": 1.04975951,
      "balance_loss_mlp": 1.02968812,
      "epoch": 0.44718331028677927,
      "flos": 16173986653440.0,
      "grad_norm": 10.492457842103649,
      "language_loss": 0.82636642,
      "learning_rate": 2.4358151389546782e-06,
      "loss": 0.84830761,
      "num_input_tokens_seen": 80066525,
      "step": 3719,
      "time_per_iteration": 2.5202276706695557
    },
    {
      "auxiliary_loss_clip": 0.01182927,
      "auxiliary_loss_mlp": 0.0102687,
      "balance_loss_clip": 1.05542302,
      "balance_loss_mlp": 1.01860905,
      "epoch": 0.4473035531774184,
      "flos": 19681238430720.0,
      "grad_norm": 2.106207601132298,
      "language_loss": 0.75686955,
      "learning_rate": 2.4350548534177035e-06,
      "loss": 0.77896756,
      "num_input_tokens_seen": 80083355,
      "step": 3720,
      "time_per_iteration": 2.4662587642669678
    },
    {
      "auxiliary_loss_clip": 0.01136352,
      "auxiliary_loss_mlp": 0.010301,
      "balance_loss_clip": 1.04614305,
      "balance_loss_mlp": 1.02228558,
      "epoch": 0.4474237960680575,
      "flos": 41427590515200.0,
      "grad_norm": 1.5456486069540167,
      "language_loss": 0.66515476,
      "learning_rate": 2.434294501883254e-06,
      "loss": 0.6868192,
      "num_input_tokens_seen": 80106450,
      "step": 3721,
      "time_per_iteration": 2.738900899887085
    },
    {
      "auxiliary_loss_clip": 0.01143144,
      "auxiliary_loss_mlp": 0.01029777,
      "balance_loss_clip": 1.04691172,
      "balance_loss_mlp": 1.02187932,
      "epoch": 0.44754403895869654,
      "flos": 22891328991360.0,
      "grad_norm": 1.7129312001303638,
      "language_loss": 0.65572894,
      "learning_rate": 2.433534084466674e-06,
      "loss": 0.67745811,
      "num_input_tokens_seen": 80125670,
      "step": 3722,
      "time_per_iteration": 2.568183183670044
    },
    {
      "auxiliary_loss_clip": 0.01176784,
      "auxiliary_loss_mlp": 0.01026838,
      "balance_loss_clip": 1.05175769,
      "balance_loss_mlp": 1.01896477,
      "epoch": 0.44766428184933565,
      "flos": 25630271832960.0,
      "grad_norm": 1.5230970619356274,
      "language_loss": 0.70447552,
      "learning_rate": 2.4327736012833178e-06,
      "loss": 0.72651172,
      "num_input_tokens_seen": 80147390,
      "step": 3723,
      "time_per_iteration": 2.5052080154418945
    },
    {
      "auxiliary_loss_clip": 0.01165577,
      "auxiliary_loss_mlp": 0.01028823,
      "balance_loss_clip": 1.05193019,
      "balance_loss_mlp": 1.0208478,
      "epoch": 0.44778452473997477,
      "flos": 20448972748800.0,
      "grad_norm": 2.064807714715758,
      "language_loss": 0.76358575,
      "learning_rate": 2.4320130524485506e-06,
      "loss": 0.78552973,
      "num_input_tokens_seen": 80166185,
      "step": 3724,
      "time_per_iteration": 2.529613733291626
    },
    {
      "auxiliary_loss_clip": 0.01146105,
      "auxiliary_loss_mlp": 0.01027992,
      "balance_loss_clip": 1.05029845,
      "balance_loss_mlp": 1.02044642,
      "epoch": 0.4479047676306138,
      "flos": 21975462984960.0,
      "grad_norm": 1.6070388774057396,
      "language_loss": 0.79664052,
      "learning_rate": 2.431252438077746e-06,
      "loss": 0.81838155,
      "num_input_tokens_seen": 80185685,
      "step": 3725,
      "time_per_iteration": 2.545245885848999
    },
    {
      "auxiliary_loss_clip": 0.01167602,
      "auxiliary_loss_mlp": 0.00762025,
      "balance_loss_clip": 1.04962873,
      "balance_loss_mlp": 1.00020874,
      "epoch": 0.44802501052125293,
      "flos": 21467219495040.0,
      "grad_norm": 4.197033547283291,
      "language_loss": 0.77635336,
      "learning_rate": 2.4304917582862906e-06,
      "loss": 0.79564959,
      "num_input_tokens_seen": 80204865,
      "step": 3726,
      "time_per_iteration": 3.357140302658081
    },
    {
      "auxiliary_loss_clip": 0.01183866,
      "auxiliary_loss_mlp": 0.01028634,
      "balance_loss_clip": 1.05628777,
      "balance_loss_mlp": 1.02091503,
      "epoch": 0.44814525341189204,
      "flos": 22126970551680.0,
      "grad_norm": 2.116973676661863,
      "language_loss": 0.87876272,
      "learning_rate": 2.4297310131895774e-06,
      "loss": 0.90088773,
      "num_input_tokens_seen": 80223410,
      "step": 3727,
      "time_per_iteration": 2.4591686725616455
    },
    {
      "auxiliary_loss_clip": 0.01168485,
      "auxiliary_loss_mlp": 0.01032906,
      "balance_loss_clip": 1.05278504,
      "balance_loss_mlp": 1.02462125,
      "epoch": 0.4482654963025311,
      "flos": 16653933204480.0,
      "grad_norm": 3.7153219381935196,
      "language_loss": 0.74190724,
      "learning_rate": 2.4289702029030113e-06,
      "loss": 0.76392108,
      "num_input_tokens_seen": 80240880,
      "step": 3728,
      "time_per_iteration": 2.494149923324585
    },
    {
      "auxiliary_loss_clip": 0.01168125,
      "auxiliary_loss_mlp": 0.01028537,
      "balance_loss_clip": 1.05263352,
      "balance_loss_mlp": 1.02049685,
      "epoch": 0.4483857391931702,
      "flos": 18841251905280.0,
      "grad_norm": 1.8776428544068313,
      "language_loss": 0.82943904,
      "learning_rate": 2.4282093275420057e-06,
      "loss": 0.85140562,
      "num_input_tokens_seen": 80259910,
      "step": 3729,
      "time_per_iteration": 2.5029714107513428
    },
    {
      "auxiliary_loss_clip": 0.01169381,
      "auxiliary_loss_mlp": 0.01034668,
      "balance_loss_clip": 1.05236161,
      "balance_loss_mlp": 1.02644897,
      "epoch": 0.4485059820838093,
      "flos": 20372590477440.0,
      "grad_norm": 2.0276213550221995,
      "language_loss": 0.70438468,
      "learning_rate": 2.4274483872219863e-06,
      "loss": 0.72642517,
      "num_input_tokens_seen": 80277270,
      "step": 3730,
      "time_per_iteration": 2.5254156589508057
    },
    {
      "auxiliary_loss_clip": 0.01163041,
      "auxiliary_loss_mlp": 0.01023881,
      "balance_loss_clip": 1.04945159,
      "balance_loss_mlp": 1.01678848,
      "epoch": 0.4486262249744484,
      "flos": 20047742853120.0,
      "grad_norm": 1.7256791028489342,
      "language_loss": 0.93629432,
      "learning_rate": 2.426687382058386e-06,
      "loss": 0.9581635,
      "num_input_tokens_seen": 80295550,
      "step": 3731,
      "time_per_iteration": 3.340275287628174
    },
    {
      "auxiliary_loss_clip": 0.01063666,
      "auxiliary_loss_mlp": 0.01004931,
      "balance_loss_clip": 1.0219686,
      "balance_loss_mlp": 1.00370932,
      "epoch": 0.4487464678650875,
      "flos": 64595684776320.0,
      "grad_norm": 0.8619712904131849,
      "language_loss": 0.59844339,
      "learning_rate": 2.425926312166649e-06,
      "loss": 0.6191293,
      "num_input_tokens_seen": 80348425,
      "step": 3732,
      "time_per_iteration": 3.679382562637329
    },
    {
      "auxiliary_loss_clip": 0.01156503,
      "auxiliary_loss_mlp": 0.01037421,
      "balance_loss_clip": 1.05011034,
      "balance_loss_mlp": 1.02886236,
      "epoch": 0.4488667107557266,
      "flos": 20769798049920.0,
      "grad_norm": 6.192275022314623,
      "language_loss": 0.73192149,
      "learning_rate": 2.42516517766223e-06,
      "loss": 0.75386071,
      "num_input_tokens_seen": 80366505,
      "step": 3733,
      "time_per_iteration": 2.5511629581451416
    },
    {
      "auxiliary_loss_clip": 0.01182641,
      "auxiliary_loss_mlp": 0.01027671,
      "balance_loss_clip": 1.05537701,
      "balance_loss_mlp": 1.01968443,
      "epoch": 0.44898695364636565,
      "flos": 23951735326080.0,
      "grad_norm": 2.16847034998956,
      "language_loss": 0.67932951,
      "learning_rate": 2.4244039786605907e-06,
      "loss": 0.70143265,
      "num_input_tokens_seen": 80387510,
      "step": 3734,
      "time_per_iteration": 2.530648946762085
    },
    {
      "auxiliary_loss_clip": 0.01125893,
      "auxiliary_loss_mlp": 0.01033397,
      "balance_loss_clip": 1.04650545,
      "balance_loss_mlp": 1.02508807,
      "epoch": 0.44910719653700476,
      "flos": 18624351628800.0,
      "grad_norm": 2.3013358157231583,
      "language_loss": 0.82571024,
      "learning_rate": 2.4236427152772055e-06,
      "loss": 0.84730315,
      "num_input_tokens_seen": 80405915,
      "step": 3735,
      "time_per_iteration": 2.5852949619293213
    },
    {
      "auxiliary_loss_clip": 0.010351,
      "auxiliary_loss_mlp": 0.01000981,
      "balance_loss_clip": 1.01919901,
      "balance_loss_mlp": 0.99984246,
      "epoch": 0.4492274394276438,
      "flos": 57033435749760.0,
      "grad_norm": 0.8217609648861407,
      "language_loss": 0.57374841,
      "learning_rate": 2.422881387627557e-06,
      "loss": 0.59410918,
      "num_input_tokens_seen": 80458365,
      "step": 3736,
      "time_per_iteration": 2.890310287475586
    },
    {
      "auxiliary_loss_clip": 0.01154044,
      "auxiliary_loss_mlp": 0.01026595,
      "balance_loss_clip": 1.05151665,
      "balance_loss_mlp": 1.01907563,
      "epoch": 0.4493476823182829,
      "flos": 23254888498560.0,
      "grad_norm": 1.660004501633974,
      "language_loss": 0.77376556,
      "learning_rate": 2.422119995827139e-06,
      "loss": 0.79557192,
      "num_input_tokens_seen": 80478490,
      "step": 3737,
      "time_per_iteration": 2.579089879989624
    },
    {
      "auxiliary_loss_clip": 0.01170235,
      "auxiliary_loss_mlp": 0.01030949,
      "balance_loss_clip": 1.05420017,
      "balance_loss_mlp": 1.02272701,
      "epoch": 0.44946792520892204,
      "flos": 15815131827840.0,
      "grad_norm": 2.3751748706337916,
      "language_loss": 0.74255991,
      "learning_rate": 2.4213585399914528e-06,
      "loss": 0.76457167,
      "num_input_tokens_seen": 80495695,
      "step": 3738,
      "time_per_iteration": 2.475811243057251
    },
    {
      "auxiliary_loss_clip": 0.01167074,
      "auxiliary_loss_mlp": 0.01024049,
      "balance_loss_clip": 1.05301154,
      "balance_loss_mlp": 1.01623487,
      "epoch": 0.4495881680995611,
      "flos": 19610063631360.0,
      "grad_norm": 1.7110419307668605,
      "language_loss": 0.85274756,
      "learning_rate": 2.4205970202360113e-06,
      "loss": 0.87465882,
      "num_input_tokens_seen": 80515260,
      "step": 3739,
      "time_per_iteration": 2.5008251667022705
    },
    {
      "auxiliary_loss_clip": 0.01116952,
      "auxiliary_loss_mlp": 0.01028766,
      "balance_loss_clip": 1.04703939,
      "balance_loss_mlp": 1.02002251,
      "epoch": 0.4497084109902002,
      "flos": 26031465815040.0,
      "grad_norm": 2.1830800775741723,
      "language_loss": 0.78032738,
      "learning_rate": 2.4198354366763354e-06,
      "loss": 0.80178457,
      "num_input_tokens_seen": 80533900,
      "step": 3740,
      "time_per_iteration": 2.63045072555542
    },
    {
      "auxiliary_loss_clip": 0.01154883,
      "auxiliary_loss_mlp": 0.010307,
      "balance_loss_clip": 1.04992461,
      "balance_loss_mlp": 1.02314186,
      "epoch": 0.4498286538808393,
      "flos": 14793688771200.0,
      "grad_norm": 1.9496122034549181,
      "language_loss": 0.78668541,
      "learning_rate": 2.4190737894279587e-06,
      "loss": 0.8085413,
      "num_input_tokens_seen": 80551270,
      "step": 3741,
      "time_per_iteration": 2.542106866836548
    },
    {
      "auxiliary_loss_clip": 0.01123846,
      "auxiliary_loss_mlp": 0.01028158,
      "balance_loss_clip": 1.04154646,
      "balance_loss_mlp": 1.01980138,
      "epoch": 0.44994889677147837,
      "flos": 15450171690240.0,
      "grad_norm": 2.3577224979803315,
      "language_loss": 0.80423629,
      "learning_rate": 2.4183120786064203e-06,
      "loss": 0.82575637,
      "num_input_tokens_seen": 80568145,
      "step": 3742,
      "time_per_iteration": 2.729404926300049
    },
    {
      "auxiliary_loss_clip": 0.01168431,
      "auxiliary_loss_mlp": 0.00761747,
      "balance_loss_clip": 1.05390978,
      "balance_loss_mlp": 1.00021648,
      "epoch": 0.4500691396621175,
      "flos": 21798316085760.0,
      "grad_norm": 2.316118362882173,
      "language_loss": 0.85375422,
      "learning_rate": 2.417550304327273e-06,
      "loss": 0.87305599,
      "num_input_tokens_seen": 80586185,
      "step": 3743,
      "time_per_iteration": 2.5212302207946777
    },
    {
      "auxiliary_loss_clip": 0.01184228,
      "auxiliary_loss_mlp": 0.01031505,
      "balance_loss_clip": 1.05560875,
      "balance_loss_mlp": 1.02297592,
      "epoch": 0.4501893825527566,
      "flos": 32382016421760.0,
      "grad_norm": 1.5746668950151628,
      "language_loss": 0.75656879,
      "learning_rate": 2.4167884667060763e-06,
      "loss": 0.7787261,
      "num_input_tokens_seen": 80608895,
      "step": 3744,
      "time_per_iteration": 2.5996124744415283
    },
    {
      "auxiliary_loss_clip": 0.0115192,
      "auxiliary_loss_mlp": 0.01039593,
      "balance_loss_clip": 1.04934537,
      "balance_loss_mlp": 1.03102231,
      "epoch": 0.45030962544339564,
      "flos": 16544944362240.0,
      "grad_norm": 2.6947670647959097,
      "language_loss": 0.87218684,
      "learning_rate": 2.4160265658584e-06,
      "loss": 0.89410204,
      "num_input_tokens_seen": 80623785,
      "step": 3745,
      "time_per_iteration": 2.51450514793396
    },
    {
      "auxiliary_loss_clip": 0.01169818,
      "auxiliary_loss_mlp": 0.01029341,
      "balance_loss_clip": 1.05292797,
      "balance_loss_mlp": 1.02104378,
      "epoch": 0.45042986833403476,
      "flos": 19573039687680.0,
      "grad_norm": 1.9568865787293412,
      "language_loss": 0.68479067,
      "learning_rate": 2.4152646018998253e-06,
      "loss": 0.70678228,
      "num_input_tokens_seen": 80642735,
      "step": 3746,
      "time_per_iteration": 2.5059092044830322
    },
    {
      "auxiliary_loss_clip": 0.01164538,
      "auxiliary_loss_mlp": 0.01026531,
      "balance_loss_clip": 1.05192351,
      "balance_loss_mlp": 1.01849663,
      "epoch": 0.45055011122467387,
      "flos": 23112467072640.0,
      "grad_norm": 1.671373828783129,
      "language_loss": 0.71769565,
      "learning_rate": 2.4145025749459403e-06,
      "loss": 0.73960632,
      "num_input_tokens_seen": 80663760,
      "step": 3747,
      "time_per_iteration": 2.542825698852539
    },
    {
      "auxiliary_loss_clip": 0.01095388,
      "auxiliary_loss_mlp": 0.0103046,
      "balance_loss_clip": 1.04750562,
      "balance_loss_mlp": 1.02216291,
      "epoch": 0.4506703541153129,
      "flos": 19934623946880.0,
      "grad_norm": 2.115509316035939,
      "language_loss": 0.69861931,
      "learning_rate": 2.413740485112344e-06,
      "loss": 0.71987778,
      "num_input_tokens_seen": 80682100,
      "step": 3748,
      "time_per_iteration": 2.7453818321228027
    },
    {
      "auxiliary_loss_clip": 0.0114622,
      "auxiliary_loss_mlp": 0.01030783,
      "balance_loss_clip": 1.05070257,
      "balance_loss_mlp": 1.0232724,
      "epoch": 0.45079059700595203,
      "flos": 19499530504320.0,
      "grad_norm": 1.557093654060581,
      "language_loss": 0.82065713,
      "learning_rate": 2.412978332514646e-06,
      "loss": 0.84242713,
      "num_input_tokens_seen": 80700880,
      "step": 3749,
      "time_per_iteration": 2.690563201904297
    },
    {
      "auxiliary_loss_clip": 0.01152998,
      "auxiliary_loss_mlp": 0.01032242,
      "balance_loss_clip": 1.0515883,
      "balance_loss_mlp": 1.02405834,
      "epoch": 0.4509108398965911,
      "flos": 27636313570560.0,
      "grad_norm": 2.028793880825944,
      "language_loss": 0.72269398,
      "learning_rate": 2.4122161172684623e-06,
      "loss": 0.74454635,
      "num_input_tokens_seen": 80721675,
      "step": 3750,
      "time_per_iteration": 2.635821580886841
    },
    {
      "auxiliary_loss_clip": 0.01154444,
      "auxiliary_loss_mlp": 0.01026486,
      "balance_loss_clip": 1.05321586,
      "balance_loss_mlp": 1.01790893,
      "epoch": 0.4510310827872302,
      "flos": 20995712640000.0,
      "grad_norm": 2.0244630287820558,
      "language_loss": 0.83929867,
      "learning_rate": 2.4114538394894216e-06,
      "loss": 0.86110795,
      "num_input_tokens_seen": 80739315,
      "step": 3751,
      "time_per_iteration": 2.5359854698181152
    },
    {
      "auxiliary_loss_clip": 0.01148524,
      "auxiliary_loss_mlp": 0.01023775,
      "balance_loss_clip": 1.04592478,
      "balance_loss_mlp": 1.01585388,
      "epoch": 0.4511513256778693,
      "flos": 16216684945920.0,
      "grad_norm": 1.772785382547845,
      "language_loss": 0.82975066,
      "learning_rate": 2.410691499293161e-06,
      "loss": 0.85147357,
      "num_input_tokens_seen": 80757470,
      "step": 3752,
      "time_per_iteration": 3.3461711406707764
    },
    {
      "auxiliary_loss_clip": 0.01165835,
      "auxiliary_loss_mlp": 0.01030081,
      "balance_loss_clip": 1.05040002,
      "balance_loss_mlp": 1.02220154,
      "epoch": 0.45127156856850836,
      "flos": 25186702780800.0,
      "grad_norm": 1.8424297677156591,
      "language_loss": 0.7385062,
      "learning_rate": 2.409929096795326e-06,
      "loss": 0.76046538,
      "num_input_tokens_seen": 80777840,
      "step": 3753,
      "time_per_iteration": 2.537637948989868
    },
    {
      "auxiliary_loss_clip": 0.01165229,
      "auxiliary_loss_mlp": 0.01032001,
      "balance_loss_clip": 1.04989612,
      "balance_loss_mlp": 1.0237577,
      "epoch": 0.4513918114591475,
      "flos": 20412523422720.0,
      "grad_norm": 1.996365031138014,
      "language_loss": 0.792328,
      "learning_rate": 2.409166632111573e-06,
      "loss": 0.8143003,
      "num_input_tokens_seen": 80795975,
      "step": 3754,
      "time_per_iteration": 2.5109753608703613
    },
    {
      "auxiliary_loss_clip": 0.0117402,
      "auxiliary_loss_mlp": 0.0102961,
      "balance_loss_clip": 1.0528717,
      "balance_loss_mlp": 1.0209012,
      "epoch": 0.4515120543497866,
      "flos": 26648482665600.0,
      "grad_norm": 1.921833505831142,
      "language_loss": 0.80469263,
      "learning_rate": 2.4084041053575674e-06,
      "loss": 0.82672888,
      "num_input_tokens_seen": 80815395,
      "step": 3755,
      "time_per_iteration": 2.5416147708892822
    },
    {
      "auxiliary_loss_clip": 0.01158654,
      "auxiliary_loss_mlp": 0.0103094,
      "balance_loss_clip": 1.05413532,
      "balance_loss_mlp": 1.0228039,
      "epoch": 0.45163229724042564,
      "flos": 20595093275520.0,
      "grad_norm": 1.977531459274755,
      "language_loss": 0.72547853,
      "learning_rate": 2.4076415166489834e-06,
      "loss": 0.74737448,
      "num_input_tokens_seen": 80834805,
      "step": 3756,
      "time_per_iteration": 2.549978256225586
    },
    {
      "auxiliary_loss_clip": 0.01129312,
      "auxiliary_loss_mlp": 0.01029098,
      "balance_loss_clip": 1.0475142,
      "balance_loss_mlp": 1.02183509,
      "epoch": 0.45175254013106475,
      "flos": 21689004021120.0,
      "grad_norm": 1.4973980060039136,
      "language_loss": 0.78973055,
      "learning_rate": 2.406878866101506e-06,
      "loss": 0.8113147,
      "num_input_tokens_seen": 80853770,
      "step": 3757,
      "time_per_iteration": 3.3884408473968506
    },
    {
      "auxiliary_loss_clip": 0.01180577,
      "auxiliary_loss_mlp": 0.0102968,
      "balance_loss_clip": 1.05456805,
      "balance_loss_mlp": 1.02171397,
      "epoch": 0.45187278302170386,
      "flos": 18878850466560.0,
      "grad_norm": 2.694086365620156,
      "language_loss": 0.78530377,
      "learning_rate": 2.4061161538308273e-06,
      "loss": 0.80740631,
      "num_input_tokens_seen": 80870615,
      "step": 3758,
      "time_per_iteration": 3.2624094486236572
    },
    {
      "auxiliary_loss_clip": 0.01167492,
      "auxiliary_loss_mlp": 0.01027073,
      "balance_loss_clip": 1.05339313,
      "balance_loss_mlp": 1.01934183,
      "epoch": 0.4519930259123429,
      "flos": 18582479349120.0,
      "grad_norm": 1.900172915747642,
      "language_loss": 0.88665944,
      "learning_rate": 2.4053533799526523e-06,
      "loss": 0.9086051,
      "num_input_tokens_seen": 80886335,
      "step": 3759,
      "time_per_iteration": 2.4971325397491455
    },
    {
      "auxiliary_loss_clip": 0.01148891,
      "auxiliary_loss_mlp": 0.0103088,
      "balance_loss_clip": 1.05199122,
      "balance_loss_mlp": 1.02358985,
      "epoch": 0.452113268802982,
      "flos": 25192377129600.0,
      "grad_norm": 1.6801494337811582,
      "language_loss": 0.86253804,
      "learning_rate": 2.404590544582691e-06,
      "loss": 0.88433576,
      "num_input_tokens_seen": 80904570,
      "step": 3760,
      "time_per_iteration": 2.5955586433410645
    },
    {
      "auxiliary_loss_clip": 0.0112755,
      "auxiliary_loss_mlp": 0.01032287,
      "balance_loss_clip": 1.04489613,
      "balance_loss_mlp": 1.02411509,
      "epoch": 0.45223351169362114,
      "flos": 39378922312320.0,
      "grad_norm": 1.5745157694883132,
      "language_loss": 0.80606765,
      "learning_rate": 2.403827647836666e-06,
      "loss": 0.82766604,
      "num_input_tokens_seen": 80925125,
      "step": 3761,
      "time_per_iteration": 2.7795450687408447
    },
    {
      "auxiliary_loss_clip": 0.01180162,
      "auxiliary_loss_mlp": 0.01025423,
      "balance_loss_clip": 1.05227709,
      "balance_loss_mlp": 1.01797581,
      "epoch": 0.4523537545842602,
      "flos": 21582169994880.0,
      "grad_norm": 2.0972195759709322,
      "language_loss": 0.69855225,
      "learning_rate": 2.4030646898303075e-06,
      "loss": 0.72060812,
      "num_input_tokens_seen": 80946615,
      "step": 3762,
      "time_per_iteration": 2.5028369426727295
    },
    {
      "auxiliary_loss_clip": 0.01158184,
      "auxiliary_loss_mlp": 0.01032132,
      "balance_loss_clip": 1.05061162,
      "balance_loss_mlp": 1.02431154,
      "epoch": 0.4524739974748993,
      "flos": 28439527547520.0,
      "grad_norm": 2.0851574561029866,
      "language_loss": 0.82166994,
      "learning_rate": 2.4023016706793566e-06,
      "loss": 0.84357309,
      "num_input_tokens_seen": 80966410,
      "step": 3763,
      "time_per_iteration": 2.6181085109710693
    },
    {
      "auxiliary_loss_clip": 0.01044042,
      "auxiliary_loss_mlp": 0.01003417,
      "balance_loss_clip": 1.01589954,
      "balance_loss_mlp": 1.00212944,
      "epoch": 0.4525942403655384,
      "flos": 61556492148480.0,
      "grad_norm": 0.7649879260504175,
      "language_loss": 0.56857371,
      "learning_rate": 2.401538590499561e-06,
      "loss": 0.58904827,
      "num_input_tokens_seen": 81026865,
      "step": 3764,
      "time_per_iteration": 3.2094645500183105
    },
    {
      "auxiliary_loss_clip": 0.01167742,
      "auxiliary_loss_mlp": 0.00761519,
      "balance_loss_clip": 1.0504601,
      "balance_loss_mlp": 1.00021195,
      "epoch": 0.45271448325617747,
      "flos": 27529838680320.0,
      "grad_norm": 5.202569655840854,
      "language_loss": 0.71258479,
      "learning_rate": 2.400775449406682e-06,
      "loss": 0.73187739,
      "num_input_tokens_seen": 81050060,
      "step": 3765,
      "time_per_iteration": 2.60562801361084
    },
    {
      "auxiliary_loss_clip": 0.01163676,
      "auxiliary_loss_mlp": 0.01031993,
      "balance_loss_clip": 1.04935014,
      "balance_loss_mlp": 1.0246613,
      "epoch": 0.4528347261468166,
      "flos": 22452608275200.0,
      "grad_norm": 1.7673253345238815,
      "language_loss": 0.73109686,
      "learning_rate": 2.400012247516485e-06,
      "loss": 0.75305355,
      "num_input_tokens_seen": 81070625,
      "step": 3766,
      "time_per_iteration": 2.5212621688842773
    },
    {
      "auxiliary_loss_clip": 0.01139534,
      "auxiliary_loss_mlp": 0.01029274,
      "balance_loss_clip": 1.04680789,
      "balance_loss_mlp": 1.02159715,
      "epoch": 0.45295496903745563,
      "flos": 21103875469440.0,
      "grad_norm": 1.7500070089493907,
      "language_loss": 0.90265751,
      "learning_rate": 2.3992489849447484e-06,
      "loss": 0.92434555,
      "num_input_tokens_seen": 81089080,
      "step": 3767,
      "time_per_iteration": 2.654156446456909
    },
    {
      "auxiliary_loss_clip": 0.01141357,
      "auxiliary_loss_mlp": 0.01030741,
      "balance_loss_clip": 1.04794121,
      "balance_loss_mlp": 1.02287936,
      "epoch": 0.45307521192809475,
      "flos": 23221168606080.0,
      "grad_norm": 2.545577789465279,
      "language_loss": 0.78824234,
      "learning_rate": 2.3984856618072584e-06,
      "loss": 0.80996335,
      "num_input_tokens_seen": 81109115,
      "step": 3768,
      "time_per_iteration": 2.6796483993530273
    },
    {
      "auxiliary_loss_clip": 0.01142118,
      "auxiliary_loss_mlp": 0.01028332,
      "balance_loss_clip": 1.0478586,
      "balance_loss_mlp": 1.02046978,
      "epoch": 0.45319545481873386,
      "flos": 15560094286080.0,
      "grad_norm": 1.9591170176243695,
      "language_loss": 0.74333227,
      "learning_rate": 2.3977222782198098e-06,
      "loss": 0.76503682,
      "num_input_tokens_seen": 81127750,
      "step": 3769,
      "time_per_iteration": 2.6364450454711914
    },
    {
      "auxiliary_loss_clip": 0.01125972,
      "auxiliary_loss_mlp": 0.01026841,
      "balance_loss_clip": 1.04405594,
      "balance_loss_mlp": 1.01839519,
      "epoch": 0.4533156977093729,
      "flos": 21944759834880.0,
      "grad_norm": 1.7229852175809637,
      "language_loss": 0.75324547,
      "learning_rate": 2.3969588342982077e-06,
      "loss": 0.77477366,
      "num_input_tokens_seen": 81147125,
      "step": 3770,
      "time_per_iteration": 2.5933375358581543
    },
    {
      "auxiliary_loss_clip": 0.01166781,
      "auxiliary_loss_mlp": 0.01027363,
      "balance_loss_clip": 1.05430102,
      "balance_loss_mlp": 1.01973379,
      "epoch": 0.453435940600012,
      "flos": 24242180699520.0,
      "grad_norm": 2.0172318808423984,
      "language_loss": 0.72666621,
      "learning_rate": 2.396195330158267e-06,
      "loss": 0.74860764,
      "num_input_tokens_seen": 81167015,
      "step": 3771,
      "time_per_iteration": 2.567723274230957
    },
    {
      "auxiliary_loss_clip": 0.01180011,
      "auxiliary_loss_mlp": 0.01028095,
      "balance_loss_clip": 1.0528878,
      "balance_loss_mlp": 1.02021551,
      "epoch": 0.45355618349065113,
      "flos": 23440367352960.0,
      "grad_norm": 1.7618341475052894,
      "language_loss": 0.79252362,
      "learning_rate": 2.3954317659158094e-06,
      "loss": 0.8146047,
      "num_input_tokens_seen": 81187350,
      "step": 3772,
      "time_per_iteration": 2.557277202606201
    },
    {
      "auxiliary_loss_clip": 0.01070173,
      "auxiliary_loss_mlp": 0.01001143,
      "balance_loss_clip": 1.01523507,
      "balance_loss_mlp": 0.99980748,
      "epoch": 0.4536764263812902,
      "flos": 66903161448960.0,
      "grad_norm": 0.8868259617590143,
      "language_loss": 0.56933814,
      "learning_rate": 2.394668141686667e-06,
      "loss": 0.59005129,
      "num_input_tokens_seen": 81249315,
      "step": 3773,
      "time_per_iteration": 3.12005352973938
    },
    {
      "auxiliary_loss_clip": 0.01161226,
      "auxiliary_loss_mlp": 0.01027976,
      "balance_loss_clip": 1.04910946,
      "balance_loss_mlp": 1.01991129,
      "epoch": 0.4537966692719293,
      "flos": 42739766254080.0,
      "grad_norm": 3.5653974189662967,
      "language_loss": 0.69459838,
      "learning_rate": 2.3939044575866813e-06,
      "loss": 0.71649045,
      "num_input_tokens_seen": 81272065,
      "step": 3774,
      "time_per_iteration": 2.715855360031128
    },
    {
      "auxiliary_loss_clip": 0.01145807,
      "auxiliary_loss_mlp": 0.00761299,
      "balance_loss_clip": 1.04776955,
      "balance_loss_mlp": 1.00016522,
      "epoch": 0.4539169121625684,
      "flos": 35549480517120.0,
      "grad_norm": 2.0428615785557596,
      "language_loss": 0.75969523,
      "learning_rate": 2.3931407137317024e-06,
      "loss": 0.77876627,
      "num_input_tokens_seen": 81292220,
      "step": 3775,
      "time_per_iteration": 2.6729605197906494
    },
    {
      "auxiliary_loss_clip": 0.01130669,
      "auxiliary_loss_mlp": 0.01030223,
      "balance_loss_clip": 1.04455352,
      "balance_loss_mlp": 1.02268887,
      "epoch": 0.45403715505320746,
      "flos": 18514716341760.0,
      "grad_norm": 1.680903914707323,
      "language_loss": 0.85047662,
      "learning_rate": 2.3923769102375907e-06,
      "loss": 0.87208557,
      "num_input_tokens_seen": 81311085,
      "step": 3776,
      "time_per_iteration": 2.5943236351013184
    },
    {
      "auxiliary_loss_clip": 0.01132686,
      "auxiliary_loss_mlp": 0.01029901,
      "balance_loss_clip": 1.04284859,
      "balance_loss_mlp": 1.02198529,
      "epoch": 0.4541573979438466,
      "flos": 25045825639680.0,
      "grad_norm": 2.2590623624294137,
      "language_loss": 0.78322482,
      "learning_rate": 2.391613047220213e-06,
      "loss": 0.8048507,
      "num_input_tokens_seen": 81330985,
      "step": 3777,
      "time_per_iteration": 2.6010138988494873
    },
    {
      "auxiliary_loss_clip": 0.01127349,
      "auxiliary_loss_mlp": 0.0102632,
      "balance_loss_clip": 1.04552877,
      "balance_loss_mlp": 1.01818955,
      "epoch": 0.4542776408344857,
      "flos": 18332397884160.0,
      "grad_norm": 1.9334093914807406,
      "language_loss": 0.78960192,
      "learning_rate": 2.390849124795447e-06,
      "loss": 0.81113863,
      "num_input_tokens_seen": 81346985,
      "step": 3778,
      "time_per_iteration": 3.4059925079345703
    },
    {
      "auxiliary_loss_clip": 0.01179921,
      "auxiliary_loss_mlp": 0.01033161,
      "balance_loss_clip": 1.05292284,
      "balance_loss_mlp": 1.02595472,
      "epoch": 0.45439788372512474,
      "flos": 20701173116160.0,
      "grad_norm": 3.7446396467162018,
      "language_loss": 0.84076071,
      "learning_rate": 2.3900851430791804e-06,
      "loss": 0.8628915,
      "num_input_tokens_seen": 81365005,
      "step": 3779,
      "time_per_iteration": 2.4777472019195557
    },
    {
      "auxiliary_loss_clip": 0.01183709,
      "auxiliary_loss_mlp": 0.01031912,
      "balance_loss_clip": 1.05449033,
      "balance_loss_mlp": 1.02393651,
      "epoch": 0.45451812661576385,
      "flos": 22309432663680.0,
      "grad_norm": 1.9315845149299342,
      "language_loss": 0.84753847,
      "learning_rate": 2.389321102187307e-06,
      "loss": 0.86969471,
      "num_input_tokens_seen": 81383785,
      "step": 3780,
      "time_per_iteration": 2.492516040802002
    },
    {
      "auxiliary_loss_clip": 0.01150123,
      "auxiliary_loss_mlp": 0.00762172,
      "balance_loss_clip": 1.05113888,
      "balance_loss_mlp": 1.00020003,
      "epoch": 0.4546383695064029,
      "flos": 21763303303680.0,
      "grad_norm": 1.9053929084710592,
      "language_loss": 0.81403887,
      "learning_rate": 2.3885570022357326e-06,
      "loss": 0.83316183,
      "num_input_tokens_seen": 81402915,
      "step": 3781,
      "time_per_iteration": 2.568053960800171
    },
    {
      "auxiliary_loss_clip": 0.01040487,
      "auxiliary_loss_mlp": 0.01000187,
      "balance_loss_clip": 1.01374555,
      "balance_loss_mlp": 0.99888211,
      "epoch": 0.454758612397042,
      "flos": 64242755694720.0,
      "grad_norm": 0.7973027278109948,
      "language_loss": 0.60883373,
      "learning_rate": 2.38779284334037e-06,
      "loss": 0.62924051,
      "num_input_tokens_seen": 81467890,
      "step": 3782,
      "time_per_iteration": 3.2188708782196045
    },
    {
      "auxiliary_loss_clip": 0.01108907,
      "auxiliary_loss_mlp": 0.01027455,
      "balance_loss_clip": 1.04292166,
      "balance_loss_mlp": 1.01944959,
      "epoch": 0.4548788552876811,
      "flos": 27304175485440.0,
      "grad_norm": 2.0087927195568245,
      "language_loss": 0.78783935,
      "learning_rate": 2.387028625617141e-06,
      "loss": 0.80920303,
      "num_input_tokens_seen": 81487105,
      "step": 3783,
      "time_per_iteration": 4.199683904647827
    },
    {
      "auxiliary_loss_clip": 0.01136009,
      "auxiliary_loss_mlp": 0.0102937,
      "balance_loss_clip": 1.0439204,
      "balance_loss_mlp": 1.02131116,
      "epoch": 0.4549990981783202,
      "flos": 22857142222080.0,
      "grad_norm": 1.9232732299776745,
      "language_loss": 0.84856898,
      "learning_rate": 2.3862643491819766e-06,
      "loss": 0.87022281,
      "num_input_tokens_seen": 81505670,
      "step": 3784,
      "time_per_iteration": 3.353611469268799
    },
    {
      "auxiliary_loss_clip": 0.01161471,
      "auxiliary_loss_mlp": 0.01031439,
      "balance_loss_clip": 1.04868889,
      "balance_loss_mlp": 1.02332664,
      "epoch": 0.4551193410689593,
      "flos": 23258587599360.0,
      "grad_norm": 1.7986853295778558,
      "language_loss": 0.84642726,
      "learning_rate": 2.3855000141508186e-06,
      "loss": 0.86835635,
      "num_input_tokens_seen": 81525825,
      "step": 3785,
      "time_per_iteration": 2.517411470413208
    },
    {
      "auxiliary_loss_clip": 0.01153024,
      "auxiliary_loss_mlp": 0.01031393,
      "balance_loss_clip": 1.05185723,
      "balance_loss_mlp": 1.02368021,
      "epoch": 0.4552395839595984,
      "flos": 20777519473920.0,
      "grad_norm": 2.1638363965872585,
      "language_loss": 0.84160626,
      "learning_rate": 2.3847356206396143e-06,
      "loss": 0.86345047,
      "num_input_tokens_seen": 81543135,
      "step": 3786,
      "time_per_iteration": 2.5667545795440674
    },
    {
      "auxiliary_loss_clip": 0.011788,
      "auxiliary_loss_mlp": 0.01026503,
      "balance_loss_clip": 1.05261099,
      "balance_loss_mlp": 1.0182476,
      "epoch": 0.45535982685023746,
      "flos": 23257510191360.0,
      "grad_norm": 1.8993395746155657,
      "language_loss": 0.785941,
      "learning_rate": 2.3839711687643227e-06,
      "loss": 0.80799401,
      "num_input_tokens_seen": 81564360,
      "step": 3787,
      "time_per_iteration": 2.493812084197998
    },
    {
      "auxiliary_loss_clip": 0.01164581,
      "auxiliary_loss_mlp": 0.01036055,
      "balance_loss_clip": 1.05120993,
      "balance_loss_mlp": 1.02697146,
      "epoch": 0.45548006974087657,
      "flos": 19646117907840.0,
      "grad_norm": 1.8965805983731625,
      "language_loss": 0.74043846,
      "learning_rate": 2.38320665864091e-06,
      "loss": 0.76244485,
      "num_input_tokens_seen": 81583710,
      "step": 3788,
      "time_per_iteration": 2.5088119506835938
    },
    {
      "auxiliary_loss_clip": 0.01109202,
      "auxiliary_loss_mlp": 0.01026415,
      "balance_loss_clip": 1.042243,
      "balance_loss_mlp": 1.01835692,
      "epoch": 0.4556003126315157,
      "flos": 20047778766720.0,
      "grad_norm": 2.18090340115957,
      "language_loss": 0.81917691,
      "learning_rate": 2.3824420903853516e-06,
      "loss": 0.84053314,
      "num_input_tokens_seen": 81602175,
      "step": 3789,
      "time_per_iteration": 2.645024061203003
    },
    {
      "auxiliary_loss_clip": 0.01163526,
      "auxiliary_loss_mlp": 0.01028044,
      "balance_loss_clip": 1.05225778,
      "balance_loss_mlp": 1.02034342,
      "epoch": 0.45572055552215474,
      "flos": 22959738443520.0,
      "grad_norm": 2.4949820159595957,
      "language_loss": 0.81972516,
      "learning_rate": 2.3816774641136324e-06,
      "loss": 0.84164083,
      "num_input_tokens_seen": 81619430,
      "step": 3790,
      "time_per_iteration": 2.5227949619293213
    },
    {
      "auxiliary_loss_clip": 0.01164165,
      "auxiliary_loss_mlp": 0.00762096,
      "balance_loss_clip": 1.05221486,
      "balance_loss_mlp": 1.00019598,
      "epoch": 0.45584079841279385,
      "flos": 33109925535360.0,
      "grad_norm": 1.801857214666825,
      "language_loss": 0.71666884,
      "learning_rate": 2.380912779941745e-06,
      "loss": 0.73593146,
      "num_input_tokens_seen": 81642550,
      "step": 3791,
      "time_per_iteration": 2.6076061725616455
    },
    {
      "auxiliary_loss_clip": 0.0116482,
      "auxiliary_loss_mlp": 0.01032322,
      "balance_loss_clip": 1.0475359,
      "balance_loss_mlp": 1.02392364,
      "epoch": 0.45596104130343296,
      "flos": 27272179445760.0,
      "grad_norm": 2.3027762903335742,
      "language_loss": 0.83261931,
      "learning_rate": 2.3801480379856918e-06,
      "loss": 0.85459077,
      "num_input_tokens_seen": 81664260,
      "step": 3792,
      "time_per_iteration": 2.561879873275757
    },
    {
      "auxiliary_loss_clip": 0.01153728,
      "auxiliary_loss_mlp": 0.01029146,
      "balance_loss_clip": 1.05192685,
      "balance_loss_mlp": 1.02167797,
      "epoch": 0.456081284194072,
      "flos": 21579799697280.0,
      "grad_norm": 1.6934736593844248,
      "language_loss": 0.8375355,
      "learning_rate": 2.379383238361484e-06,
      "loss": 0.85936421,
      "num_input_tokens_seen": 81683620,
      "step": 3793,
      "time_per_iteration": 2.5379748344421387
    },
    {
      "auxiliary_loss_clip": 0.01162525,
      "auxiliary_loss_mlp": 0.01030195,
      "balance_loss_clip": 1.04995668,
      "balance_loss_mlp": 1.02285802,
      "epoch": 0.4562015270847111,
      "flos": 35918822113920.0,
      "grad_norm": 2.0553376732504582,
      "language_loss": 0.79870224,
      "learning_rate": 2.3786183811851407e-06,
      "loss": 0.82062936,
      "num_input_tokens_seen": 81704325,
      "step": 3794,
      "time_per_iteration": 2.6312150955200195
    },
    {
      "auxiliary_loss_clip": 0.01180709,
      "auxiliary_loss_mlp": 0.01036963,
      "balance_loss_clip": 1.05374098,
      "balance_loss_mlp": 1.02926183,
      "epoch": 0.45632176997535023,
      "flos": 13589783602560.0,
      "grad_norm": 1.5802467703076075,
      "language_loss": 0.80032527,
      "learning_rate": 2.3778534665726892e-06,
      "loss": 0.82250202,
      "num_input_tokens_seen": 81721155,
      "step": 3795,
      "time_per_iteration": 2.450556993484497
    },
    {
      "auxiliary_loss_clip": 0.01155335,
      "auxiliary_loss_mlp": 0.01029725,
      "balance_loss_clip": 1.04955709,
      "balance_loss_mlp": 1.02194107,
      "epoch": 0.4564420128659893,
      "flos": 32635401937920.0,
      "grad_norm": 1.8152062035181948,
      "language_loss": 0.72958767,
      "learning_rate": 2.377088494640168e-06,
      "loss": 0.75143832,
      "num_input_tokens_seen": 81742905,
      "step": 3796,
      "time_per_iteration": 2.5993428230285645
    },
    {
      "auxiliary_loss_clip": 0.01161029,
      "auxiliary_loss_mlp": 0.01026251,
      "balance_loss_clip": 1.05149388,
      "balance_loss_mlp": 1.01860404,
      "epoch": 0.4565622557566284,
      "flos": 20377690208640.0,
      "grad_norm": 1.792088329194526,
      "language_loss": 0.78275025,
      "learning_rate": 2.3763234655036216e-06,
      "loss": 0.80462307,
      "num_input_tokens_seen": 81762105,
      "step": 3797,
      "time_per_iteration": 2.4977407455444336
    },
    {
      "auxiliary_loss_clip": 0.01135705,
      "auxiliary_loss_mlp": 0.01035473,
      "balance_loss_clip": 1.04448485,
      "balance_loss_mlp": 1.02767086,
      "epoch": 0.45668249864726745,
      "flos": 25374372364800.0,
      "grad_norm": 2.1049223235504786,
      "language_loss": 0.8714186,
      "learning_rate": 2.3755583792791046e-06,
      "loss": 0.89313036,
      "num_input_tokens_seen": 81781975,
      "step": 3798,
      "time_per_iteration": 2.6019344329833984
    },
    {
      "auxiliary_loss_clip": 0.01167196,
      "auxiliary_loss_mlp": 0.01035724,
      "balance_loss_clip": 1.05087304,
      "balance_loss_mlp": 1.02815986,
      "epoch": 0.45680274153790656,
      "flos": 15559806977280.0,
      "grad_norm": 1.8613305574932022,
      "language_loss": 0.74481493,
      "learning_rate": 2.3747932360826803e-06,
      "loss": 0.76684415,
      "num_input_tokens_seen": 81798905,
      "step": 3799,
      "time_per_iteration": 2.472931146621704
    },
    {
      "auxiliary_loss_clip": 0.01165934,
      "auxiliary_loss_mlp": 0.01024436,
      "balance_loss_clip": 1.05413246,
      "balance_loss_mlp": 1.01612139,
      "epoch": 0.4569229844285457,
      "flos": 19792884879360.0,
      "grad_norm": 2.1837763210766377,
      "language_loss": 0.82427132,
      "learning_rate": 2.3740280360304205e-06,
      "loss": 0.84617507,
      "num_input_tokens_seen": 81816630,
      "step": 3800,
      "time_per_iteration": 2.4984776973724365
    },
    {
      "auxiliary_loss_clip": 0.01139938,
      "auxiliary_loss_mlp": 0.01032198,
      "balance_loss_clip": 1.0512948,
      "balance_loss_mlp": 1.02480721,
      "epoch": 0.45704322731918473,
      "flos": 24093941270400.0,
      "grad_norm": 1.7945169905518796,
      "language_loss": 0.68115139,
      "learning_rate": 2.3732627792384038e-06,
      "loss": 0.70287275,
      "num_input_tokens_seen": 81837700,
      "step": 3801,
      "time_per_iteration": 2.592421770095825
    },
    {
      "auxiliary_loss_clip": 0.01177858,
      "auxiliary_loss_mlp": 0.01023501,
      "balance_loss_clip": 1.05098438,
      "balance_loss_mlp": 1.01615763,
      "epoch": 0.45716347020982384,
      "flos": 31317803245440.0,
      "grad_norm": 1.8708389401223504,
      "language_loss": 0.75378042,
      "learning_rate": 2.3724974658227207e-06,
      "loss": 0.77579403,
      "num_input_tokens_seen": 81858490,
      "step": 3802,
      "time_per_iteration": 2.544110059738159
    },
    {
      "auxiliary_loss_clip": 0.01150097,
      "auxiliary_loss_mlp": 0.00762382,
      "balance_loss_clip": 1.05009663,
      "balance_loss_mlp": 1.00019372,
      "epoch": 0.45728371310046295,
      "flos": 26501392471680.0,
      "grad_norm": 1.8379132774020817,
      "language_loss": 0.71100938,
      "learning_rate": 2.3717320958994687e-06,
      "loss": 0.73013413,
      "num_input_tokens_seen": 81876050,
      "step": 3803,
      "time_per_iteration": 2.571352005004883
    },
    {
      "auxiliary_loss_clip": 0.01136345,
      "auxiliary_loss_mlp": 0.01036969,
      "balance_loss_clip": 1.04312027,
      "balance_loss_mlp": 1.02887511,
      "epoch": 0.457403955991102,
      "flos": 17929408222080.0,
      "grad_norm": 1.8578063434708478,
      "language_loss": 0.70411956,
      "learning_rate": 2.3709666695847534e-06,
      "loss": 0.72585273,
      "num_input_tokens_seen": 81894230,
      "step": 3804,
      "time_per_iteration": 3.422168731689453
    },
    {
      "auxiliary_loss_clip": 0.01116562,
      "auxiliary_loss_mlp": 0.0102792,
      "balance_loss_clip": 1.04452276,
      "balance_loss_mlp": 1.02060986,
      "epoch": 0.4575241988817411,
      "flos": 42230660837760.0,
      "grad_norm": 1.5551688135112223,
      "language_loss": 0.70166409,
      "learning_rate": 2.370201186994689e-06,
      "loss": 0.72310889,
      "num_input_tokens_seen": 81917915,
      "step": 3805,
      "time_per_iteration": 2.8034238815307617
    },
    {
      "auxiliary_loss_clip": 0.01142309,
      "auxiliary_loss_mlp": 0.01028631,
      "balance_loss_clip": 1.0493536,
      "balance_loss_mlp": 1.02066207,
      "epoch": 0.45764444177238023,
      "flos": 30117309868800.0,
      "grad_norm": 1.9353637129858212,
      "language_loss": 0.69575,
      "learning_rate": 2.369435648245399e-06,
      "loss": 0.71745944,
      "num_input_tokens_seen": 81938130,
      "step": 3806,
      "time_per_iteration": 2.6137192249298096
    },
    {
      "auxiliary_loss_clip": 0.01149112,
      "auxiliary_loss_mlp": 0.01031643,
      "balance_loss_clip": 1.04974532,
      "balance_loss_mlp": 1.02370358,
      "epoch": 0.4577646846630193,
      "flos": 24060293205120.0,
      "grad_norm": 1.636798247198128,
      "language_loss": 0.85214365,
      "learning_rate": 2.368670053453015e-06,
      "loss": 0.8739512,
      "num_input_tokens_seen": 81959820,
      "step": 3807,
      "time_per_iteration": 2.563408851623535
    },
    {
      "auxiliary_loss_clip": 0.01170345,
      "auxiliary_loss_mlp": 0.01027881,
      "balance_loss_clip": 1.05289364,
      "balance_loss_mlp": 1.02008426,
      "epoch": 0.4578849275536584,
      "flos": 17418578952960.0,
      "grad_norm": 2.0851831387036346,
      "language_loss": 0.74079561,
      "learning_rate": 2.3679044027336757e-06,
      "loss": 0.76277786,
      "num_input_tokens_seen": 81975710,
      "step": 3808,
      "time_per_iteration": 2.463207483291626
    },
    {
      "auxiliary_loss_clip": 0.0117976,
      "auxiliary_loss_mlp": 0.01029225,
      "balance_loss_clip": 1.05269849,
      "balance_loss_mlp": 1.02129769,
      "epoch": 0.4580051704442975,
      "flos": 13510169107200.0,
      "grad_norm": 2.6388820831048654,
      "language_loss": 0.69075012,
      "learning_rate": 2.3671386962035326e-06,
      "loss": 0.71283996,
      "num_input_tokens_seen": 81993180,
      "step": 3809,
      "time_per_iteration": 3.9721713066101074
    },
    {
      "auxiliary_loss_clip": 0.01165379,
      "auxiliary_loss_mlp": 0.01032264,
      "balance_loss_clip": 1.0510819,
      "balance_loss_mlp": 1.02402639,
      "epoch": 0.45812541333493656,
      "flos": 18037606965120.0,
      "grad_norm": 2.0458670841621833,
      "language_loss": 0.68644786,
      "learning_rate": 2.3663729339787405e-06,
      "loss": 0.70842433,
      "num_input_tokens_seen": 82010115,
      "step": 3810,
      "time_per_iteration": 3.324791669845581
    },
    {
      "auxiliary_loss_clip": 0.01177861,
      "auxiliary_loss_mlp": 0.01033448,
      "balance_loss_clip": 1.05096436,
      "balance_loss_mlp": 1.02426934,
      "epoch": 0.45824565622557567,
      "flos": 20222196232320.0,
      "grad_norm": 2.713546743892018,
      "language_loss": 0.73684931,
      "learning_rate": 2.365607116175466e-06,
      "loss": 0.75896239,
      "num_input_tokens_seen": 82025540,
      "step": 3811,
      "time_per_iteration": 2.4307312965393066
    },
    {
      "auxiliary_loss_clip": 0.01179175,
      "auxiliary_loss_mlp": 0.01026513,
      "balance_loss_clip": 1.05358887,
      "balance_loss_mlp": 1.01863337,
      "epoch": 0.4583658991162148,
      "flos": 19864885691520.0,
      "grad_norm": 2.1698646906564405,
      "language_loss": 0.66494989,
      "learning_rate": 2.3648412429098825e-06,
      "loss": 0.68700677,
      "num_input_tokens_seen": 82043890,
      "step": 3812,
      "time_per_iteration": 2.4761080741882324
    },
    {
      "auxiliary_loss_clip": 0.01135933,
      "auxiliary_loss_mlp": 0.01040697,
      "balance_loss_clip": 1.05037785,
      "balance_loss_mlp": 1.03250158,
      "epoch": 0.45848614200685384,
      "flos": 21029935322880.0,
      "grad_norm": 1.949084677617244,
      "language_loss": 0.81928158,
      "learning_rate": 2.364075314298172e-06,
      "loss": 0.84104788,
      "num_input_tokens_seen": 82061345,
      "step": 3813,
      "time_per_iteration": 2.575054407119751
    },
    {
      "auxiliary_loss_clip": 0.01166659,
      "auxiliary_loss_mlp": 0.00761607,
      "balance_loss_clip": 1.05042815,
      "balance_loss_mlp": 1.00015831,
      "epoch": 0.45860638489749295,
      "flos": 21069293650560.0,
      "grad_norm": 1.8160267284931182,
      "language_loss": 0.70252156,
      "learning_rate": 2.3633093304565267e-06,
      "loss": 0.72180426,
      "num_input_tokens_seen": 82080400,
      "step": 3814,
      "time_per_iteration": 2.5105133056640625
    },
    {
      "auxiliary_loss_clip": 0.0118707,
      "auxiliary_loss_mlp": 0.01029752,
      "balance_loss_clip": 1.05795026,
      "balance_loss_mlp": 1.02178264,
      "epoch": 0.458726627788132,
      "flos": 26833889692800.0,
      "grad_norm": 1.8442737339484487,
      "language_loss": 0.63193285,
      "learning_rate": 2.3625432915011443e-06,
      "loss": 0.65410107,
      "num_input_tokens_seen": 82102310,
      "step": 3815,
      "time_per_iteration": 2.5119123458862305
    },
    {
      "auxiliary_loss_clip": 0.01144357,
      "auxiliary_loss_mlp": 0.0102406,
      "balance_loss_clip": 1.04796159,
      "balance_loss_mlp": 1.01630497,
      "epoch": 0.4588468706787711,
      "flos": 24097927680000.0,
      "grad_norm": 1.6874672628438991,
      "language_loss": 0.6479466,
      "learning_rate": 2.3617771975482334e-06,
      "loss": 0.66963077,
      "num_input_tokens_seen": 82121140,
      "step": 3816,
      "time_per_iteration": 2.5616648197174072
    },
    {
      "auxiliary_loss_clip": 0.01120326,
      "auxiliary_loss_mlp": 0.01029604,
      "balance_loss_clip": 1.0471251,
      "balance_loss_mlp": 1.02266622,
      "epoch": 0.4589671135694102,
      "flos": 17889331622400.0,
      "grad_norm": 1.5948200773387111,
      "language_loss": 0.7441653,
      "learning_rate": 2.3610110487140083e-06,
      "loss": 0.76566458,
      "num_input_tokens_seen": 82139575,
      "step": 3817,
      "time_per_iteration": 2.5911097526550293
    },
    {
      "auxiliary_loss_clip": 0.01148852,
      "auxiliary_loss_mlp": 0.01026335,
      "balance_loss_clip": 1.04917026,
      "balance_loss_mlp": 1.01888466,
      "epoch": 0.4590873564600493,
      "flos": 25626967781760.0,
      "grad_norm": 1.8560785538958022,
      "language_loss": 0.80775893,
      "learning_rate": 2.360244845114695e-06,
      "loss": 0.82951081,
      "num_input_tokens_seen": 82159195,
      "step": 3818,
      "time_per_iteration": 2.590060234069824
    },
    {
      "auxiliary_loss_clip": 0.01145822,
      "auxiliary_loss_mlp": 0.01026081,
      "balance_loss_clip": 1.05029237,
      "balance_loss_mlp": 1.01785541,
      "epoch": 0.4592075993506884,
      "flos": 18514788168960.0,
      "grad_norm": 2.7665287631213835,
      "language_loss": 0.68278855,
      "learning_rate": 2.3594785868665245e-06,
      "loss": 0.70450759,
      "num_input_tokens_seen": 82175500,
      "step": 3819,
      "time_per_iteration": 2.5120010375976562
    },
    {
      "auxiliary_loss_clip": 0.01138766,
      "auxiliary_loss_mlp": 0.00762228,
      "balance_loss_clip": 1.04785728,
      "balance_loss_mlp": 1.00021803,
      "epoch": 0.4593278422413275,
      "flos": 20631111638400.0,
      "grad_norm": 2.0117359522772884,
      "language_loss": 0.80365419,
      "learning_rate": 2.3587122740857386e-06,
      "loss": 0.82266414,
      "num_input_tokens_seen": 82192600,
      "step": 3820,
      "time_per_iteration": 2.579968214035034
    },
    {
      "auxiliary_loss_clip": 0.01166148,
      "auxiliary_loss_mlp": 0.0102856,
      "balance_loss_clip": 1.05194831,
      "balance_loss_mlp": 1.02125573,
      "epoch": 0.45944808513196655,
      "flos": 21358517961600.0,
      "grad_norm": 1.667297621637011,
      "language_loss": 0.7783882,
      "learning_rate": 2.357945906888586e-06,
      "loss": 0.80033535,
      "num_input_tokens_seen": 82212040,
      "step": 3821,
      "time_per_iteration": 2.5351154804229736
    },
    {
      "auxiliary_loss_clip": 0.01167951,
      "auxiliary_loss_mlp": 0.01028102,
      "balance_loss_clip": 1.05287886,
      "balance_loss_mlp": 1.0196085,
      "epoch": 0.45956832802260567,
      "flos": 21427789340160.0,
      "grad_norm": 2.1802845660241017,
      "language_loss": 0.79489684,
      "learning_rate": 2.357179485391324e-06,
      "loss": 0.81685734,
      "num_input_tokens_seen": 82229895,
      "step": 3822,
      "time_per_iteration": 2.5007171630859375
    },
    {
      "auxiliary_loss_clip": 0.01178089,
      "auxiliary_loss_mlp": 0.01035217,
      "balance_loss_clip": 1.05351627,
      "balance_loss_mlp": 1.02780795,
      "epoch": 0.4596885709132448,
      "flos": 22382654538240.0,
      "grad_norm": 1.8434412700223115,
      "language_loss": 0.86401498,
      "learning_rate": 2.3564130097102173e-06,
      "loss": 0.8861481,
      "num_input_tokens_seen": 82249550,
      "step": 3823,
      "time_per_iteration": 2.481332302093506
    },
    {
      "auxiliary_loss_clip": 0.01144941,
      "auxiliary_loss_mlp": 0.01027494,
      "balance_loss_clip": 1.05209458,
      "balance_loss_mlp": 1.01951897,
      "epoch": 0.45980881380388383,
      "flos": 28981957806720.0,
      "grad_norm": 4.074730774137933,
      "language_loss": 0.74850452,
      "learning_rate": 2.355646479961541e-06,
      "loss": 0.7702288,
      "num_input_tokens_seen": 82268860,
      "step": 3824,
      "time_per_iteration": 2.612250328063965
    },
    {
      "auxiliary_loss_clip": 0.01178564,
      "auxiliary_loss_mlp": 0.01033317,
      "balance_loss_clip": 1.05194402,
      "balance_loss_mlp": 1.02534223,
      "epoch": 0.45992905669452294,
      "flos": 33396599980800.0,
      "grad_norm": 1.878286627401379,
      "language_loss": 0.7144261,
      "learning_rate": 2.354879896261576e-06,
      "loss": 0.73654485,
      "num_input_tokens_seen": 82289070,
      "step": 3825,
      "time_per_iteration": 2.5806679725646973
    },
    {
      "auxiliary_loss_clip": 0.01135159,
      "auxiliary_loss_mlp": 0.01028494,
      "balance_loss_clip": 1.05120504,
      "balance_loss_mlp": 1.02030456,
      "epoch": 0.46004929958516205,
      "flos": 36318184502400.0,
      "grad_norm": 5.522962680050381,
      "language_loss": 0.56786686,
      "learning_rate": 2.3541132587266133e-06,
      "loss": 0.58950341,
      "num_input_tokens_seen": 82311790,
      "step": 3826,
      "time_per_iteration": 2.697087049484253
    },
    {
      "auxiliary_loss_clip": 0.01142246,
      "auxiliary_loss_mlp": 0.01034288,
      "balance_loss_clip": 1.04981399,
      "balance_loss_mlp": 1.02683115,
      "epoch": 0.4601695424758011,
      "flos": 17238451224960.0,
      "grad_norm": 1.7571933025394342,
      "language_loss": 0.69240391,
      "learning_rate": 2.3533465674729515e-06,
      "loss": 0.71416926,
      "num_input_tokens_seen": 82329020,
      "step": 3827,
      "time_per_iteration": 2.564023733139038
    },
    {
      "auxiliary_loss_clip": 0.01182661,
      "auxiliary_loss_mlp": 0.01031914,
      "balance_loss_clip": 1.05570102,
      "balance_loss_mlp": 1.02357578,
      "epoch": 0.4602897853664402,
      "flos": 15888425529600.0,
      "grad_norm": 1.9792846762446406,
      "language_loss": 0.72898108,
      "learning_rate": 2.352579822616895e-06,
      "loss": 0.75112683,
      "num_input_tokens_seen": 82346455,
      "step": 3828,
      "time_per_iteration": 2.4639105796813965
    },
    {
      "auxiliary_loss_clip": 0.01153657,
      "auxiliary_loss_mlp": 0.0102472,
      "balance_loss_clip": 1.05189013,
      "balance_loss_mlp": 1.01701283,
      "epoch": 0.4604100282570793,
      "flos": 25412617370880.0,
      "grad_norm": 1.9379010216362937,
      "language_loss": 0.77922392,
      "learning_rate": 2.351813024274761e-06,
      "loss": 0.80100763,
      "num_input_tokens_seen": 82367810,
      "step": 3829,
      "time_per_iteration": 2.583096742630005
    },
    {
      "auxiliary_loss_clip": 0.01140975,
      "auxiliary_loss_mlp": 0.0103261,
      "balance_loss_clip": 1.04955423,
      "balance_loss_mlp": 1.02446246,
      "epoch": 0.4605302711477184,
      "flos": 27630711048960.0,
      "grad_norm": 1.7621040674671256,
      "language_loss": 0.73219037,
      "learning_rate": 2.3510461725628693e-06,
      "loss": 0.75392628,
      "num_input_tokens_seen": 82388275,
      "step": 3830,
      "time_per_iteration": 3.399188280105591
    },
    {
      "auxiliary_loss_clip": 0.0114183,
      "auxiliary_loss_mlp": 0.01032528,
      "balance_loss_clip": 1.05097222,
      "balance_loss_mlp": 1.02465415,
      "epoch": 0.4606505140383575,
      "flos": 23839657914240.0,
      "grad_norm": 1.9365507423421742,
      "language_loss": 0.70671248,
      "learning_rate": 2.350279267597554e-06,
      "loss": 0.72845608,
      "num_input_tokens_seen": 82408915,
      "step": 3831,
      "time_per_iteration": 2.6219053268432617
    },
    {
      "auxiliary_loss_clip": 0.01166094,
      "auxiliary_loss_mlp": 0.01034252,
      "balance_loss_clip": 1.05276155,
      "balance_loss_mlp": 1.02593708,
      "epoch": 0.46077075692899655,
      "flos": 16107013745280.0,
      "grad_norm": 1.943939318866429,
      "language_loss": 0.82441229,
      "learning_rate": 2.3495123094951515e-06,
      "loss": 0.84641576,
      "num_input_tokens_seen": 82427260,
      "step": 3832,
      "time_per_iteration": 2.494882822036743
    },
    {
      "auxiliary_loss_clip": 0.01146752,
      "auxiliary_loss_mlp": 0.01034446,
      "balance_loss_clip": 1.05080533,
      "balance_loss_mlp": 1.02626824,
      "epoch": 0.46089099981963566,
      "flos": 48798147634560.0,
      "grad_norm": 4.90343666593536,
      "language_loss": 0.75708663,
      "learning_rate": 2.34874529837201e-06,
      "loss": 0.7788986,
      "num_input_tokens_seen": 82450805,
      "step": 3833,
      "time_per_iteration": 2.812138795852661
    },
    {
      "auxiliary_loss_clip": 0.01104341,
      "auxiliary_loss_mlp": 0.01027854,
      "balance_loss_clip": 1.04446983,
      "balance_loss_mlp": 1.02024853,
      "epoch": 0.46101124271027477,
      "flos": 19099234362240.0,
      "grad_norm": 1.9892262990228236,
      "language_loss": 0.79303271,
      "learning_rate": 2.347978234344483e-06,
      "loss": 0.81435472,
      "num_input_tokens_seen": 82467010,
      "step": 3834,
      "time_per_iteration": 2.6123430728912354
    },
    {
      "auxiliary_loss_clip": 0.01171124,
      "auxiliary_loss_mlp": 0.01038386,
      "balance_loss_clip": 1.05386639,
      "balance_loss_mlp": 1.02892077,
      "epoch": 0.4611314856009138,
      "flos": 39347931853440.0,
      "grad_norm": 1.6233061100839423,
      "language_loss": 0.68726754,
      "learning_rate": 2.347211117528935e-06,
      "loss": 0.70936263,
      "num_input_tokens_seen": 82489310,
      "step": 3835,
      "time_per_iteration": 3.4536683559417725
    },
    {
      "auxiliary_loss_clip": 0.0114966,
      "auxiliary_loss_mlp": 0.01031977,
      "balance_loss_clip": 1.05498195,
      "balance_loss_mlp": 1.02394795,
      "epoch": 0.46125172849155294,
      "flos": 20810772489600.0,
      "grad_norm": 1.5283580941011288,
      "language_loss": 0.71693909,
      "learning_rate": 2.3464439480417374e-06,
      "loss": 0.73875546,
      "num_input_tokens_seen": 82508830,
      "step": 3836,
      "time_per_iteration": 3.4179673194885254
    },
    {
      "auxiliary_loss_clip": 0.0117152,
      "auxiliary_loss_mlp": 0.01024943,
      "balance_loss_clip": 1.05468369,
      "balance_loss_mlp": 1.01671147,
      "epoch": 0.46137197138219205,
      "flos": 17930808852480.0,
      "grad_norm": 2.8021381161504975,
      "language_loss": 0.77459174,
      "learning_rate": 2.3456767259992676e-06,
      "loss": 0.79655635,
      "num_input_tokens_seen": 82526475,
      "step": 3837,
      "time_per_iteration": 2.4996180534362793
    },
    {
      "auxiliary_loss_clip": 0.01181886,
      "auxiliary_loss_mlp": 0.00761881,
      "balance_loss_clip": 1.05393863,
      "balance_loss_mlp": 1.00018942,
      "epoch": 0.4614922142728311,
      "flos": 16836610798080.0,
      "grad_norm": 2.1802875904668464,
      "language_loss": 0.88705122,
      "learning_rate": 2.3449094515179135e-06,
      "loss": 0.9064889,
      "num_input_tokens_seen": 82543935,
      "step": 3838,
      "time_per_iteration": 2.446885585784912
    },
    {
      "auxiliary_loss_clip": 0.01155503,
      "auxiliary_loss_mlp": 0.01032577,
      "balance_loss_clip": 1.04916298,
      "balance_loss_mlp": 1.02518606,
      "epoch": 0.4616124571634702,
      "flos": 26614906427520.0,
      "grad_norm": 1.6630711569477297,
      "language_loss": 0.81721914,
      "learning_rate": 2.34414212471407e-06,
      "loss": 0.83909988,
      "num_input_tokens_seen": 82563730,
      "step": 3839,
      "time_per_iteration": 2.577338695526123
    },
    {
      "auxiliary_loss_clip": 0.01172094,
      "auxiliary_loss_mlp": 0.0102852,
      "balance_loss_clip": 1.05262852,
      "balance_loss_mlp": 1.02053905,
      "epoch": 0.4617327000541093,
      "flos": 20340127560960.0,
      "grad_norm": 2.0464323185763496,
      "language_loss": 0.72511089,
      "learning_rate": 2.3433747457041394e-06,
      "loss": 0.74711698,
      "num_input_tokens_seen": 82582435,
      "step": 3840,
      "time_per_iteration": 2.4908883571624756
    },
    {
      "auxiliary_loss_clip": 0.01136729,
      "auxiliary_loss_mlp": 0.01031925,
      "balance_loss_clip": 1.05061829,
      "balance_loss_mlp": 1.02337766,
      "epoch": 0.4618529429447484,
      "flos": 29570749545600.0,
      "grad_norm": 1.8362775405161145,
      "language_loss": 0.8485496,
      "learning_rate": 2.342607314604533e-06,
      "loss": 0.87023616,
      "num_input_tokens_seen": 82602185,
      "step": 3841,
      "time_per_iteration": 2.6505284309387207
    },
    {
      "auxiliary_loss_clip": 0.01164147,
      "auxiliary_loss_mlp": 0.01037053,
      "balance_loss_clip": 1.05355704,
      "balance_loss_mlp": 1.02855885,
      "epoch": 0.4619731858353875,
      "flos": 19787030962560.0,
      "grad_norm": 2.019988810713038,
      "language_loss": 0.83939052,
      "learning_rate": 2.3418398315316694e-06,
      "loss": 0.86140245,
      "num_input_tokens_seen": 82620005,
      "step": 3842,
      "time_per_iteration": 2.5004162788391113
    },
    {
      "auxiliary_loss_clip": 0.01180829,
      "auxiliary_loss_mlp": 0.01031336,
      "balance_loss_clip": 1.05473757,
      "balance_loss_mlp": 1.02297378,
      "epoch": 0.4620934287260266,
      "flos": 18951138587520.0,
      "grad_norm": 2.2762700472927446,
      "language_loss": 0.78476191,
      "learning_rate": 2.3410722966019755e-06,
      "loss": 0.80688357,
      "num_input_tokens_seen": 82635120,
      "step": 3843,
      "time_per_iteration": 2.461857557296753
    },
    {
      "auxiliary_loss_clip": 0.0116631,
      "auxiliary_loss_mlp": 0.01030328,
      "balance_loss_clip": 1.05304205,
      "balance_loss_mlp": 1.02240109,
      "epoch": 0.46221367161666566,
      "flos": 37341674634240.0,
      "grad_norm": 1.6584498640583603,
      "language_loss": 0.65746927,
      "learning_rate": 2.3403047099318848e-06,
      "loss": 0.67943561,
      "num_input_tokens_seen": 82659190,
      "step": 3844,
      "time_per_iteration": 2.643996000289917
    },
    {
      "auxiliary_loss_clip": 0.01118465,
      "auxiliary_loss_mlp": 0.01030928,
      "balance_loss_clip": 1.04554474,
      "balance_loss_mlp": 1.0228399,
      "epoch": 0.46233391450730477,
      "flos": 14428549065600.0,
      "grad_norm": 2.1344106447397198,
      "language_loss": 0.75169057,
      "learning_rate": 2.3395370716378405e-06,
      "loss": 0.77318454,
      "num_input_tokens_seen": 82676635,
      "step": 3845,
      "time_per_iteration": 2.60945987701416
    },
    {
      "auxiliary_loss_clip": 0.01166518,
      "auxiliary_loss_mlp": 0.01029116,
      "balance_loss_clip": 1.05021822,
      "balance_loss_mlp": 1.02145708,
      "epoch": 0.4624541573979438,
      "flos": 22493044010880.0,
      "grad_norm": 2.0350382128080815,
      "language_loss": 0.71876955,
      "learning_rate": 2.338769381836292e-06,
      "loss": 0.74072593,
      "num_input_tokens_seen": 82696245,
      "step": 3846,
      "time_per_iteration": 2.5322506427764893
    },
    {
      "auxiliary_loss_clip": 0.01136035,
      "auxiliary_loss_mlp": 0.0103327,
      "balance_loss_clip": 1.05113339,
      "balance_loss_mlp": 1.02528572,
      "epoch": 0.46257440028858293,
      "flos": 14465070218880.0,
      "grad_norm": 1.9842742541243525,
      "language_loss": 0.73507619,
      "learning_rate": 2.3380016406436984e-06,
      "loss": 0.75676924,
      "num_input_tokens_seen": 82713725,
      "step": 3847,
      "time_per_iteration": 2.5452163219451904
    },
    {
      "auxiliary_loss_clip": 0.01122415,
      "auxiliary_loss_mlp": 0.01027074,
      "balance_loss_clip": 1.05018234,
      "balance_loss_mlp": 1.01945007,
      "epoch": 0.46269464317922204,
      "flos": 23332204523520.0,
      "grad_norm": 1.854488492573768,
      "language_loss": 0.81254613,
      "learning_rate": 2.337233848176524e-06,
      "loss": 0.83404106,
      "num_input_tokens_seen": 82731495,
      "step": 3848,
      "time_per_iteration": 2.6263160705566406
    },
    {
      "auxiliary_loss_clip": 0.01114858,
      "auxiliary_loss_mlp": 0.01029851,
      "balance_loss_clip": 1.04478276,
      "balance_loss_mlp": 1.02151215,
      "epoch": 0.4628148860698611,
      "flos": 18552027594240.0,
      "grad_norm": 3.887915044141574,
      "language_loss": 0.83574039,
      "learning_rate": 2.3364660045512435e-06,
      "loss": 0.85718751,
      "num_input_tokens_seen": 82750255,
      "step": 3849,
      "time_per_iteration": 2.5974502563476562
    },
    {
      "auxiliary_loss_clip": 0.01047175,
      "auxiliary_loss_mlp": 0.01003698,
      "balance_loss_clip": 1.01627493,
      "balance_loss_mlp": 1.00236845,
      "epoch": 0.4629351289605002,
      "flos": 70667569670400.0,
      "grad_norm": 0.735644602631247,
      "language_loss": 0.58212018,
      "learning_rate": 2.335698109884337e-06,
      "loss": 0.60262895,
      "num_input_tokens_seen": 82815460,
      "step": 3850,
      "time_per_iteration": 3.263976573944092
    },
    {
      "auxiliary_loss_clip": 0.01028483,
      "auxiliary_loss_mlp": 0.01002967,
      "balance_loss_clip": 1.01904535,
      "balance_loss_mlp": 1.00165558,
      "epoch": 0.4630553718511393,
      "flos": 59687200465920.0,
      "grad_norm": 0.7835813656838914,
      "language_loss": 0.59870625,
      "learning_rate": 2.334930164292294e-06,
      "loss": 0.61902076,
      "num_input_tokens_seen": 82878010,
      "step": 3851,
      "time_per_iteration": 3.2982001304626465
    },
    {
      "auxiliary_loss_clip": 0.01120005,
      "auxiliary_loss_mlp": 0.01028392,
      "balance_loss_clip": 1.04660332,
      "balance_loss_mlp": 1.02034557,
      "epoch": 0.4631756147417784,
      "flos": 15960605909760.0,
      "grad_norm": 2.6162478676432075,
      "language_loss": 0.80263382,
      "learning_rate": 2.334162167891612e-06,
      "loss": 0.82411778,
      "num_input_tokens_seen": 82895275,
      "step": 3852,
      "time_per_iteration": 2.5861377716064453
    },
    {
      "auxiliary_loss_clip": 0.01154361,
      "auxiliary_loss_mlp": 0.01027958,
      "balance_loss_clip": 1.04895639,
      "balance_loss_mlp": 1.01970232,
      "epoch": 0.4632958576324175,
      "flos": 16472907636480.0,
      "grad_norm": 2.020150749778654,
      "language_loss": 0.7513212,
      "learning_rate": 2.333394120798795e-06,
      "loss": 0.77314442,
      "num_input_tokens_seen": 82914010,
      "step": 3853,
      "time_per_iteration": 2.533970832824707
    },
    {
      "auxiliary_loss_clip": 0.01154735,
      "auxiliary_loss_mlp": 0.01030445,
      "balance_loss_clip": 1.04984856,
      "balance_loss_mlp": 1.02265453,
      "epoch": 0.4634161005230566,
      "flos": 22346492520960.0,
      "grad_norm": 2.9327089702818276,
      "language_loss": 0.72339869,
      "learning_rate": 2.3326260231303545e-06,
      "loss": 0.74525052,
      "num_input_tokens_seen": 82932610,
      "step": 3854,
      "time_per_iteration": 2.5625243186950684
    },
    {
      "auxiliary_loss_clip": 0.01181348,
      "auxiliary_loss_mlp": 0.01029292,
      "balance_loss_clip": 1.05629075,
      "balance_loss_mlp": 1.02116251,
      "epoch": 0.46353634341369565,
      "flos": 15742233175680.0,
      "grad_norm": 1.6820933465103396,
      "language_loss": 0.86562353,
      "learning_rate": 2.331857875002811e-06,
      "loss": 0.88772988,
      "num_input_tokens_seen": 82951210,
      "step": 3855,
      "time_per_iteration": 3.26835298538208
    },
    {
      "auxiliary_loss_clip": 0.01156371,
      "auxiliary_loss_mlp": 0.01037177,
      "balance_loss_clip": 1.05407727,
      "balance_loss_mlp": 1.02972639,
      "epoch": 0.46365658630433476,
      "flos": 28329820433280.0,
      "grad_norm": 1.6209306286721221,
      "language_loss": 0.76182193,
      "learning_rate": 2.3310896765326916e-06,
      "loss": 0.78375739,
      "num_input_tokens_seen": 82972210,
      "step": 3856,
      "time_per_iteration": 2.6177046298980713
    },
    {
      "auxiliary_loss_clip": 0.01132306,
      "auxiliary_loss_mlp": 0.01032283,
      "balance_loss_clip": 1.04655886,
      "balance_loss_mlp": 1.02448034,
      "epoch": 0.46377682919497387,
      "flos": 24608074590720.0,
      "grad_norm": 1.7482383253248859,
      "language_loss": 0.83774626,
      "learning_rate": 2.330321427836531e-06,
      "loss": 0.85939217,
      "num_input_tokens_seen": 82994080,
      "step": 3857,
      "time_per_iteration": 2.668135643005371
    },
    {
      "auxiliary_loss_clip": 0.01165068,
      "auxiliary_loss_mlp": 0.01026937,
      "balance_loss_clip": 1.0520407,
      "balance_loss_mlp": 1.01901591,
      "epoch": 0.4638970720856129,
      "flos": 19060953442560.0,
      "grad_norm": 1.7030701615808672,
      "language_loss": 0.82673222,
      "learning_rate": 2.3295531290308733e-06,
      "loss": 0.8486523,
      "num_input_tokens_seen": 83012230,
      "step": 3858,
      "time_per_iteration": 2.5214805603027344
    },
    {
      "auxiliary_loss_clip": 0.01184065,
      "auxiliary_loss_mlp": 0.00762034,
      "balance_loss_clip": 1.05560243,
      "balance_loss_mlp": 1.00020409,
      "epoch": 0.46401731497625204,
      "flos": 18471012468480.0,
      "grad_norm": 3.061230269254775,
      "language_loss": 0.7583459,
      "learning_rate": 2.3287847802322678e-06,
      "loss": 0.77780688,
      "num_input_tokens_seen": 83027800,
      "step": 3859,
      "time_per_iteration": 2.4616923332214355
    },
    {
      "auxiliary_loss_clip": 0.01156663,
      "auxiliary_loss_mlp": 0.01032126,
      "balance_loss_clip": 1.05344009,
      "balance_loss_mlp": 1.02259493,
      "epoch": 0.4641375578668911,
      "flos": 26067053214720.0,
      "grad_norm": 1.6102642622424008,
      "language_loss": 0.84160447,
      "learning_rate": 2.3280163815572723e-06,
      "loss": 0.86349237,
      "num_input_tokens_seen": 83048395,
      "step": 3860,
      "time_per_iteration": 2.575424909591675
    },
    {
      "auxiliary_loss_clip": 0.01145241,
      "auxiliary_loss_mlp": 0.01035803,
      "balance_loss_clip": 1.05088365,
      "balance_loss_mlp": 1.02759552,
      "epoch": 0.4642578007575302,
      "flos": 19570382081280.0,
      "grad_norm": 1.9479080923588787,
      "language_loss": 0.77277839,
      "learning_rate": 2.3272479331224522e-06,
      "loss": 0.7945888,
      "num_input_tokens_seen": 83065825,
      "step": 3861,
      "time_per_iteration": 4.074430227279663
    },
    {
      "auxiliary_loss_clip": 0.01184241,
      "auxiliary_loss_mlp": 0.0102881,
      "balance_loss_clip": 1.0551939,
      "balance_loss_mlp": 1.02019763,
      "epoch": 0.4643780436481693,
      "flos": 28186249772160.0,
      "grad_norm": 1.5040459854793349,
      "language_loss": 0.77775913,
      "learning_rate": 2.3264794350443817e-06,
      "loss": 0.79988968,
      "num_input_tokens_seen": 83087920,
      "step": 3862,
      "time_per_iteration": 3.3948593139648438
    },
    {
      "auxiliary_loss_clip": 0.01166774,
      "auxiliary_loss_mlp": 0.0102899,
      "balance_loss_clip": 1.04932857,
      "balance_loss_mlp": 1.02095485,
      "epoch": 0.46449828653880837,
      "flos": 25375270204800.0,
      "grad_norm": 1.8656347976096819,
      "language_loss": 0.78355986,
      "learning_rate": 2.3257108874396396e-06,
      "loss": 0.80551744,
      "num_input_tokens_seen": 83109015,
      "step": 3863,
      "time_per_iteration": 2.559690237045288
    },
    {
      "auxiliary_loss_clip": 0.01153099,
      "auxiliary_loss_mlp": 0.01030471,
      "balance_loss_clip": 1.0503372,
      "balance_loss_mlp": 1.02258539,
      "epoch": 0.4646185294294475,
      "flos": 16034330574720.0,
      "grad_norm": 1.9078937896235535,
      "language_loss": 0.73988312,
      "learning_rate": 2.3249422904248152e-06,
      "loss": 0.76171887,
      "num_input_tokens_seen": 83127450,
      "step": 3864,
      "time_per_iteration": 2.539729356765747
    },
    {
      "auxiliary_loss_clip": 0.01170326,
      "auxiliary_loss_mlp": 0.01031704,
      "balance_loss_clip": 1.05176425,
      "balance_loss_mlp": 1.02340722,
      "epoch": 0.4647387723200866,
      "flos": 26363101109760.0,
      "grad_norm": 1.396668412005556,
      "language_loss": 0.87376034,
      "learning_rate": 2.324173644116504e-06,
      "loss": 0.89578062,
      "num_input_tokens_seen": 83150300,
      "step": 3865,
      "time_per_iteration": 2.576395273208618
    },
    {
      "auxiliary_loss_clip": 0.01166155,
      "auxiliary_loss_mlp": 0.01032706,
      "balance_loss_clip": 1.05370855,
      "balance_loss_mlp": 1.02443874,
      "epoch": 0.46485901521072565,
      "flos": 27160209774720.0,
      "grad_norm": 1.6024242067384387,
      "language_loss": 0.8130765,
      "learning_rate": 2.3234049486313087e-06,
      "loss": 0.83506507,
      "num_input_tokens_seen": 83171750,
      "step": 3866,
      "time_per_iteration": 2.5759479999542236
    },
    {
      "auxiliary_loss_clip": 0.01169408,
      "auxiliary_loss_mlp": 0.01030055,
      "balance_loss_clip": 1.05458808,
      "balance_loss_mlp": 1.02262187,
      "epoch": 0.46497925810136476,
      "flos": 24279851088000.0,
      "grad_norm": 1.6703055656197032,
      "language_loss": 0.75951242,
      "learning_rate": 2.322636204085839e-06,
      "loss": 0.78150707,
      "num_input_tokens_seen": 83191820,
      "step": 3867,
      "time_per_iteration": 2.6413400173187256
    },
    {
      "auxiliary_loss_clip": 0.01145257,
      "auxiliary_loss_mlp": 0.01033367,
      "balance_loss_clip": 1.04784262,
      "balance_loss_mlp": 1.02516568,
      "epoch": 0.46509950099200387,
      "flos": 16253134272000.0,
      "grad_norm": 2.4902576683788293,
      "language_loss": 0.78843534,
      "learning_rate": 2.3218674105967143e-06,
      "loss": 0.81022155,
      "num_input_tokens_seen": 83210085,
      "step": 3868,
      "time_per_iteration": 2.5719077587127686
    },
    {
      "auxiliary_loss_clip": 0.01145195,
      "auxiliary_loss_mlp": 0.01027638,
      "balance_loss_clip": 1.04978371,
      "balance_loss_mlp": 1.01969254,
      "epoch": 0.4652197438826429,
      "flos": 23442270773760.0,
      "grad_norm": 1.571348074885005,
      "language_loss": 0.83407432,
      "learning_rate": 2.3210985682805593e-06,
      "loss": 0.85580266,
      "num_input_tokens_seen": 83231865,
      "step": 3869,
      "time_per_iteration": 2.604243516921997
    },
    {
      "auxiliary_loss_clip": 0.01180655,
      "auxiliary_loss_mlp": 0.01032471,
      "balance_loss_clip": 1.05503464,
      "balance_loss_mlp": 1.02449846,
      "epoch": 0.46533998677328203,
      "flos": 16216397637120.0,
      "grad_norm": 2.288223247883593,
      "language_loss": 0.67969233,
      "learning_rate": 2.320329677254007e-06,
      "loss": 0.70182353,
      "num_input_tokens_seen": 83249195,
      "step": 3870,
      "time_per_iteration": 2.44983172416687
    },
    {
      "auxiliary_loss_clip": 0.01182752,
      "auxiliary_loss_mlp": 0.01028868,
      "balance_loss_clip": 1.05563772,
      "balance_loss_mlp": 1.02067196,
      "epoch": 0.46546022966392114,
      "flos": 21141869080320.0,
      "grad_norm": 2.373526451628856,
      "language_loss": 0.72720945,
      "learning_rate": 2.319560737633697e-06,
      "loss": 0.74932563,
      "num_input_tokens_seen": 83267915,
      "step": 3871,
      "time_per_iteration": 2.4718477725982666
    },
    {
      "auxiliary_loss_clip": 0.01142748,
      "auxiliary_loss_mlp": 0.01034066,
      "balance_loss_clip": 1.04659867,
      "balance_loss_mlp": 1.02585256,
      "epoch": 0.4655804725545602,
      "flos": 41171942442240.0,
      "grad_norm": 1.8333824291975287,
      "language_loss": 0.67830509,
      "learning_rate": 2.3187917495362775e-06,
      "loss": 0.70007324,
      "num_input_tokens_seen": 83292325,
      "step": 3872,
      "time_per_iteration": 2.7449910640716553
    },
    {
      "auxiliary_loss_clip": 0.01123195,
      "auxiliary_loss_mlp": 0.01038279,
      "balance_loss_clip": 1.04899788,
      "balance_loss_mlp": 1.0303638,
      "epoch": 0.4657007154451993,
      "flos": 19570956698880.0,
      "grad_norm": 2.718375067912322,
      "language_loss": 0.76879859,
      "learning_rate": 2.318022713078403e-06,
      "loss": 0.79041338,
      "num_input_tokens_seen": 83306905,
      "step": 3873,
      "time_per_iteration": 2.5777699947357178
    },
    {
      "auxiliary_loss_clip": 0.01149723,
      "auxiliary_loss_mlp": 0.01029197,
      "balance_loss_clip": 1.04953873,
      "balance_loss_mlp": 1.02126372,
      "epoch": 0.4658209583358384,
      "flos": 15517826956800.0,
      "grad_norm": 2.2555585599478487,
      "language_loss": 0.85335803,
      "learning_rate": 2.3172536283767354e-06,
      "loss": 0.87514722,
      "num_input_tokens_seen": 83320665,
      "step": 3874,
      "time_per_iteration": 2.503894329071045
    },
    {
      "auxiliary_loss_clip": 0.01134832,
      "auxiliary_loss_mlp": 0.01032194,
      "balance_loss_clip": 1.05067825,
      "balance_loss_mlp": 1.02386713,
      "epoch": 0.4659412012264775,
      "flos": 14903180403840.0,
      "grad_norm": 2.2452279364130336,
      "language_loss": 0.81134903,
      "learning_rate": 2.3164844955479447e-06,
      "loss": 0.83301926,
      "num_input_tokens_seen": 83336475,
      "step": 3875,
      "time_per_iteration": 2.534827709197998
    },
    {
      "auxiliary_loss_clip": 0.01132594,
      "auxiliary_loss_mlp": 0.01027273,
      "balance_loss_clip": 1.04896784,
      "balance_loss_mlp": 1.01851118,
      "epoch": 0.4660614441171166,
      "flos": 24425612478720.0,
      "grad_norm": 1.7910436498652131,
      "language_loss": 0.70575571,
      "learning_rate": 2.3157153147087082e-06,
      "loss": 0.72735441,
      "num_input_tokens_seen": 83358365,
      "step": 3876,
      "time_per_iteration": 2.6902196407318115
    },
    {
      "auxiliary_loss_clip": 0.01133269,
      "auxiliary_loss_mlp": 0.01034404,
      "balance_loss_clip": 1.05277824,
      "balance_loss_mlp": 1.02636063,
      "epoch": 0.46618168700775564,
      "flos": 22091095843200.0,
      "grad_norm": 1.7588504894944452,
      "language_loss": 0.83163708,
      "learning_rate": 2.314946085975709e-06,
      "loss": 0.8533138,
      "num_input_tokens_seen": 83377345,
      "step": 3877,
      "time_per_iteration": 2.6133341789245605
    },
    {
      "auxiliary_loss_clip": 0.01128548,
      "auxiliary_loss_mlp": 0.01025281,
      "balance_loss_clip": 1.04870522,
      "balance_loss_mlp": 1.01761556,
      "epoch": 0.46630192989839475,
      "flos": 26176975810560.0,
      "grad_norm": 2.119438216964303,
      "language_loss": 0.82493347,
      "learning_rate": 2.3141768094656393e-06,
      "loss": 0.84647179,
      "num_input_tokens_seen": 83395920,
      "step": 3878,
      "time_per_iteration": 2.608243942260742
    },
    {
      "auxiliary_loss_clip": 0.01102321,
      "auxiliary_loss_mlp": 0.01030093,
      "balance_loss_clip": 1.04278803,
      "balance_loss_mlp": 1.02242756,
      "epoch": 0.46642217278903386,
      "flos": 11509622150400.0,
      "grad_norm": 2.222217430583525,
      "language_loss": 0.83117402,
      "learning_rate": 2.3134074852951966e-06,
      "loss": 0.85249817,
      "num_input_tokens_seen": 83412510,
      "step": 3879,
      "time_per_iteration": 2.6582043170928955
    },
    {
      "auxiliary_loss_clip": 0.01119769,
      "auxiliary_loss_mlp": 0.01027718,
      "balance_loss_clip": 1.04538953,
      "balance_loss_mlp": 1.01982975,
      "epoch": 0.4665424156796729,
      "flos": 32306819299200.0,
      "grad_norm": 1.6155251800985762,
      "language_loss": 0.77568078,
      "learning_rate": 2.312638113581088e-06,
      "loss": 0.79715562,
      "num_input_tokens_seen": 83432995,
      "step": 3880,
      "time_per_iteration": 2.7272143363952637
    },
    {
      "auxiliary_loss_clip": 0.01166962,
      "auxiliary_loss_mlp": 0.01033986,
      "balance_loss_clip": 1.05121613,
      "balance_loss_mlp": 1.02550411,
      "epoch": 0.46666265857031203,
      "flos": 18436179254400.0,
      "grad_norm": 2.3547796064904456,
      "language_loss": 0.78509176,
      "learning_rate": 2.311868694440027e-06,
      "loss": 0.80710119,
      "num_input_tokens_seen": 83447415,
      "step": 3881,
      "time_per_iteration": 3.337514638900757
    },
    {
      "auxiliary_loss_clip": 0.01070161,
      "auxiliary_loss_mlp": 0.0100918,
      "balance_loss_clip": 1.01617968,
      "balance_loss_mlp": 1.00783348,
      "epoch": 0.46678290146095114,
      "flos": 68438989221120.0,
      "grad_norm": 0.727560183523074,
      "language_loss": 0.62505478,
      "learning_rate": 2.3110992279887323e-06,
      "loss": 0.64584816,
      "num_input_tokens_seen": 83519340,
      "step": 3882,
      "time_per_iteration": 3.198573112487793
    },
    {
      "auxiliary_loss_clip": 0.01142617,
      "auxiliary_loss_mlp": 0.0102913,
      "balance_loss_clip": 1.04973936,
      "balance_loss_mlp": 1.0206542,
      "epoch": 0.4669031443515902,
      "flos": 17712507945600.0,
      "grad_norm": 2.1322390358433614,
      "language_loss": 0.84469318,
      "learning_rate": 2.310329714343932e-06,
      "loss": 0.86641067,
      "num_input_tokens_seen": 83535490,
      "step": 3883,
      "time_per_iteration": 2.5484302043914795
    },
    {
      "auxiliary_loss_clip": 0.01148289,
      "auxiliary_loss_mlp": 0.01030643,
      "balance_loss_clip": 1.05179524,
      "balance_loss_mlp": 1.02249503,
      "epoch": 0.4670233872422293,
      "flos": 23947748916480.0,
      "grad_norm": 1.8761355265439668,
      "language_loss": 0.82099986,
      "learning_rate": 2.309560153622361e-06,
      "loss": 0.84278911,
      "num_input_tokens_seen": 83552400,
      "step": 3884,
      "time_per_iteration": 2.542273759841919
    },
    {
      "auxiliary_loss_clip": 0.01141768,
      "auxiliary_loss_mlp": 0.01035226,
      "balance_loss_clip": 1.05225778,
      "balance_loss_mlp": 1.02674401,
      "epoch": 0.4671436301328684,
      "flos": 28111268131200.0,
      "grad_norm": 2.006038674657599,
      "language_loss": 0.74236071,
      "learning_rate": 2.3087905459407602e-06,
      "loss": 0.76413071,
      "num_input_tokens_seen": 83571340,
      "step": 3885,
      "time_per_iteration": 2.665985107421875
    },
    {
      "auxiliary_loss_clip": 0.01059669,
      "auxiliary_loss_mlp": 0.01004384,
      "balance_loss_clip": 1.01644313,
      "balance_loss_mlp": 1.00301266,
      "epoch": 0.46726387302350747,
      "flos": 69369684566400.0,
      "grad_norm": 0.7928181832434987,
      "language_loss": 0.62891114,
      "learning_rate": 2.3080208914158795e-06,
      "loss": 0.64955163,
      "num_input_tokens_seen": 83634340,
      "step": 3886,
      "time_per_iteration": 3.1286234855651855
    },
    {
      "auxiliary_loss_clip": 0.01148523,
      "auxiliary_loss_mlp": 0.01031732,
      "balance_loss_clip": 1.0526526,
      "balance_loss_mlp": 1.02390289,
      "epoch": 0.4673841159141466,
      "flos": 25519666878720.0,
      "grad_norm": 2.0306475785040248,
      "language_loss": 0.7225371,
      "learning_rate": 2.3072511901644753e-06,
      "loss": 0.74433964,
      "num_input_tokens_seen": 83653410,
      "step": 3887,
      "time_per_iteration": 4.093341827392578
    },
    {
      "auxiliary_loss_clip": 0.01181802,
      "auxiliary_loss_mlp": 0.01028334,
      "balance_loss_clip": 1.05630398,
      "balance_loss_mlp": 1.02094936,
      "epoch": 0.4675043588047857,
      "flos": 24499265316480.0,
      "grad_norm": 2.2318956981478046,
      "language_loss": 0.81010908,
      "learning_rate": 2.306481442303309e-06,
      "loss": 0.83221042,
      "num_input_tokens_seen": 83672985,
      "step": 3888,
      "time_per_iteration": 3.3080251216888428
    },
    {
      "auxiliary_loss_clip": 0.01166338,
      "auxiliary_loss_mlp": 0.01033419,
      "balance_loss_clip": 1.05092645,
      "balance_loss_mlp": 1.02476466,
      "epoch": 0.46762460169542475,
      "flos": 20960771685120.0,
      "grad_norm": 1.694845992512799,
      "language_loss": 0.73154181,
      "learning_rate": 2.3057116479491515e-06,
      "loss": 0.75353932,
      "num_input_tokens_seen": 83692395,
      "step": 3889,
      "time_per_iteration": 2.518341064453125
    },
    {
      "auxiliary_loss_clip": 0.01163209,
      "auxiliary_loss_mlp": 0.0103255,
      "balance_loss_clip": 1.04902935,
      "balance_loss_mlp": 1.02393103,
      "epoch": 0.46774484458606386,
      "flos": 19171666137600.0,
      "grad_norm": 1.8973026362773473,
      "language_loss": 0.76272416,
      "learning_rate": 2.30494180721878e-06,
      "loss": 0.78468168,
      "num_input_tokens_seen": 83709735,
      "step": 3890,
      "time_per_iteration": 2.543682098388672
    },
    {
      "auxiliary_loss_clip": 0.01164175,
      "auxiliary_loss_mlp": 0.01028183,
      "balance_loss_clip": 1.05160105,
      "balance_loss_mlp": 1.01975513,
      "epoch": 0.4678650874767029,
      "flos": 17967689141760.0,
      "grad_norm": 1.740823265077849,
      "language_loss": 0.89593226,
      "learning_rate": 2.3041719202289794e-06,
      "loss": 0.91785592,
      "num_input_tokens_seen": 83725910,
      "step": 3891,
      "time_per_iteration": 2.483513832092285
    },
    {
      "auxiliary_loss_clip": 0.01169813,
      "auxiliary_loss_mlp": 0.01034545,
      "balance_loss_clip": 1.05443358,
      "balance_loss_mlp": 1.02679658,
      "epoch": 0.467985330367342,
      "flos": 21360816432000.0,
      "grad_norm": 1.713767806777814,
      "language_loss": 0.80195999,
      "learning_rate": 2.30340198709654e-06,
      "loss": 0.82400358,
      "num_input_tokens_seen": 83745745,
      "step": 3892,
      "time_per_iteration": 2.5282087326049805
    },
    {
      "auxiliary_loss_clip": 0.01154199,
      "auxiliary_loss_mlp": 0.01031831,
      "balance_loss_clip": 1.04837322,
      "balance_loss_mlp": 1.02406406,
      "epoch": 0.46810557325798113,
      "flos": 20521835487360.0,
      "grad_norm": 1.8396250235333085,
      "language_loss": 0.74722648,
      "learning_rate": 2.3026320079382605e-06,
      "loss": 0.76908678,
      "num_input_tokens_seen": 83762680,
      "step": 3893,
      "time_per_iteration": 2.5444350242614746
    },
    {
      "auxiliary_loss_clip": 0.01180976,
      "auxiliary_loss_mlp": 0.01028823,
      "balance_loss_clip": 1.05601525,
      "balance_loss_mlp": 1.02129221,
      "epoch": 0.4682258161486202,
      "flos": 30117848572800.0,
      "grad_norm": 1.9798895464621107,
      "language_loss": 0.76180089,
      "learning_rate": 2.3018619828709454e-06,
      "loss": 0.78389883,
      "num_input_tokens_seen": 83784220,
      "step": 3894,
      "time_per_iteration": 2.5717382431030273
    },
    {
      "auxiliary_loss_clip": 0.01165505,
      "auxiliary_loss_mlp": 0.00761788,
      "balance_loss_clip": 1.0550313,
      "balance_loss_mlp": 1.00018752,
      "epoch": 0.4683460590392593,
      "flos": 25293357239040.0,
      "grad_norm": 1.8050647033380942,
      "language_loss": 0.82000065,
      "learning_rate": 2.3010919120114084e-06,
      "loss": 0.83927357,
      "num_input_tokens_seen": 83800750,
      "step": 3895,
      "time_per_iteration": 2.52693772315979
    },
    {
      "auxiliary_loss_clip": 0.01161918,
      "auxiliary_loss_mlp": 0.01026811,
      "balance_loss_clip": 1.0485121,
      "balance_loss_mlp": 1.01837659,
      "epoch": 0.4684663019298984,
      "flos": 15368330551680.0,
      "grad_norm": 2.0693518644685294,
      "language_loss": 0.66208112,
      "learning_rate": 2.3003217954764672e-06,
      "loss": 0.68396848,
      "num_input_tokens_seen": 83815455,
      "step": 3896,
      "time_per_iteration": 2.482342481613159
    },
    {
      "auxiliary_loss_clip": 0.01165969,
      "auxiliary_loss_mlp": 0.01030838,
      "balance_loss_clip": 1.04876697,
      "balance_loss_mlp": 1.02214134,
      "epoch": 0.46858654482053747,
      "flos": 27778842737280.0,
      "grad_norm": 1.59640223895007,
      "language_loss": 0.78956497,
      "learning_rate": 2.299551633382949e-06,
      "loss": 0.81153297,
      "num_input_tokens_seen": 83835765,
      "step": 3897,
      "time_per_iteration": 2.5802297592163086
    },
    {
      "auxiliary_loss_clip": 0.01146702,
      "auxiliary_loss_mlp": 0.01039033,
      "balance_loss_clip": 1.04998469,
      "balance_loss_mlp": 1.03057492,
      "epoch": 0.4687067877111766,
      "flos": 18040623707520.0,
      "grad_norm": 1.8357212993738075,
      "language_loss": 0.85143244,
      "learning_rate": 2.2987814258476854e-06,
      "loss": 0.87328982,
      "num_input_tokens_seen": 83853565,
      "step": 3898,
      "time_per_iteration": 2.531841516494751
    },
    {
      "auxiliary_loss_clip": 0.0112491,
      "auxiliary_loss_mlp": 0.01034142,
      "balance_loss_clip": 1.04381132,
      "balance_loss_mlp": 1.02574348,
      "epoch": 0.4688270306018157,
      "flos": 16977380198400.0,
      "grad_norm": 2.217319480354246,
      "language_loss": 0.68041766,
      "learning_rate": 2.2980111729875177e-06,
      "loss": 0.70200813,
      "num_input_tokens_seen": 83869815,
      "step": 3899,
      "time_per_iteration": 2.5848748683929443
    },
    {
      "auxiliary_loss_clip": 0.01149228,
      "auxiliary_loss_mlp": 0.01026652,
      "balance_loss_clip": 1.0512588,
      "balance_loss_mlp": 1.01851654,
      "epoch": 0.46894727349245474,
      "flos": 17821640442240.0,
      "grad_norm": 1.9478627951223486,
      "language_loss": 0.82462668,
      "learning_rate": 2.2972408749192917e-06,
      "loss": 0.84638548,
      "num_input_tokens_seen": 83887545,
      "step": 3900,
      "time_per_iteration": 2.517995595932007
    },
    {
      "auxiliary_loss_clip": 0.01166631,
      "auxiliary_loss_mlp": 0.00762115,
      "balance_loss_clip": 1.05607009,
      "balance_loss_mlp": 1.00016665,
      "epoch": 0.46906751638309385,
      "flos": 21471349559040.0,
      "grad_norm": 1.6827975520367686,
      "language_loss": 0.67046928,
      "learning_rate": 2.296470531759861e-06,
      "loss": 0.68975675,
      "num_input_tokens_seen": 83905645,
      "step": 3901,
      "time_per_iteration": 2.530261278152466
    },
    {
      "auxiliary_loss_clip": 0.01130221,
      "auxiliary_loss_mlp": 0.01028318,
      "balance_loss_clip": 1.04549503,
      "balance_loss_mlp": 1.01994991,
      "epoch": 0.46918775927373296,
      "flos": 20337829090560.0,
      "grad_norm": 2.7352540610051905,
      "language_loss": 0.7970295,
      "learning_rate": 2.2957001436260866e-06,
      "loss": 0.81861484,
      "num_input_tokens_seen": 83922705,
      "step": 3902,
      "time_per_iteration": 2.5584282875061035
    },
    {
      "auxiliary_loss_clip": 0.01150579,
      "auxiliary_loss_mlp": 0.01030141,
      "balance_loss_clip": 1.05079937,
      "balance_loss_mlp": 1.02219021,
      "epoch": 0.469308002164372,
      "flos": 18403249461120.0,
      "grad_norm": 1.6243092819596887,
      "language_loss": 0.73037952,
      "learning_rate": 2.294929710634836e-06,
      "loss": 0.75218666,
      "num_input_tokens_seen": 83940795,
      "step": 3903,
      "time_per_iteration": 2.5465402603149414
    },
    {
      "auxiliary_loss_clip": 0.01162254,
      "auxiliary_loss_mlp": 0.01030381,
      "balance_loss_clip": 1.04742312,
      "balance_loss_mlp": 1.02244198,
      "epoch": 0.46942824505501113,
      "flos": 37962067363200.0,
      "grad_norm": 3.567047069823146,
      "language_loss": 0.61552107,
      "learning_rate": 2.2941592329029823e-06,
      "loss": 0.63744736,
      "num_input_tokens_seen": 83961900,
      "step": 3904,
      "time_per_iteration": 2.6581130027770996
    },
    {
      "auxiliary_loss_clip": 0.01158489,
      "auxiliary_loss_mlp": 0.01033347,
      "balance_loss_clip": 1.04820371,
      "balance_loss_mlp": 1.02488303,
      "epoch": 0.46954848794565024,
      "flos": 21872507627520.0,
      "grad_norm": 1.838903502739008,
      "language_loss": 0.78900576,
      "learning_rate": 2.2933887105474067e-06,
      "loss": 0.81092405,
      "num_input_tokens_seen": 83980075,
      "step": 3905,
      "time_per_iteration": 2.5137081146240234
    },
    {
      "auxiliary_loss_clip": 0.01164159,
      "auxiliary_loss_mlp": 0.01026212,
      "balance_loss_clip": 1.05430579,
      "balance_loss_mlp": 1.01901197,
      "epoch": 0.4696687308362893,
      "flos": 22016545165440.0,
      "grad_norm": 1.5572191279671832,
      "language_loss": 0.81063116,
      "learning_rate": 2.2926181436849974e-06,
      "loss": 0.83253485,
      "num_input_tokens_seen": 83999430,
      "step": 3906,
      "time_per_iteration": 2.506103992462158
    },
    {
      "auxiliary_loss_clip": 0.01166203,
      "auxiliary_loss_mlp": 0.01028823,
      "balance_loss_clip": 1.05250573,
      "balance_loss_mlp": 1.0205977,
      "epoch": 0.4697889737269284,
      "flos": 21613663244160.0,
      "grad_norm": 1.563402698540988,
      "language_loss": 0.72724342,
      "learning_rate": 2.2918475324326478e-06,
      "loss": 0.74919373,
      "num_input_tokens_seen": 84019150,
      "step": 3907,
      "time_per_iteration": 3.364996910095215
    },
    {
      "auxiliary_loss_clip": 0.01169787,
      "auxiliary_loss_mlp": 0.00762393,
      "balance_loss_clip": 1.05277348,
      "balance_loss_mlp": 1.00017929,
      "epoch": 0.46990921661756746,
      "flos": 25228323665280.0,
      "grad_norm": 1.9375476284409454,
      "language_loss": 0.90829891,
      "learning_rate": 2.2910768769072603e-06,
      "loss": 0.92762077,
      "num_input_tokens_seen": 84037930,
      "step": 3908,
      "time_per_iteration": 2.5534403324127197
    },
    {
      "auxiliary_loss_clip": 0.01160707,
      "auxiliary_loss_mlp": 0.01036995,
      "balance_loss_clip": 1.0519259,
      "balance_loss_mlp": 1.02862597,
      "epoch": 0.47002945950820657,
      "flos": 13844031045120.0,
      "grad_norm": 1.8554245246851973,
      "language_loss": 0.75986612,
      "learning_rate": 2.2903061772257417e-06,
      "loss": 0.78184313,
      "num_input_tokens_seen": 84055915,
      "step": 3909,
      "time_per_iteration": 2.490959405899048
    },
    {
      "auxiliary_loss_clip": 0.01164989,
      "auxiliary_loss_mlp": 0.01027991,
      "balance_loss_clip": 1.05202198,
      "balance_loss_mlp": 1.01993203,
      "epoch": 0.4701497023988457,
      "flos": 26247001374720.0,
      "grad_norm": 1.5129640812035055,
      "language_loss": 0.78246135,
      "learning_rate": 2.289535433505007e-06,
      "loss": 0.80439115,
      "num_input_tokens_seen": 84077270,
      "step": 3910,
      "time_per_iteration": 2.5463273525238037
    },
    {
      "auxiliary_loss_clip": 0.01151413,
      "auxiliary_loss_mlp": 0.01030139,
      "balance_loss_clip": 1.04913914,
      "balance_loss_mlp": 1.02227712,
      "epoch": 0.47026994528948474,
      "flos": 25629517647360.0,
      "grad_norm": 1.636889011396557,
      "language_loss": 0.6361804,
      "learning_rate": 2.2887646458619767e-06,
      "loss": 0.65799594,
      "num_input_tokens_seen": 84098635,
      "step": 3911,
      "time_per_iteration": 2.603100061416626
    },
    {
      "auxiliary_loss_clip": 0.01142425,
      "auxiliary_loss_mlp": 0.01030627,
      "balance_loss_clip": 1.04917002,
      "balance_loss_mlp": 1.02240753,
      "epoch": 0.47039018818012385,
      "flos": 20554406144640.0,
      "grad_norm": 2.0021553836547783,
      "language_loss": 0.76944518,
      "learning_rate": 2.2879938144135797e-06,
      "loss": 0.79117572,
      "num_input_tokens_seen": 84114740,
      "step": 3912,
      "time_per_iteration": 2.5624849796295166
    },
    {
      "auxiliary_loss_clip": 0.01132748,
      "auxiliary_loss_mlp": 0.00761644,
      "balance_loss_clip": 1.04590869,
      "balance_loss_mlp": 1.0001657,
      "epoch": 0.47051043107076296,
      "flos": 21577249831680.0,
      "grad_norm": 1.5192092177692131,
      "language_loss": 0.7509439,
      "learning_rate": 2.2872229392767496e-06,
      "loss": 0.7698878,
      "num_input_tokens_seen": 84134845,
      "step": 3913,
      "time_per_iteration": 4.132529973983765
    },
    {
      "auxiliary_loss_clip": 0.01168875,
      "auxiliary_loss_mlp": 0.01032893,
      "balance_loss_clip": 1.05104733,
      "balance_loss_mlp": 1.02448249,
      "epoch": 0.470630673961402,
      "flos": 18953185662720.0,
      "grad_norm": 1.7540675212771286,
      "language_loss": 0.74724478,
      "learning_rate": 2.286452020568428e-06,
      "loss": 0.76926249,
      "num_input_tokens_seen": 84152920,
      "step": 3914,
      "time_per_iteration": 3.252521514892578
    },
    {
      "auxiliary_loss_clip": 0.01181644,
      "auxiliary_loss_mlp": 0.01033029,
      "balance_loss_clip": 1.05153704,
      "balance_loss_mlp": 1.02470779,
      "epoch": 0.4707509168520411,
      "flos": 19938969492480.0,
      "grad_norm": 2.2520114601687493,
      "language_loss": 0.73179603,
      "learning_rate": 2.2856810584055637e-06,
      "loss": 0.75394285,
      "num_input_tokens_seen": 84170455,
      "step": 3915,
      "time_per_iteration": 2.4646458625793457
    },
    {
      "auxiliary_loss_clip": 0.01166099,
      "auxiliary_loss_mlp": 0.01027052,
      "balance_loss_clip": 1.05092239,
      "balance_loss_mlp": 1.01913071,
      "epoch": 0.47087115974268023,
      "flos": 40118754741120.0,
      "grad_norm": 1.4780928351736522,
      "language_loss": 0.67742598,
      "learning_rate": 2.2849100529051085e-06,
      "loss": 0.69935751,
      "num_input_tokens_seen": 84197390,
      "step": 3916,
      "time_per_iteration": 2.726388931274414
    },
    {
      "auxiliary_loss_clip": 0.01178774,
      "auxiliary_loss_mlp": 0.0103831,
      "balance_loss_clip": 1.05240011,
      "balance_loss_mlp": 1.0309751,
      "epoch": 0.4709914026333193,
      "flos": 13552723745280.0,
      "grad_norm": 2.7057552742525353,
      "language_loss": 0.80281681,
      "learning_rate": 2.284139004184026e-06,
      "loss": 0.82498765,
      "num_input_tokens_seen": 84214620,
      "step": 3917,
      "time_per_iteration": 2.472241163253784
    },
    {
      "auxiliary_loss_clip": 0.01178922,
      "auxiliary_loss_mlp": 0.01035868,
      "balance_loss_clip": 1.05321789,
      "balance_loss_mlp": 1.02791691,
      "epoch": 0.4711116455239584,
      "flos": 19974628719360.0,
      "grad_norm": 2.189669136642054,
      "language_loss": 0.74927092,
      "learning_rate": 2.2833679123592814e-06,
      "loss": 0.77141881,
      "num_input_tokens_seen": 84231880,
      "step": 3918,
      "time_per_iteration": 2.475940704345703
    },
    {
      "auxiliary_loss_clip": 0.01154049,
      "auxiliary_loss_mlp": 0.0103268,
      "balance_loss_clip": 1.05231667,
      "balance_loss_mlp": 1.02465129,
      "epoch": 0.4712318884145975,
      "flos": 32124824064000.0,
      "grad_norm": 1.6228785313046032,
      "language_loss": 0.63101077,
      "learning_rate": 2.2825967775478508e-06,
      "loss": 0.65287811,
      "num_input_tokens_seen": 84252980,
      "step": 3919,
      "time_per_iteration": 2.6257011890411377
    },
    {
      "auxiliary_loss_clip": 0.01177611,
      "auxiliary_loss_mlp": 0.01031302,
      "balance_loss_clip": 1.05179477,
      "balance_loss_mlp": 1.02304125,
      "epoch": 0.47135213130523657,
      "flos": 20047850593920.0,
      "grad_norm": 1.8486734145210084,
      "language_loss": 0.83338773,
      "learning_rate": 2.2818255998667135e-06,
      "loss": 0.85547686,
      "num_input_tokens_seen": 84271490,
      "step": 3920,
      "time_per_iteration": 2.4669673442840576
    },
    {
      "auxiliary_loss_clip": 0.01162315,
      "auxiliary_loss_mlp": 0.01032574,
      "balance_loss_clip": 1.05178928,
      "balance_loss_mlp": 1.02514184,
      "epoch": 0.4714723741958757,
      "flos": 19426990988160.0,
      "grad_norm": 1.5635842243028915,
      "language_loss": 0.78991145,
      "learning_rate": 2.2810543794328566e-06,
      "loss": 0.81186038,
      "num_input_tokens_seen": 84290525,
      "step": 3921,
      "time_per_iteration": 2.4861972332000732
    },
    {
      "auxiliary_loss_clip": 0.01168633,
      "auxiliary_loss_mlp": 0.01031298,
      "balance_loss_clip": 1.05235636,
      "balance_loss_mlp": 1.02368021,
      "epoch": 0.4715926170865148,
      "flos": 20373883367040.0,
      "grad_norm": 3.004643065864307,
      "language_loss": 0.82328463,
      "learning_rate": 2.2802831163632735e-06,
      "loss": 0.84528387,
      "num_input_tokens_seen": 84309245,
      "step": 3922,
      "time_per_iteration": 2.5522544384002686
    },
    {
      "auxiliary_loss_clip": 0.01112703,
      "auxiliary_loss_mlp": 0.01031126,
      "balance_loss_clip": 1.04564762,
      "balance_loss_mlp": 1.02285922,
      "epoch": 0.47171285997715384,
      "flos": 22672884430080.0,
      "grad_norm": 1.6170425431261597,
      "language_loss": 0.7428416,
      "learning_rate": 2.279511810774965e-06,
      "loss": 0.76427996,
      "num_input_tokens_seen": 84330775,
      "step": 3923,
      "time_per_iteration": 2.6586432456970215
    },
    {
      "auxiliary_loss_clip": 0.01180909,
      "auxiliary_loss_mlp": 0.01031565,
      "balance_loss_clip": 1.0543195,
      "balance_loss_mlp": 1.02401948,
      "epoch": 0.47183310286779295,
      "flos": 21105419754240.0,
      "grad_norm": 1.7469766430134295,
      "language_loss": 0.71685362,
      "learning_rate": 2.2787404627849364e-06,
      "loss": 0.73897839,
      "num_input_tokens_seen": 84349985,
      "step": 3924,
      "time_per_iteration": 2.4851696491241455
    },
    {
      "auxiliary_loss_clip": 0.01149304,
      "auxiliary_loss_mlp": 0.01030756,
      "balance_loss_clip": 1.04878771,
      "balance_loss_mlp": 1.02307296,
      "epoch": 0.471953345758432,
      "flos": 21726566668800.0,
      "grad_norm": 1.612684360113646,
      "language_loss": 0.79078019,
      "learning_rate": 2.277969072510202e-06,
      "loss": 0.81258082,
      "num_input_tokens_seen": 84368965,
      "step": 3925,
      "time_per_iteration": 2.538074493408203
    },
    {
      "auxiliary_loss_clip": 0.01151399,
      "auxiliary_loss_mlp": 0.01027254,
      "balance_loss_clip": 1.04948401,
      "balance_loss_mlp": 1.01990521,
      "epoch": 0.4720735886490711,
      "flos": 19861078849920.0,
      "grad_norm": 1.5485895590699696,
      "language_loss": 0.81604922,
      "learning_rate": 2.2771976400677803e-06,
      "loss": 0.83783573,
      "num_input_tokens_seen": 84387795,
      "step": 3926,
      "time_per_iteration": 2.538931369781494
    },
    {
      "auxiliary_loss_clip": 0.01114525,
      "auxiliary_loss_mlp": 0.01028004,
      "balance_loss_clip": 1.04662442,
      "balance_loss_mlp": 1.02002609,
      "epoch": 0.47219383153971023,
      "flos": 19171809792000.0,
      "grad_norm": 1.692485970786244,
      "language_loss": 0.78992987,
      "learning_rate": 2.2764261655746965e-06,
      "loss": 0.81135511,
      "num_input_tokens_seen": 84405290,
      "step": 3927,
      "time_per_iteration": 2.603553295135498
    },
    {
      "auxiliary_loss_clip": 0.01132171,
      "auxiliary_loss_mlp": 0.01032791,
      "balance_loss_clip": 1.04592681,
      "balance_loss_mlp": 1.02417803,
      "epoch": 0.4723140744303493,
      "flos": 23224005780480.0,
      "grad_norm": 1.556848264405677,
      "language_loss": 0.75647151,
      "learning_rate": 2.2756546491479832e-06,
      "loss": 0.77812111,
      "num_input_tokens_seen": 84426205,
      "step": 3928,
      "time_per_iteration": 2.6090219020843506
    },
    {
      "auxiliary_loss_clip": 0.01179429,
      "auxiliary_loss_mlp": 0.00761806,
      "balance_loss_clip": 1.05232882,
      "balance_loss_mlp": 1.00017488,
      "epoch": 0.4724343173209884,
      "flos": 18223265387520.0,
      "grad_norm": 2.041596191964343,
      "language_loss": 0.8020885,
      "learning_rate": 2.274883090904679e-06,
      "loss": 0.8215009,
      "num_input_tokens_seen": 84443970,
      "step": 3929,
      "time_per_iteration": 2.469534397125244
    },
    {
      "auxiliary_loss_clip": 0.01183273,
      "auxiliary_loss_mlp": 0.01036216,
      "balance_loss_clip": 1.05540276,
      "balance_loss_mlp": 1.02800214,
      "epoch": 0.4725545602116275,
      "flos": 21251037490560.0,
      "grad_norm": 2.232378255633167,
      "language_loss": 0.67507076,
      "learning_rate": 2.2741114909618283e-06,
      "loss": 0.69726562,
      "num_input_tokens_seen": 84459865,
      "step": 3930,
      "time_per_iteration": 2.4906420707702637
    },
    {
      "auxiliary_loss_clip": 0.01136227,
      "auxiliary_loss_mlp": 0.0102574,
      "balance_loss_clip": 1.04800224,
      "balance_loss_mlp": 1.0182898,
      "epoch": 0.47267480310226656,
      "flos": 21434002392960.0,
      "grad_norm": 1.6590227297045423,
      "language_loss": 0.71863639,
      "learning_rate": 2.2733398494364828e-06,
      "loss": 0.74025613,
      "num_input_tokens_seen": 84479110,
      "step": 3931,
      "time_per_iteration": 2.593660354614258
    },
    {
      "auxiliary_loss_clip": 0.01143926,
      "auxiliary_loss_mlp": 0.01033488,
      "balance_loss_clip": 1.05067146,
      "balance_loss_mlp": 1.02590084,
      "epoch": 0.47279504599290567,
      "flos": 18770508069120.0,
      "grad_norm": 1.8491693413299166,
      "language_loss": 0.84393948,
      "learning_rate": 2.272568166445699e-06,
      "loss": 0.8657136,
      "num_input_tokens_seen": 84497675,
      "step": 3932,
      "time_per_iteration": 2.5517945289611816
    },
    {
      "auxiliary_loss_clip": 0.01163522,
      "auxiliary_loss_mlp": 0.01028343,
      "balance_loss_clip": 1.04981983,
      "balance_loss_mlp": 1.0206182,
      "epoch": 0.4729152888835448,
      "flos": 21105742976640.0,
      "grad_norm": 1.8267886887332803,
      "language_loss": 0.64324546,
      "learning_rate": 2.271796442106541e-06,
      "loss": 0.66516411,
      "num_input_tokens_seen": 84517030,
      "step": 3933,
      "time_per_iteration": 3.355485439300537
    },
    {
      "auxiliary_loss_clip": 0.01035772,
      "auxiliary_loss_mlp": 0.01002797,
      "balance_loss_clip": 1.0136143,
      "balance_loss_mlp": 1.00149214,
      "epoch": 0.47303553177418384,
      "flos": 70201877840640.0,
      "grad_norm": 0.7978838220985509,
      "language_loss": 0.56495929,
      "learning_rate": 2.271024676536079e-06,
      "loss": 0.58534503,
      "num_input_tokens_seen": 84577290,
      "step": 3934,
      "time_per_iteration": 3.1282217502593994
    },
    {
      "auxiliary_loss_clip": 0.01155895,
      "auxiliary_loss_mlp": 0.01031006,
      "balance_loss_clip": 1.05285645,
      "balance_loss_mlp": 1.0225302,
      "epoch": 0.47315577466482295,
      "flos": 22455122227200.0,
      "grad_norm": 2.228985135556569,
      "language_loss": 0.73288155,
      "learning_rate": 2.2702528698513894e-06,
      "loss": 0.75475055,
      "num_input_tokens_seen": 84598415,
      "step": 3935,
      "time_per_iteration": 2.6323509216308594
    },
    {
      "auxiliary_loss_clip": 0.01150695,
      "auxiliary_loss_mlp": 0.01034604,
      "balance_loss_clip": 1.04723334,
      "balance_loss_mlp": 1.02688825,
      "epoch": 0.47327601755546206,
      "flos": 24352857480960.0,
      "grad_norm": 1.7967169162338479,
      "language_loss": 0.78633618,
      "learning_rate": 2.269481022169554e-06,
      "loss": 0.80818909,
      "num_input_tokens_seen": 84617010,
      "step": 3936,
      "time_per_iteration": 2.647722005844116
    },
    {
      "auxiliary_loss_clip": 0.01155327,
      "auxiliary_loss_mlp": 0.01029941,
      "balance_loss_clip": 1.04822898,
      "balance_loss_mlp": 1.02164364,
      "epoch": 0.4733962604461011,
      "flos": 22926772736640.0,
      "grad_norm": 1.681395721417986,
      "language_loss": 0.80698836,
      "learning_rate": 2.2687091336076614e-06,
      "loss": 0.82884097,
      "num_input_tokens_seen": 84636350,
      "step": 3937,
      "time_per_iteration": 2.618433713912964
    },
    {
      "auxiliary_loss_clip": 0.01162932,
      "auxiliary_loss_mlp": 0.01028249,
      "balance_loss_clip": 1.05132699,
      "balance_loss_mlp": 1.02042902,
      "epoch": 0.4735165033367402,
      "flos": 18327369980160.0,
      "grad_norm": 1.7405370142244998,
      "language_loss": 0.80115747,
      "learning_rate": 2.267937204282807e-06,
      "loss": 0.82306933,
      "num_input_tokens_seen": 84653490,
      "step": 3938,
      "time_per_iteration": 2.48189377784729
    },
    {
      "auxiliary_loss_clip": 0.01172503,
      "auxiliary_loss_mlp": 0.01038909,
      "balance_loss_clip": 1.05408669,
      "balance_loss_mlp": 1.03039742,
      "epoch": 0.4736367462273793,
      "flos": 23037018554880.0,
      "grad_norm": 2.262137563712265,
      "language_loss": 0.78977883,
      "learning_rate": 2.2671652343120926e-06,
      "loss": 0.81189293,
      "num_input_tokens_seen": 84673965,
      "step": 3939,
      "time_per_iteration": 4.048202037811279
    },
    {
      "auxiliary_loss_clip": 0.01181121,
      "auxiliary_loss_mlp": 0.010318,
      "balance_loss_clip": 1.05631661,
      "balance_loss_mlp": 1.02448988,
      "epoch": 0.4737569891180184,
      "flos": 25374336451200.0,
      "grad_norm": 1.845085334500575,
      "language_loss": 0.80522966,
      "learning_rate": 2.2663932238126236e-06,
      "loss": 0.82735884,
      "num_input_tokens_seen": 84692525,
      "step": 3940,
      "time_per_iteration": 2.5288443565368652
    },
    {
      "auxiliary_loss_clip": 0.01160506,
      "auxiliary_loss_mlp": 0.01033134,
      "balance_loss_clip": 1.04750323,
      "balance_loss_mlp": 1.02542675,
      "epoch": 0.4738772320086575,
      "flos": 25849326925440.0,
      "grad_norm": 1.4772941871666148,
      "language_loss": 0.80332726,
      "learning_rate": 2.265621172901515e-06,
      "loss": 0.82526368,
      "num_input_tokens_seen": 84715640,
      "step": 3941,
      "time_per_iteration": 3.367682933807373
    },
    {
      "auxiliary_loss_clip": 0.011825,
      "auxiliary_loss_mlp": 0.01031108,
      "balance_loss_clip": 1.05643082,
      "balance_loss_mlp": 1.02335346,
      "epoch": 0.47399747489929656,
      "flos": 27564420499200.0,
      "grad_norm": 1.9920965809376945,
      "language_loss": 0.70915246,
      "learning_rate": 2.2648490816958854e-06,
      "loss": 0.73128849,
      "num_input_tokens_seen": 84736635,
      "step": 3942,
      "time_per_iteration": 2.5433263778686523
    },
    {
      "auxiliary_loss_clip": 0.01162805,
      "auxiliary_loss_mlp": 0.01031769,
      "balance_loss_clip": 1.04856837,
      "balance_loss_mlp": 1.02352023,
      "epoch": 0.47411771778993567,
      "flos": 24863650836480.0,
      "grad_norm": 2.0759731521740363,
      "language_loss": 0.73060918,
      "learning_rate": 2.264076950312861e-06,
      "loss": 0.75255495,
      "num_input_tokens_seen": 84755445,
      "step": 3943,
      "time_per_iteration": 2.532470703125
    },
    {
      "auxiliary_loss_clip": 0.01151498,
      "auxiliary_loss_mlp": 0.01030839,
      "balance_loss_clip": 1.04935503,
      "balance_loss_mlp": 1.02327478,
      "epoch": 0.4742379606805748,
      "flos": 22748009725440.0,
      "grad_norm": 1.8387574114134273,
      "language_loss": 0.82601106,
      "learning_rate": 2.2633047788695727e-06,
      "loss": 0.84783447,
      "num_input_tokens_seen": 84775750,
      "step": 3944,
      "time_per_iteration": 2.5444822311401367
    },
    {
      "auxiliary_loss_clip": 0.01151438,
      "auxiliary_loss_mlp": 0.01031708,
      "balance_loss_clip": 1.053509,
      "balance_loss_mlp": 1.02447212,
      "epoch": 0.47435820357121383,
      "flos": 19681130689920.0,
      "grad_norm": 1.9076267188105778,
      "language_loss": 0.64342898,
      "learning_rate": 2.262532567483159e-06,
      "loss": 0.66526043,
      "num_input_tokens_seen": 84794310,
      "step": 3945,
      "time_per_iteration": 2.546480178833008
    },
    {
      "auxiliary_loss_clip": 0.01180131,
      "auxiliary_loss_mlp": 0.00761289,
      "balance_loss_clip": 1.05452442,
      "balance_loss_mlp": 1.0001812,
      "epoch": 0.47447844646185294,
      "flos": 25228718714880.0,
      "grad_norm": 1.850369250283981,
      "language_loss": 0.80350924,
      "learning_rate": 2.2617603162707635e-06,
      "loss": 0.82292342,
      "num_input_tokens_seen": 84814720,
      "step": 3946,
      "time_per_iteration": 2.5039408206939697
    },
    {
      "auxiliary_loss_clip": 0.01178933,
      "auxiliary_loss_mlp": 0.01028509,
      "balance_loss_clip": 1.0538367,
      "balance_loss_mlp": 1.02099919,
      "epoch": 0.47459868935249205,
      "flos": 24570619683840.0,
      "grad_norm": 2.219563561605798,
      "language_loss": 0.8281334,
      "learning_rate": 2.2609880253495363e-06,
      "loss": 0.85020781,
      "num_input_tokens_seen": 84834355,
      "step": 3947,
      "time_per_iteration": 2.497504472732544
    },
    {
      "auxiliary_loss_clip": 0.01142446,
      "auxiliary_loss_mlp": 0.010273,
      "balance_loss_clip": 1.04655635,
      "balance_loss_mlp": 1.01913977,
      "epoch": 0.4747189322431311,
      "flos": 20558500295040.0,
      "grad_norm": 1.9542831626803954,
      "language_loss": 0.86350775,
      "learning_rate": 2.260215694836633e-06,
      "loss": 0.88520521,
      "num_input_tokens_seen": 84853530,
      "step": 3948,
      "time_per_iteration": 2.579791784286499
    },
    {
      "auxiliary_loss_clip": 0.01121982,
      "auxiliary_loss_mlp": 0.00761169,
      "balance_loss_clip": 1.04382873,
      "balance_loss_mlp": 1.00016689,
      "epoch": 0.4748391751337702,
      "flos": 25995231970560.0,
      "grad_norm": 1.8404678040867195,
      "language_loss": 0.64315259,
      "learning_rate": 2.2594433248492157e-06,
      "loss": 0.66198415,
      "num_input_tokens_seen": 84872505,
      "step": 3949,
      "time_per_iteration": 2.6583266258239746
    },
    {
      "auxiliary_loss_clip": 0.01169968,
      "auxiliary_loss_mlp": 0.01029759,
      "balance_loss_clip": 1.05209148,
      "balance_loss_mlp": 1.02173054,
      "epoch": 0.47495941802440933,
      "flos": 22821052032000.0,
      "grad_norm": 1.6853420303300406,
      "language_loss": 0.79790032,
      "learning_rate": 2.2586709155044527e-06,
      "loss": 0.81989753,
      "num_input_tokens_seen": 84893105,
      "step": 3950,
      "time_per_iteration": 2.54876971244812
    },
    {
      "auxiliary_loss_clip": 0.0118089,
      "auxiliary_loss_mlp": 0.01034768,
      "balance_loss_clip": 1.05388212,
      "balance_loss_mlp": 1.02684951,
      "epoch": 0.4750796609150484,
      "flos": 27891782075520.0,
      "grad_norm": 2.312447648838554,
      "language_loss": 0.76019019,
      "learning_rate": 2.2578984669195167e-06,
      "loss": 0.78234684,
      "num_input_tokens_seen": 84914070,
      "step": 3951,
      "time_per_iteration": 2.554088592529297
    },
    {
      "auxiliary_loss_clip": 0.01162092,
      "auxiliary_loss_mlp": 0.01029979,
      "balance_loss_clip": 1.04919946,
      "balance_loss_mlp": 1.0227077,
      "epoch": 0.4751999038056875,
      "flos": 35660085471360.0,
      "grad_norm": 1.9391014752637712,
      "language_loss": 0.67957413,
      "learning_rate": 2.2571259792115887e-06,
      "loss": 0.70149481,
      "num_input_tokens_seen": 84935290,
      "step": 3952,
      "time_per_iteration": 2.6680448055267334
    },
    {
      "auxiliary_loss_clip": 0.01159763,
      "auxiliary_loss_mlp": 0.01026857,
      "balance_loss_clip": 1.05053854,
      "balance_loss_mlp": 1.01947165,
      "epoch": 0.4753201466963266,
      "flos": 22090880361600.0,
      "grad_norm": 1.639231989016119,
      "language_loss": 0.79299831,
      "learning_rate": 2.2563534524978544e-06,
      "loss": 0.81486458,
      "num_input_tokens_seen": 84952760,
      "step": 3953,
      "time_per_iteration": 2.533249616622925
    },
    {
      "auxiliary_loss_clip": 0.01133074,
      "auxiliary_loss_mlp": 0.01025517,
      "balance_loss_clip": 1.05083966,
      "balance_loss_mlp": 1.0182395,
      "epoch": 0.47544038958696566,
      "flos": 30190854965760.0,
      "grad_norm": 1.5925792225058881,
      "language_loss": 0.70638573,
      "learning_rate": 2.2555808868955052e-06,
      "loss": 0.72797161,
      "num_input_tokens_seen": 84974890,
      "step": 3954,
      "time_per_iteration": 2.690821409225464
    },
    {
      "auxiliary_loss_clip": 0.01121853,
      "auxiliary_loss_mlp": 0.01028344,
      "balance_loss_clip": 1.04829121,
      "balance_loss_mlp": 1.02013588,
      "epoch": 0.47556063247760477,
      "flos": 23472219738240.0,
      "grad_norm": 1.9772267978023321,
      "language_loss": 0.73416579,
      "learning_rate": 2.254808282521738e-06,
      "loss": 0.75566781,
      "num_input_tokens_seen": 84993640,
      "step": 3955,
      "time_per_iteration": 2.6859192848205566
    },
    {
      "auxiliary_loss_clip": 0.01138808,
      "auxiliary_loss_mlp": 0.00761118,
      "balance_loss_clip": 1.04716563,
      "balance_loss_mlp": 1.00017762,
      "epoch": 0.4756808753682438,
      "flos": 25155209531520.0,
      "grad_norm": 1.6719420765735047,
      "language_loss": 0.8130362,
      "learning_rate": 2.2540356394937573e-06,
      "loss": 0.83203542,
      "num_input_tokens_seen": 85012340,
      "step": 3956,
      "time_per_iteration": 2.622450113296509
    },
    {
      "auxiliary_loss_clip": 0.01140015,
      "auxiliary_loss_mlp": 0.01026401,
      "balance_loss_clip": 1.04704762,
      "balance_loss_mlp": 1.01821756,
      "epoch": 0.47580111825888294,
      "flos": 15669729573120.0,
      "grad_norm": 2.160331826371299,
      "language_loss": 0.84287542,
      "learning_rate": 2.253262957928772e-06,
      "loss": 0.86453962,
      "num_input_tokens_seen": 85029225,
      "step": 3957,
      "time_per_iteration": 2.5811872482299805
    },
    {
      "auxiliary_loss_clip": 0.01142045,
      "auxiliary_loss_mlp": 0.01033455,
      "balance_loss_clip": 1.0449307,
      "balance_loss_mlp": 1.02541447,
      "epoch": 0.47592136114952205,
      "flos": 17636556637440.0,
      "grad_norm": 1.6487403302546868,
      "language_loss": 0.71700585,
      "learning_rate": 2.2524902379439976e-06,
      "loss": 0.73876083,
      "num_input_tokens_seen": 85047895,
      "step": 3958,
      "time_per_iteration": 2.5213100910186768
    },
    {
      "auxiliary_loss_clip": 0.01012743,
      "auxiliary_loss_mlp": 0.01008677,
      "balance_loss_clip": 1.01643276,
      "balance_loss_mlp": 1.0072341,
      "epoch": 0.4760416040401611,
      "flos": 61417159292160.0,
      "grad_norm": 0.7399358780206213,
      "language_loss": 0.63730258,
      "learning_rate": 2.251717479656655e-06,
      "loss": 0.65751678,
      "num_input_tokens_seen": 85112690,
      "step": 3959,
      "time_per_iteration": 4.097006797790527
    },
    {
      "auxiliary_loss_clip": 0.01180465,
      "auxiliary_loss_mlp": 0.01028744,
      "balance_loss_clip": 1.05373633,
      "balance_loss_mlp": 1.02110267,
      "epoch": 0.4761618469308002,
      "flos": 18405871153920.0,
      "grad_norm": 1.7915878874984799,
      "language_loss": 0.76179051,
      "learning_rate": 2.2509446831839704e-06,
      "loss": 0.78388256,
      "num_input_tokens_seen": 85132130,
      "step": 3960,
      "time_per_iteration": 2.6185150146484375
    },
    {
      "auxiliary_loss_clip": 0.01152576,
      "auxiliary_loss_mlp": 0.01027053,
      "balance_loss_clip": 1.04839694,
      "balance_loss_mlp": 1.01924467,
      "epoch": 0.4762820898214393,
      "flos": 18040911016320.0,
      "grad_norm": 2.0561947348066494,
      "language_loss": 0.82206821,
      "learning_rate": 2.250171848643177e-06,
      "loss": 0.84386456,
      "num_input_tokens_seen": 85149420,
      "step": 3961,
      "time_per_iteration": 2.521286725997925
    },
    {
      "auxiliary_loss_clip": 0.01148423,
      "auxiliary_loss_mlp": 0.01034296,
      "balance_loss_clip": 1.05215073,
      "balance_loss_mlp": 1.02659178,
      "epoch": 0.4764023327120784,
      "flos": 19318253541120.0,
      "grad_norm": 2.8442630293970743,
      "language_loss": 0.86016786,
      "learning_rate": 2.249398976151513e-06,
      "loss": 0.88199508,
      "num_input_tokens_seen": 85166970,
      "step": 3962,
      "time_per_iteration": 2.5312082767486572
    },
    {
      "auxiliary_loss_clip": 0.01179057,
      "auxiliary_loss_mlp": 0.01029551,
      "balance_loss_clip": 1.05504215,
      "balance_loss_mlp": 1.02121794,
      "epoch": 0.4765225756027175,
      "flos": 22747255539840.0,
      "grad_norm": 2.086182106234943,
      "language_loss": 0.78521729,
      "learning_rate": 2.248626065826223e-06,
      "loss": 0.80730337,
      "num_input_tokens_seen": 85185175,
      "step": 3963,
      "time_per_iteration": 2.492340326309204
    },
    {
      "auxiliary_loss_clip": 0.01071053,
      "auxiliary_loss_mlp": 0.01002114,
      "balance_loss_clip": 1.01742625,
      "balance_loss_mlp": 1.00067794,
      "epoch": 0.4766428184933566,
      "flos": 65933392106880.0,
      "grad_norm": 0.759086812486289,
      "language_loss": 0.62603271,
      "learning_rate": 2.2478531177845564e-06,
      "loss": 0.6467644,
      "num_input_tokens_seen": 85246170,
      "step": 3964,
      "time_per_iteration": 3.061748743057251
    },
    {
      "auxiliary_loss_clip": 0.01155685,
      "auxiliary_loss_mlp": 0.01031652,
      "balance_loss_clip": 1.05315292,
      "balance_loss_mlp": 1.02368879,
      "epoch": 0.47676306138399566,
      "flos": 24136495908480.0,
      "grad_norm": 1.9100977605636438,
      "language_loss": 0.8522768,
      "learning_rate": 2.247080132143769e-06,
      "loss": 0.87415022,
      "num_input_tokens_seen": 85268525,
      "step": 3965,
      "time_per_iteration": 4.066251516342163
    },
    {
      "auxiliary_loss_clip": 0.01135427,
      "auxiliary_loss_mlp": 0.01029643,
      "balance_loss_clip": 1.04628849,
      "balance_loss_mlp": 1.02154887,
      "epoch": 0.47688330427463477,
      "flos": 12604322995200.0,
      "grad_norm": 2.2003461394593375,
      "language_loss": 0.69576621,
      "learning_rate": 2.246307109021121e-06,
      "loss": 0.71741688,
      "num_input_tokens_seen": 85285930,
      "step": 3966,
      "time_per_iteration": 2.59199595451355
    },
    {
      "auxiliary_loss_clip": 0.01144144,
      "auxiliary_loss_mlp": 0.01034986,
      "balance_loss_clip": 1.04627824,
      "balance_loss_mlp": 1.02715135,
      "epoch": 0.4770035471652739,
      "flos": 21390585828480.0,
      "grad_norm": 1.6214443206927447,
      "language_loss": 0.82301164,
      "learning_rate": 2.2455340485338817e-06,
      "loss": 0.84480298,
      "num_input_tokens_seen": 85303565,
      "step": 3967,
      "time_per_iteration": 3.5031399726867676
    },
    {
      "auxiliary_loss_clip": 0.01165322,
      "auxiliary_loss_mlp": 0.01032454,
      "balance_loss_clip": 1.05085135,
      "balance_loss_mlp": 1.02508664,
      "epoch": 0.47712379005591293,
      "flos": 25156251025920.0,
      "grad_norm": 1.771508718412266,
      "language_loss": 0.67901975,
      "learning_rate": 2.244760950799322e-06,
      "loss": 0.70099747,
      "num_input_tokens_seen": 85321835,
      "step": 3968,
      "time_per_iteration": 2.557223081588745
    },
    {
      "auxiliary_loss_clip": 0.01125212,
      "auxiliary_loss_mlp": 0.01027342,
      "balance_loss_clip": 1.04864621,
      "balance_loss_mlp": 1.0201385,
      "epoch": 0.47724403294655204,
      "flos": 22054323294720.0,
      "grad_norm": 2.398224335324407,
      "language_loss": 0.72330874,
      "learning_rate": 2.2439878159347203e-06,
      "loss": 0.7448343,
      "num_input_tokens_seen": 85341260,
      "step": 3969,
      "time_per_iteration": 2.6182138919830322
    },
    {
      "auxiliary_loss_clip": 0.01069775,
      "auxiliary_loss_mlp": 0.01001497,
      "balance_loss_clip": 1.01640606,
      "balance_loss_mlp": 1.0000844,
      "epoch": 0.4773642758371911,
      "flos": 70229387658240.0,
      "grad_norm": 1.9717527003684086,
      "language_loss": 0.55326581,
      "learning_rate": 2.2432146440573616e-06,
      "loss": 0.57397848,
      "num_input_tokens_seen": 85407220,
      "step": 3970,
      "time_per_iteration": 3.182111978530884
    },
    {
      "auxiliary_loss_clip": 0.01149075,
      "auxiliary_loss_mlp": 0.01027192,
      "balance_loss_clip": 1.05274391,
      "balance_loss_mlp": 1.01931798,
      "epoch": 0.4774845187278302,
      "flos": 23548602009600.0,
      "grad_norm": 1.8719651561898045,
      "language_loss": 0.66609597,
      "learning_rate": 2.242441435284534e-06,
      "loss": 0.68785858,
      "num_input_tokens_seen": 85426095,
      "step": 3971,
      "time_per_iteration": 2.579507350921631
    },
    {
      "auxiliary_loss_clip": 0.01164238,
      "auxiliary_loss_mlp": 0.01035617,
      "balance_loss_clip": 1.0516454,
      "balance_loss_mlp": 1.02785099,
      "epoch": 0.4776047616184693,
      "flos": 23075371301760.0,
      "grad_norm": 2.6058485311740522,
      "language_loss": 0.85106385,
      "learning_rate": 2.2416681897335337e-06,
      "loss": 0.87306249,
      "num_input_tokens_seen": 85444245,
      "step": 3972,
      "time_per_iteration": 2.546626091003418
    },
    {
      "auxiliary_loss_clip": 0.01122509,
      "auxiliary_loss_mlp": 0.01028217,
      "balance_loss_clip": 1.04949832,
      "balance_loss_mlp": 1.02015209,
      "epoch": 0.4777250045091084,
      "flos": 31898119374720.0,
      "grad_norm": 1.7717607357995335,
      "language_loss": 0.67126673,
      "learning_rate": 2.240894907521661e-06,
      "loss": 0.692774,
      "num_input_tokens_seen": 85463325,
      "step": 3973,
      "time_per_iteration": 2.7341911792755127
    },
    {
      "auxiliary_loss_clip": 0.01149909,
      "auxiliary_loss_mlp": 0.01029443,
      "balance_loss_clip": 1.0495162,
      "balance_loss_mlp": 1.02166462,
      "epoch": 0.4778452473997475,
      "flos": 24278163148800.0,
      "grad_norm": 2.1234857826427653,
      "language_loss": 0.63661569,
      "learning_rate": 2.240121588766223e-06,
      "loss": 0.65840924,
      "num_input_tokens_seen": 85483375,
      "step": 3974,
      "time_per_iteration": 2.556644916534424
    },
    {
      "auxiliary_loss_clip": 0.01147063,
      "auxiliary_loss_mlp": 0.01023514,
      "balance_loss_clip": 1.04932594,
      "balance_loss_mlp": 1.01643252,
      "epoch": 0.4779654902903866,
      "flos": 31575031516800.0,
      "grad_norm": 4.044783973308583,
      "language_loss": 0.71096683,
      "learning_rate": 2.239348233584531e-06,
      "loss": 0.73267257,
      "num_input_tokens_seen": 85504230,
      "step": 3975,
      "time_per_iteration": 2.6246495246887207
    },
    {
      "auxiliary_loss_clip": 0.01167684,
      "auxiliary_loss_mlp": 0.01032355,
      "balance_loss_clip": 1.05257654,
      "balance_loss_mlp": 1.02463603,
      "epoch": 0.47808573318102565,
      "flos": 19500428344320.0,
      "grad_norm": 1.778489207228484,
      "language_loss": 0.81412137,
      "learning_rate": 2.2385748420939013e-06,
      "loss": 0.83612174,
      "num_input_tokens_seen": 85523425,
      "step": 3976,
      "time_per_iteration": 2.489778757095337
    },
    {
      "auxiliary_loss_clip": 0.01179546,
      "auxiliary_loss_mlp": 0.01033407,
      "balance_loss_clip": 1.05774784,
      "balance_loss_mlp": 1.02578712,
      "epoch": 0.47820597607166476,
      "flos": 22601135013120.0,
      "grad_norm": 1.870072484383098,
      "language_loss": 0.72178924,
      "learning_rate": 2.2378014144116583e-06,
      "loss": 0.74391878,
      "num_input_tokens_seen": 85542235,
      "step": 3977,
      "time_per_iteration": 2.4951300621032715
    },
    {
      "auxiliary_loss_clip": 0.01181846,
      "auxiliary_loss_mlp": 0.01032158,
      "balance_loss_clip": 1.05438972,
      "balance_loss_mlp": 1.0244987,
      "epoch": 0.4783262189623039,
      "flos": 23003011353600.0,
      "grad_norm": 1.9253909897351225,
      "language_loss": 0.79754639,
      "learning_rate": 2.23702795065513e-06,
      "loss": 0.81968641,
      "num_input_tokens_seen": 85561815,
      "step": 3978,
      "time_per_iteration": 2.50985050201416
    },
    {
      "auxiliary_loss_clip": 0.01061374,
      "auxiliary_loss_mlp": 0.01001495,
      "balance_loss_clip": 1.0169214,
      "balance_loss_mlp": 1.00005841,
      "epoch": 0.47844646185294293,
      "flos": 49772801226240.0,
      "grad_norm": 0.977430210277476,
      "language_loss": 0.6746949,
      "learning_rate": 2.2362544509416493e-06,
      "loss": 0.69532359,
      "num_input_tokens_seen": 85613930,
      "step": 3979,
      "time_per_iteration": 2.9790618419647217
    },
    {
      "auxiliary_loss_clip": 0.01145246,
      "auxiliary_loss_mlp": 0.01027929,
      "balance_loss_clip": 1.04822421,
      "balance_loss_mlp": 1.02035332,
      "epoch": 0.47856670474358204,
      "flos": 20229558520320.0,
      "grad_norm": 2.768240885639337,
      "language_loss": 0.82600975,
      "learning_rate": 2.2354809153885572e-06,
      "loss": 0.84774148,
      "num_input_tokens_seen": 85631000,
      "step": 3980,
      "time_per_iteration": 2.553560495376587
    },
    {
      "auxiliary_loss_clip": 0.01163607,
      "auxiliary_loss_mlp": 0.01032151,
      "balance_loss_clip": 1.05143261,
      "balance_loss_mlp": 1.02402043,
      "epoch": 0.47868694763422115,
      "flos": 20990936131200.0,
      "grad_norm": 1.8690246954365137,
      "language_loss": 0.83501577,
      "learning_rate": 2.234707344113197e-06,
      "loss": 0.85697341,
      "num_input_tokens_seen": 85649095,
      "step": 3981,
      "time_per_iteration": 2.521613121032715
    },
    {
      "auxiliary_loss_clip": 0.01176529,
      "auxiliary_loss_mlp": 0.01026044,
      "balance_loss_clip": 1.05378509,
      "balance_loss_mlp": 1.01900196,
      "epoch": 0.4788071905248602,
      "flos": 19026551191680.0,
      "grad_norm": 1.6489277543723717,
      "language_loss": 0.776752,
      "learning_rate": 2.233933737232919e-06,
      "loss": 0.7987777,
      "num_input_tokens_seen": 85666875,
      "step": 3982,
      "time_per_iteration": 2.466125011444092
    },
    {
      "auxiliary_loss_clip": 0.01111844,
      "auxiliary_loss_mlp": 0.0076095,
      "balance_loss_clip": 1.04427624,
      "balance_loss_mlp": 1.00019252,
      "epoch": 0.4789274334154993,
      "flos": 23002221254400.0,
      "grad_norm": 1.5473221794891645,
      "language_loss": 0.78299975,
      "learning_rate": 2.2331600948650793e-06,
      "loss": 0.80172765,
      "num_input_tokens_seen": 85687020,
      "step": 3983,
      "time_per_iteration": 2.6191110610961914
    },
    {
      "auxiliary_loss_clip": 0.01124425,
      "auxiliary_loss_mlp": 0.00762052,
      "balance_loss_clip": 1.04793549,
      "balance_loss_mlp": 1.00017416,
      "epoch": 0.4790476763061384,
      "flos": 23075586783360.0,
      "grad_norm": 1.4388350378835197,
      "language_loss": 0.79908121,
      "learning_rate": 2.2323864171270386e-06,
      "loss": 0.81794596,
      "num_input_tokens_seen": 85708290,
      "step": 3984,
      "time_per_iteration": 2.6517903804779053
    },
    {
      "auxiliary_loss_clip": 0.0113907,
      "auxiliary_loss_mlp": 0.01028397,
      "balance_loss_clip": 1.04728818,
      "balance_loss_mlp": 1.02009392,
      "epoch": 0.4791679191967775,
      "flos": 21179288073600.0,
      "grad_norm": 1.949776765138691,
      "language_loss": 0.72600627,
      "learning_rate": 2.231612704136164e-06,
      "loss": 0.7476809,
      "num_input_tokens_seen": 85728660,
      "step": 3985,
      "time_per_iteration": 3.4202473163604736
    },
    {
      "auxiliary_loss_clip": 0.01158696,
      "auxiliary_loss_mlp": 0.01026233,
      "balance_loss_clip": 1.04895258,
      "balance_loss_mlp": 1.01763225,
      "epoch": 0.4792881620874166,
      "flos": 22301495758080.0,
      "grad_norm": 4.268952485456957,
      "language_loss": 0.74631119,
      "learning_rate": 2.2308389560098253e-06,
      "loss": 0.76816052,
      "num_input_tokens_seen": 85745035,
      "step": 3986,
      "time_per_iteration": 2.5143215656280518
    },
    {
      "auxiliary_loss_clip": 0.01136238,
      "auxiliary_loss_mlp": 0.01027575,
      "balance_loss_clip": 1.04976749,
      "balance_loss_mlp": 1.0199964,
      "epoch": 0.47940840497805565,
      "flos": 17420877423360.0,
      "grad_norm": 2.827476928955933,
      "language_loss": 0.7683782,
      "learning_rate": 2.2300651728654008e-06,
      "loss": 0.79001629,
      "num_input_tokens_seen": 85760295,
      "step": 3987,
      "time_per_iteration": 2.5738797187805176
    },
    {
      "auxiliary_loss_clip": 0.01058171,
      "auxiliary_loss_mlp": 0.00751276,
      "balance_loss_clip": 1.01714182,
      "balance_loss_mlp": 1.00010014,
      "epoch": 0.47952864786869476,
      "flos": 65358175708800.0,
      "grad_norm": 0.7351302843937069,
      "language_loss": 0.60197484,
      "learning_rate": 2.229291354820272e-06,
      "loss": 0.62006938,
      "num_input_tokens_seen": 85821305,
      "step": 3988,
      "time_per_iteration": 3.136117935180664
    },
    {
      "auxiliary_loss_clip": 0.01161535,
      "auxiliary_loss_mlp": 0.01034505,
      "balance_loss_clip": 1.04976618,
      "balance_loss_mlp": 1.02638721,
      "epoch": 0.47964889075933387,
      "flos": 16799802336000.0,
      "grad_norm": 1.981617809978044,
      "language_loss": 0.75704062,
      "learning_rate": 2.228517501991828e-06,
      "loss": 0.77900106,
      "num_input_tokens_seen": 85840105,
      "step": 3989,
      "time_per_iteration": 2.503363847732544
    },
    {
      "auxiliary_loss_clip": 0.01048167,
      "auxiliary_loss_mlp": 0.01002562,
      "balance_loss_clip": 1.01621389,
      "balance_loss_mlp": 1.00117326,
      "epoch": 0.4797691336499729,
      "flos": 70079244808320.0,
      "grad_norm": 0.8063816426794026,
      "language_loss": 0.61041033,
      "learning_rate": 2.22774361449746e-06,
      "loss": 0.63091755,
      "num_input_tokens_seen": 85896585,
      "step": 3990,
      "time_per_iteration": 3.15537428855896
    },
    {
      "auxiliary_loss_clip": 0.01103873,
      "auxiliary_loss_mlp": 0.01027612,
      "balance_loss_clip": 1.04462826,
      "balance_loss_mlp": 1.01975036,
      "epoch": 0.47988937654061203,
      "flos": 18953329317120.0,
      "grad_norm": 2.8388554812268874,
      "language_loss": 0.70909339,
      "learning_rate": 2.2269696924545668e-06,
      "loss": 0.73040825,
      "num_input_tokens_seen": 85914415,
      "step": 3991,
      "time_per_iteration": 4.190083980560303
    },
    {
      "auxiliary_loss_clip": 0.01141036,
      "auxiliary_loss_mlp": 0.01030184,
      "balance_loss_clip": 1.05404794,
      "balance_loss_mlp": 1.02288258,
      "epoch": 0.48000961943125114,
      "flos": 14461981649280.0,
      "grad_norm": 2.2797506574674467,
      "language_loss": 0.78162849,
      "learning_rate": 2.2261957359805523e-06,
      "loss": 0.80334073,
      "num_input_tokens_seen": 85931650,
      "step": 3992,
      "time_per_iteration": 2.55216121673584
    },
    {
      "auxiliary_loss_clip": 0.0117966,
      "auxiliary_loss_mlp": 0.01030247,
      "balance_loss_clip": 1.05448627,
      "balance_loss_mlp": 1.02274823,
      "epoch": 0.4801298623218902,
      "flos": 27051149105280.0,
      "grad_norm": 3.106254225084741,
      "language_loss": 0.74114382,
      "learning_rate": 2.225421745192823e-06,
      "loss": 0.76324296,
      "num_input_tokens_seen": 85951805,
      "step": 3993,
      "time_per_iteration": 3.3045389652252197
    },
    {
      "auxiliary_loss_clip": 0.01164053,
      "auxiliary_loss_mlp": 0.01028641,
      "balance_loss_clip": 1.05241632,
      "balance_loss_mlp": 1.0208385,
      "epoch": 0.4802501052125293,
      "flos": 26355236031360.0,
      "grad_norm": 1.9796454466552096,
      "language_loss": 0.78063267,
      "learning_rate": 2.2246477202087955e-06,
      "loss": 0.80255961,
      "num_input_tokens_seen": 85972485,
      "step": 3994,
      "time_per_iteration": 2.6068384647369385
    },
    {
      "auxiliary_loss_clip": 0.01148765,
      "auxiliary_loss_mlp": 0.01029241,
      "balance_loss_clip": 1.04744458,
      "balance_loss_mlp": 1.02200198,
      "epoch": 0.4803703481031684,
      "flos": 20993916960000.0,
      "grad_norm": 1.5316409024774031,
      "language_loss": 0.83059919,
      "learning_rate": 2.223873661145887e-06,
      "loss": 0.85237926,
      "num_input_tokens_seen": 85992540,
      "step": 3995,
      "time_per_iteration": 2.5599682331085205
    },
    {
      "auxiliary_loss_clip": 0.01146067,
      "auxiliary_loss_mlp": 0.00760812,
      "balance_loss_clip": 1.05174065,
      "balance_loss_mlp": 1.00016689,
      "epoch": 0.4804905909938075,
      "flos": 20703722981760.0,
      "grad_norm": 1.6193879549098928,
      "language_loss": 0.71145433,
      "learning_rate": 2.2230995681215226e-06,
      "loss": 0.73052317,
      "num_input_tokens_seen": 86012065,
      "step": 3996,
      "time_per_iteration": 2.5559229850769043
    },
    {
      "auxiliary_loss_clip": 0.01131958,
      "auxiliary_loss_mlp": 0.01028398,
      "balance_loss_clip": 1.04652822,
      "balance_loss_mlp": 1.02088237,
      "epoch": 0.4806108338844466,
      "flos": 16654831044480.0,
      "grad_norm": 1.9900032715070965,
      "language_loss": 0.77919853,
      "learning_rate": 2.2223254412531305e-06,
      "loss": 0.80080205,
      "num_input_tokens_seen": 86029435,
      "step": 3997,
      "time_per_iteration": 2.5612854957580566
    },
    {
      "auxiliary_loss_clip": 0.0113526,
      "auxiliary_loss_mlp": 0.01026685,
      "balance_loss_clip": 1.04583359,
      "balance_loss_mlp": 1.01975298,
      "epoch": 0.4807310767750857,
      "flos": 20011329440640.0,
      "grad_norm": 1.8612498124672565,
      "language_loss": 0.82181007,
      "learning_rate": 2.221551280658146e-06,
      "loss": 0.84342957,
      "num_input_tokens_seen": 86048495,
      "step": 3998,
      "time_per_iteration": 2.5735697746276855
    },
    {
      "auxiliary_loss_clip": 0.01119965,
      "auxiliary_loss_mlp": 0.01027391,
      "balance_loss_clip": 1.04767525,
      "balance_loss_mlp": 1.01991081,
      "epoch": 0.48085131966572475,
      "flos": 23185257984000.0,
      "grad_norm": 1.8153344623258607,
      "language_loss": 0.74120212,
      "learning_rate": 2.2207770864540085e-06,
      "loss": 0.76267564,
      "num_input_tokens_seen": 86067470,
      "step": 3999,
      "time_per_iteration": 2.6267201900482178
    },
    {
      "auxiliary_loss_clip": 0.01143297,
      "auxiliary_loss_mlp": 0.01026951,
      "balance_loss_clip": 1.04958999,
      "balance_loss_mlp": 1.01960206,
      "epoch": 0.48097156255636386,
      "flos": 20558643949440.0,
      "grad_norm": 1.7871746016833943,
      "language_loss": 0.72524142,
      "learning_rate": 2.220002858758162e-06,
      "loss": 0.74694383,
      "num_input_tokens_seen": 86085460,
      "step": 4000,
      "time_per_iteration": 2.555229663848877
    },
    {
      "auxiliary_loss_clip": 0.01060842,
      "auxiliary_loss_mlp": 0.0100089,
      "balance_loss_clip": 1.01636076,
      "balance_loss_mlp": 0.99946576,
      "epoch": 0.481091805447003,
      "flos": 70511608817280.0,
      "grad_norm": 0.8747779589611542,
      "language_loss": 0.60924417,
      "learning_rate": 2.2192285976880573e-06,
      "loss": 0.62986153,
      "num_input_tokens_seen": 86149715,
      "step": 4001,
      "time_per_iteration": 3.1436238288879395
    },
    {
      "auxiliary_loss_clip": 0.01136058,
      "auxiliary_loss_mlp": 0.00760389,
      "balance_loss_clip": 1.04692721,
      "balance_loss_mlp": 1.0001905,
      "epoch": 0.48121204833764203,
      "flos": 36428214839040.0,
      "grad_norm": 1.475812490575636,
      "language_loss": 0.80399537,
      "learning_rate": 2.2184543033611485e-06,
      "loss": 0.8229599,
      "num_input_tokens_seen": 86170795,
      "step": 4002,
      "time_per_iteration": 2.750775098800659
    },
    {
      "auxiliary_loss_clip": 0.01165655,
      "auxiliary_loss_mlp": 0.01030245,
      "balance_loss_clip": 1.05215633,
      "balance_loss_mlp": 1.0227828,
      "epoch": 0.48133229122828114,
      "flos": 27490264871040.0,
      "grad_norm": 2.012272104266677,
      "language_loss": 0.81680131,
      "learning_rate": 2.2176799758948957e-06,
      "loss": 0.83876026,
      "num_input_tokens_seen": 86190955,
      "step": 4003,
      "time_per_iteration": 2.5892810821533203
    },
    {
      "auxiliary_loss_clip": 0.01145912,
      "auxiliary_loss_mlp": 0.01030518,
      "balance_loss_clip": 1.05083501,
      "balance_loss_mlp": 1.02284098,
      "epoch": 0.4814525341189202,
      "flos": 43072802179200.0,
      "grad_norm": 2.7804680343752044,
      "language_loss": 0.73161995,
      "learning_rate": 2.2169056154067635e-06,
      "loss": 0.75338423,
      "num_input_tokens_seen": 86214875,
      "step": 4004,
      "time_per_iteration": 2.755753517150879
    },
    {
      "auxiliary_loss_clip": 0.01164706,
      "auxiliary_loss_mlp": 0.00761444,
      "balance_loss_clip": 1.05307865,
      "balance_loss_mlp": 1.0001719,
      "epoch": 0.4815727770095593,
      "flos": 24236901400320.0,
      "grad_norm": 1.6214464525549772,
      "language_loss": 0.82187021,
      "learning_rate": 2.216131222014222e-06,
      "loss": 0.84113169,
      "num_input_tokens_seen": 86232950,
      "step": 4005,
      "time_per_iteration": 2.5263490676879883
    },
    {
      "auxiliary_loss_clip": 0.01132763,
      "auxiliary_loss_mlp": 0.01031151,
      "balance_loss_clip": 1.05056143,
      "balance_loss_mlp": 1.02309871,
      "epoch": 0.4816930199001984,
      "flos": 18113630100480.0,
      "grad_norm": 2.2252178630512023,
      "language_loss": 0.80090868,
      "learning_rate": 2.2153567958347455e-06,
      "loss": 0.82254785,
      "num_input_tokens_seen": 86249160,
      "step": 4006,
      "time_per_iteration": 2.563032865524292
    },
    {
      "auxiliary_loss_clip": 0.01150592,
      "auxiliary_loss_mlp": 0.01029494,
      "balance_loss_clip": 1.0513953,
      "balance_loss_mlp": 1.02192986,
      "epoch": 0.48181326279083747,
      "flos": 17274720983040.0,
      "grad_norm": 2.108983949883118,
      "language_loss": 0.79924417,
      "learning_rate": 2.214582336985815e-06,
      "loss": 0.82104504,
      "num_input_tokens_seen": 86267060,
      "step": 4007,
      "time_per_iteration": 2.5239100456237793
    },
    {
      "auxiliary_loss_clip": 0.01140299,
      "auxiliary_loss_mlp": 0.01029063,
      "balance_loss_clip": 1.04837453,
      "balance_loss_mlp": 1.02068257,
      "epoch": 0.4819335056814766,
      "flos": 14903252231040.0,
      "grad_norm": 2.1543547897025124,
      "language_loss": 0.66224939,
      "learning_rate": 2.2138078455849142e-06,
      "loss": 0.68394303,
      "num_input_tokens_seen": 86285055,
      "step": 4008,
      "time_per_iteration": 2.5297999382019043
    },
    {
      "auxiliary_loss_clip": 0.01168726,
      "auxiliary_loss_mlp": 0.01028633,
      "balance_loss_clip": 1.05400479,
      "balance_loss_mlp": 1.02078342,
      "epoch": 0.4820537485721157,
      "flos": 19244888012160.0,
      "grad_norm": 1.9603910524836532,
      "language_loss": 0.78604567,
      "learning_rate": 2.2130333217495334e-06,
      "loss": 0.80801928,
      "num_input_tokens_seen": 86304225,
      "step": 4009,
      "time_per_iteration": 2.5136220455169678
    },
    {
      "auxiliary_loss_clip": 0.01143948,
      "auxiliary_loss_mlp": 0.01023803,
      "balance_loss_clip": 1.04816818,
      "balance_loss_mlp": 1.01555359,
      "epoch": 0.48217399146275475,
      "flos": 16033791870720.0,
      "grad_norm": 2.3351533644879954,
      "language_loss": 0.67210323,
      "learning_rate": 2.2122587655971665e-06,
      "loss": 0.69378072,
      "num_input_tokens_seen": 86319170,
      "step": 4010,
      "time_per_iteration": 2.5136637687683105
    },
    {
      "auxiliary_loss_clip": 0.01150734,
      "auxiliary_loss_mlp": 0.01029284,
      "balance_loss_clip": 1.04866862,
      "balance_loss_mlp": 1.02146971,
      "epoch": 0.48229423435339386,
      "flos": 24134197438080.0,
      "grad_norm": 1.600429004435232,
      "language_loss": 0.63934451,
      "learning_rate": 2.211484177245314e-06,
      "loss": 0.66114467,
      "num_input_tokens_seen": 86338760,
      "step": 4011,
      "time_per_iteration": 3.331915855407715
    },
    {
      "auxiliary_loss_clip": 0.0118332,
      "auxiliary_loss_mlp": 0.01033481,
      "balance_loss_clip": 1.05671453,
      "balance_loss_mlp": 1.02586675,
      "epoch": 0.48241447724403297,
      "flos": 23805435231360.0,
      "grad_norm": 2.519162730710436,
      "language_loss": 0.72179806,
      "learning_rate": 2.21070955681148e-06,
      "loss": 0.7439661,
      "num_input_tokens_seen": 86357865,
      "step": 4012,
      "time_per_iteration": 2.508892297744751
    },
    {
      "auxiliary_loss_clip": 0.01130899,
      "auxiliary_loss_mlp": 0.01031211,
      "balance_loss_clip": 1.04912531,
      "balance_loss_mlp": 1.0235101,
      "epoch": 0.482534720134672,
      "flos": 23110312256640.0,
      "grad_norm": 1.5960944280797,
      "language_loss": 0.78454077,
      "learning_rate": 2.209934904413174e-06,
      "loss": 0.80616188,
      "num_input_tokens_seen": 86379470,
      "step": 4013,
      "time_per_iteration": 2.6428723335266113
    },
    {
      "auxiliary_loss_clip": 0.01099578,
      "auxiliary_loss_mlp": 0.01028502,
      "balance_loss_clip": 1.03646159,
      "balance_loss_mlp": 1.02073538,
      "epoch": 0.48265496302531113,
      "flos": 20923819568640.0,
      "grad_norm": 2.698081441431407,
      "language_loss": 0.71742678,
      "learning_rate": 2.2091602201679095e-06,
      "loss": 0.7387076,
      "num_input_tokens_seen": 86399080,
      "step": 4014,
      "time_per_iteration": 2.668179750442505
    },
    {
      "auxiliary_loss_clip": 0.01142136,
      "auxiliary_loss_mlp": 0.01028103,
      "balance_loss_clip": 1.05055273,
      "balance_loss_mlp": 1.02077174,
      "epoch": 0.48277520591595025,
      "flos": 15231152511360.0,
      "grad_norm": 2.6054109924025974,
      "language_loss": 0.83522624,
      "learning_rate": 2.208385504193206e-06,
      "loss": 0.85692859,
      "num_input_tokens_seen": 86416580,
      "step": 4015,
      "time_per_iteration": 2.577590227127075
    },
    {
      "auxiliary_loss_clip": 0.01179164,
      "auxiliary_loss_mlp": 0.01034312,
      "balance_loss_clip": 1.05312657,
      "balance_loss_mlp": 1.0264976,
      "epoch": 0.4828954488065893,
      "flos": 17858664385920.0,
      "grad_norm": 2.632757901985482,
      "language_loss": 0.80917168,
      "learning_rate": 2.2076107566065873e-06,
      "loss": 0.8313064,
      "num_input_tokens_seen": 86434365,
      "step": 4016,
      "time_per_iteration": 3.2712457180023193
    },
    {
      "auxiliary_loss_clip": 0.01163751,
      "auxiliary_loss_mlp": 0.01027248,
      "balance_loss_clip": 1.05134606,
      "balance_loss_mlp": 1.01972044,
      "epoch": 0.4830156916972284,
      "flos": 32087405070720.0,
      "grad_norm": 2.8871880787574415,
      "language_loss": 0.75517869,
      "learning_rate": 2.2068359775255816e-06,
      "loss": 0.7770887,
      "num_input_tokens_seen": 86452675,
      "step": 4017,
      "time_per_iteration": 3.335500478744507
    },
    {
      "auxiliary_loss_clip": 0.01115586,
      "auxiliary_loss_mlp": 0.01024103,
      "balance_loss_clip": 1.04447579,
      "balance_loss_mlp": 1.01703119,
      "epoch": 0.48313593458786747,
      "flos": 21871717528320.0,
      "grad_norm": 2.600462468368549,
      "language_loss": 0.78452688,
      "learning_rate": 2.206061167067723e-06,
      "loss": 0.80592376,
      "num_input_tokens_seen": 86470785,
      "step": 4018,
      "time_per_iteration": 2.616544008255005
    },
    {
      "auxiliary_loss_clip": 0.01132228,
      "auxiliary_loss_mlp": 0.01033255,
      "balance_loss_clip": 1.04560328,
      "balance_loss_mlp": 1.02524459,
      "epoch": 0.4832561774785066,
      "flos": 22601206840320.0,
      "grad_norm": 2.2237475934678677,
      "language_loss": 0.79455584,
      "learning_rate": 2.205286325350549e-06,
      "loss": 0.81621069,
      "num_input_tokens_seen": 86489850,
      "step": 4019,
      "time_per_iteration": 3.404973030090332
    },
    {
      "auxiliary_loss_clip": 0.01123015,
      "auxiliary_loss_mlp": 0.01023122,
      "balance_loss_clip": 1.0470891,
      "balance_loss_mlp": 1.01544762,
      "epoch": 0.4833764203691457,
      "flos": 13437342282240.0,
      "grad_norm": 2.028423215309233,
      "language_loss": 0.72353923,
      "learning_rate": 2.204511452491603e-06,
      "loss": 0.7450006,
      "num_input_tokens_seen": 86506475,
      "step": 4020,
      "time_per_iteration": 2.5949995517730713
    },
    {
      "auxiliary_loss_clip": 0.01176376,
      "auxiliary_loss_mlp": 0.01025069,
      "balance_loss_clip": 1.05371487,
      "balance_loss_mlp": 1.01763046,
      "epoch": 0.48349666325978474,
      "flos": 44128036955520.0,
      "grad_norm": 2.1442901975976203,
      "language_loss": 0.74869215,
      "learning_rate": 2.2037365486084316e-06,
      "loss": 0.77070665,
      "num_input_tokens_seen": 86529715,
      "step": 4021,
      "time_per_iteration": 2.69012451171875
    },
    {
      "auxiliary_loss_clip": 0.0114027,
      "auxiliary_loss_mlp": 0.010277,
      "balance_loss_clip": 1.0456481,
      "balance_loss_mlp": 1.01982319,
      "epoch": 0.48361690615042385,
      "flos": 26028377245440.0,
      "grad_norm": 1.9142871364194984,
      "language_loss": 0.77770281,
      "learning_rate": 2.2029616138185886e-06,
      "loss": 0.79938257,
      "num_input_tokens_seen": 86548715,
      "step": 4022,
      "time_per_iteration": 2.643929958343506
    },
    {
      "auxiliary_loss_clip": 0.01134375,
      "auxiliary_loss_mlp": 0.01029914,
      "balance_loss_clip": 1.05152273,
      "balance_loss_mlp": 1.02279437,
      "epoch": 0.48373714904106296,
      "flos": 22273306560000.0,
      "grad_norm": 1.6174966182640038,
      "language_loss": 0.82555622,
      "learning_rate": 2.202186648239629e-06,
      "loss": 0.84719908,
      "num_input_tokens_seen": 86568650,
      "step": 4023,
      "time_per_iteration": 2.62048602104187
    },
    {
      "auxiliary_loss_clip": 0.0115994,
      "auxiliary_loss_mlp": 0.01031759,
      "balance_loss_clip": 1.05097759,
      "balance_loss_mlp": 1.02420735,
      "epoch": 0.483857391931702,
      "flos": 28292293699200.0,
      "grad_norm": 1.6498759540523387,
      "language_loss": 0.71230006,
      "learning_rate": 2.201411651989117e-06,
      "loss": 0.73421705,
      "num_input_tokens_seen": 86590630,
      "step": 4024,
      "time_per_iteration": 2.59281063079834
    },
    {
      "auxiliary_loss_clip": 0.01148538,
      "auxiliary_loss_mlp": 0.0076175,
      "balance_loss_clip": 1.05190039,
      "balance_loss_mlp": 1.00017858,
      "epoch": 0.48397763482234113,
      "flos": 27418048577280.0,
      "grad_norm": 1.8402177395924462,
      "language_loss": 0.78306925,
      "learning_rate": 2.2006366251846167e-06,
      "loss": 0.80217218,
      "num_input_tokens_seen": 86611270,
      "step": 4025,
      "time_per_iteration": 2.624256134033203
    },
    {
      "auxiliary_loss_clip": 0.0115027,
      "auxiliary_loss_mlp": 0.01031118,
      "balance_loss_clip": 1.05340743,
      "balance_loss_mlp": 1.02426982,
      "epoch": 0.48409787771298024,
      "flos": 16797252470400.0,
      "grad_norm": 1.7143266876349377,
      "language_loss": 0.7562052,
      "learning_rate": 2.1998615679436997e-06,
      "loss": 0.77801907,
      "num_input_tokens_seen": 86628810,
      "step": 4026,
      "time_per_iteration": 2.5462794303894043
    },
    {
      "auxiliary_loss_clip": 0.01154383,
      "auxiliary_loss_mlp": 0.01032655,
      "balance_loss_clip": 1.04938698,
      "balance_loss_mlp": 1.02521586,
      "epoch": 0.4842181206036193,
      "flos": 25083496028160.0,
      "grad_norm": 3.2010503417048475,
      "language_loss": 0.77529085,
      "learning_rate": 2.199086480383942e-06,
      "loss": 0.79716128,
      "num_input_tokens_seen": 86648185,
      "step": 4027,
      "time_per_iteration": 2.6159818172454834
    },
    {
      "auxiliary_loss_clip": 0.01161395,
      "auxiliary_loss_mlp": 0.0103394,
      "balance_loss_clip": 1.0515337,
      "balance_loss_mlp": 1.02512407,
      "epoch": 0.4843383634942584,
      "flos": 30372311496960.0,
      "grad_norm": 3.055089041942076,
      "language_loss": 0.67505139,
      "learning_rate": 2.1983113626229234e-06,
      "loss": 0.69700474,
      "num_input_tokens_seen": 86667435,
      "step": 4028,
      "time_per_iteration": 2.6492321491241455
    },
    {
      "auxiliary_loss_clip": 0.01127701,
      "auxiliary_loss_mlp": 0.00761316,
      "balance_loss_clip": 1.04470658,
      "balance_loss_mlp": 1.00017357,
      "epoch": 0.4844586063848975,
      "flos": 20413564917120.0,
      "grad_norm": 1.6573761412155803,
      "language_loss": 0.78257048,
      "learning_rate": 2.1975362147782293e-06,
      "loss": 0.80146062,
      "num_input_tokens_seen": 86686630,
      "step": 4029,
      "time_per_iteration": 2.5921969413757324
    },
    {
      "auxiliary_loss_clip": 0.01043839,
      "auxiliary_loss_mlp": 0.01004685,
      "balance_loss_clip": 1.01488161,
      "balance_loss_mlp": 1.00338602,
      "epoch": 0.48457884927553657,
      "flos": 70303722854400.0,
      "grad_norm": 0.6908391657320504,
      "language_loss": 0.54187286,
      "learning_rate": 2.196761036967448e-06,
      "loss": 0.56235814,
      "num_input_tokens_seen": 86754595,
      "step": 4030,
      "time_per_iteration": 3.257030487060547
    },
    {
      "auxiliary_loss_clip": 0.01160573,
      "auxiliary_loss_mlp": 0.01033433,
      "balance_loss_clip": 1.05137587,
      "balance_loss_mlp": 1.02609241,
      "epoch": 0.4846990921661757,
      "flos": 19934516206080.0,
      "grad_norm": 1.8272301791107664,
      "language_loss": 0.77643198,
      "learning_rate": 2.1959858293081743e-06,
      "loss": 0.79837203,
      "num_input_tokens_seen": 86773730,
      "step": 4031,
      "time_per_iteration": 2.5692660808563232
    },
    {
      "auxiliary_loss_clip": 0.01133803,
      "auxiliary_loss_mlp": 0.01041287,
      "balance_loss_clip": 1.05005598,
      "balance_loss_mlp": 1.03341889,
      "epoch": 0.4848193350568148,
      "flos": 23075945919360.0,
      "grad_norm": 1.6055250886791677,
      "language_loss": 0.75976408,
      "learning_rate": 2.1952105919180056e-06,
      "loss": 0.781515,
      "num_input_tokens_seen": 86792985,
      "step": 4032,
      "time_per_iteration": 2.578372001647949
    },
    {
      "auxiliary_loss_clip": 0.01149847,
      "auxiliary_loss_mlp": 0.01033414,
      "balance_loss_clip": 1.05311418,
      "balance_loss_mlp": 1.02529001,
      "epoch": 0.48493957794745385,
      "flos": 22455481363200.0,
      "grad_norm": 2.349521395535041,
      "language_loss": 0.67889404,
      "learning_rate": 2.1944353249145456e-06,
      "loss": 0.70072675,
      "num_input_tokens_seen": 86812095,
      "step": 4033,
      "time_per_iteration": 2.579235792160034
    },
    {
      "auxiliary_loss_clip": 0.01179153,
      "auxiliary_loss_mlp": 0.01029148,
      "balance_loss_clip": 1.0559746,
      "balance_loss_mlp": 1.02179539,
      "epoch": 0.48505982083809296,
      "flos": 25046112948480.0,
      "grad_norm": 1.6261506679807713,
      "language_loss": 0.74237657,
      "learning_rate": 2.193660028415401e-06,
      "loss": 0.76445955,
      "num_input_tokens_seen": 86832875,
      "step": 4034,
      "time_per_iteration": 2.5190722942352295
    },
    {
      "auxiliary_loss_clip": 0.0114249,
      "auxiliary_loss_mlp": 0.01031383,
      "balance_loss_clip": 1.04914093,
      "balance_loss_mlp": 1.02362275,
      "epoch": 0.485180063728732,
      "flos": 26761386090240.0,
      "grad_norm": 1.722517719795405,
      "language_loss": 0.82385945,
      "learning_rate": 2.1928847025381852e-06,
      "loss": 0.8455981,
      "num_input_tokens_seen": 86853480,
      "step": 4035,
      "time_per_iteration": 2.588080644607544
    },
    {
      "auxiliary_loss_clip": 0.011608,
      "auxiliary_loss_mlp": 0.01031167,
      "balance_loss_clip": 1.0478096,
      "balance_loss_mlp": 1.02326345,
      "epoch": 0.4853003066193711,
      "flos": 24059143969920.0,
      "grad_norm": 1.5252970715865397,
      "language_loss": 0.83826762,
      "learning_rate": 2.192109347400512e-06,
      "loss": 0.86018729,
      "num_input_tokens_seen": 86873695,
      "step": 4036,
      "time_per_iteration": 2.5357491970062256
    },
    {
      "auxiliary_loss_clip": 0.0115259,
      "auxiliary_loss_mlp": 0.01035075,
      "balance_loss_clip": 1.04936695,
      "balance_loss_mlp": 1.02652144,
      "epoch": 0.48542054951001024,
      "flos": 23076376882560.0,
      "grad_norm": 1.7408563411096505,
      "language_loss": 0.78937685,
      "learning_rate": 2.191333963120004e-06,
      "loss": 0.81125355,
      "num_input_tokens_seen": 86892675,
      "step": 4037,
      "time_per_iteration": 3.404409885406494
    },
    {
      "auxiliary_loss_clip": 0.01149877,
      "auxiliary_loss_mlp": 0.01029875,
      "balance_loss_clip": 1.05166602,
      "balance_loss_mlp": 1.02238822,
      "epoch": 0.4855407924006493,
      "flos": 25664889565440.0,
      "grad_norm": 2.4861297603864636,
      "language_loss": 0.70156872,
      "learning_rate": 2.190558549814286e-06,
      "loss": 0.72336626,
      "num_input_tokens_seen": 86912835,
      "step": 4038,
      "time_per_iteration": 2.597511053085327
    },
    {
      "auxiliary_loss_clip": 0.01144261,
      "auxiliary_loss_mlp": 0.01035472,
      "balance_loss_clip": 1.04641986,
      "balance_loss_mlp": 1.02795005,
      "epoch": 0.4856610352912884,
      "flos": 23987933256960.0,
      "grad_norm": 1.737694827150912,
      "language_loss": 0.79473281,
      "learning_rate": 2.1897831076009872e-06,
      "loss": 0.81653011,
      "num_input_tokens_seen": 86932475,
      "step": 4039,
      "time_per_iteration": 2.5654983520507812
    },
    {
      "auxiliary_loss_clip": 0.0116489,
      "auxiliary_loss_mlp": 0.01027156,
      "balance_loss_clip": 1.05261707,
      "balance_loss_mlp": 1.01962805,
      "epoch": 0.4857812781819275,
      "flos": 24096814358400.0,
      "grad_norm": 1.6355243623738087,
      "language_loss": 0.7973761,
      "learning_rate": 2.1890076365977426e-06,
      "loss": 0.8192966,
      "num_input_tokens_seen": 86952300,
      "step": 4040,
      "time_per_iteration": 2.539324998855591
    },
    {
      "auxiliary_loss_clip": 0.01048204,
      "auxiliary_loss_mlp": 0.01002121,
      "balance_loss_clip": 1.01729321,
      "balance_loss_mlp": 1.00082195,
      "epoch": 0.48590152107256657,
      "flos": 56266635185280.0,
      "grad_norm": 0.858785060791233,
      "language_loss": 0.52764708,
      "learning_rate": 2.188232136922189e-06,
      "loss": 0.5481503,
      "num_input_tokens_seen": 87010420,
      "step": 4041,
      "time_per_iteration": 3.053109884262085
    },
    {
      "auxiliary_loss_clip": 0.01099899,
      "auxiliary_loss_mlp": 0.01026392,
      "balance_loss_clip": 1.04441988,
      "balance_loss_mlp": 1.01746917,
      "epoch": 0.4860217639632057,
      "flos": 20046988667520.0,
      "grad_norm": 1.8051528523886142,
      "language_loss": 0.75476456,
      "learning_rate": 2.187456608691971e-06,
      "loss": 0.77602744,
      "num_input_tokens_seen": 87029295,
      "step": 4042,
      "time_per_iteration": 3.4495980739593506
    },
    {
      "auxiliary_loss_clip": 0.01140106,
      "auxiliary_loss_mlp": 0.01034752,
      "balance_loss_clip": 1.05177724,
      "balance_loss_mlp": 1.0268898,
      "epoch": 0.4861420068538448,
      "flos": 17822143232640.0,
      "grad_norm": 2.190687452590082,
      "language_loss": 0.87439227,
      "learning_rate": 2.1866810520247334e-06,
      "loss": 0.89614081,
      "num_input_tokens_seen": 87048165,
      "step": 4043,
      "time_per_iteration": 3.3202764987945557
    },
    {
      "auxiliary_loss_clip": 0.0116759,
      "auxiliary_loss_mlp": 0.01033175,
      "balance_loss_clip": 1.05128634,
      "balance_loss_mlp": 1.02531874,
      "epoch": 0.48626224974448384,
      "flos": 26250125857920.0,
      "grad_norm": 1.7056525803082725,
      "language_loss": 0.64610386,
      "learning_rate": 2.185905467038129e-06,
      "loss": 0.66811144,
      "num_input_tokens_seen": 87067070,
      "step": 4044,
      "time_per_iteration": 2.564095973968506
    },
    {
      "auxiliary_loss_clip": 0.01175131,
      "auxiliary_loss_mlp": 0.01030285,
      "balance_loss_clip": 1.05414701,
      "balance_loss_mlp": 1.02281344,
      "epoch": 0.48638249263512295,
      "flos": 22054502862720.0,
      "grad_norm": 1.6522589853284468,
      "language_loss": 0.77471197,
      "learning_rate": 2.1851298538498127e-06,
      "loss": 0.7967661,
      "num_input_tokens_seen": 87086785,
      "step": 4045,
      "time_per_iteration": 3.2632195949554443
    },
    {
      "auxiliary_loss_clip": 0.01170786,
      "auxiliary_loss_mlp": 0.00762248,
      "balance_loss_clip": 1.05473697,
      "balance_loss_mlp": 1.00017881,
      "epoch": 0.48650273552576206,
      "flos": 25119945354240.0,
      "grad_norm": 1.9718942086976732,
      "language_loss": 0.80125046,
      "learning_rate": 2.184354212577446e-06,
      "loss": 0.82058084,
      "num_input_tokens_seen": 87107090,
      "step": 4046,
      "time_per_iteration": 2.567845106124878
    },
    {
      "auxiliary_loss_clip": 0.01179402,
      "auxiliary_loss_mlp": 0.01023495,
      "balance_loss_clip": 1.05189753,
      "balance_loss_mlp": 1.01557994,
      "epoch": 0.4866229784164011,
      "flos": 17456931699840.0,
      "grad_norm": 2.6059000385717512,
      "language_loss": 0.624017,
      "learning_rate": 2.1835785433386907e-06,
      "loss": 0.64604604,
      "num_input_tokens_seen": 87125905,
      "step": 4047,
      "time_per_iteration": 2.4632349014282227
    },
    {
      "auxiliary_loss_clip": 0.01128825,
      "auxiliary_loss_mlp": 0.01033269,
      "balance_loss_clip": 1.04939997,
      "balance_loss_mlp": 1.02525806,
      "epoch": 0.48674322130704023,
      "flos": 23331127115520.0,
      "grad_norm": 1.8593704666182245,
      "language_loss": 0.65319127,
      "learning_rate": 2.182802846251216e-06,
      "loss": 0.6748122,
      "num_input_tokens_seen": 87146175,
      "step": 4048,
      "time_per_iteration": 2.6029865741729736
    },
    {
      "auxiliary_loss_clip": 0.01140859,
      "auxiliary_loss_mlp": 0.01027971,
      "balance_loss_clip": 1.04788971,
      "balance_loss_mlp": 1.02032661,
      "epoch": 0.4868634641976793,
      "flos": 28804344030720.0,
      "grad_norm": 1.7979211317055557,
      "language_loss": 0.72281069,
      "learning_rate": 2.182027121432696e-06,
      "loss": 0.74449897,
      "num_input_tokens_seen": 87166800,
      "step": 4049,
      "time_per_iteration": 2.724167585372925
    },
    {
      "auxiliary_loss_clip": 0.01182192,
      "auxiliary_loss_mlp": 0.01027452,
      "balance_loss_clip": 1.05471921,
      "balance_loss_mlp": 1.01927471,
      "epoch": 0.4869837070883184,
      "flos": 19025976574080.0,
      "grad_norm": 2.1716943160475806,
      "language_loss": 0.82376295,
      "learning_rate": 2.1812513690008054e-06,
      "loss": 0.84585941,
      "num_input_tokens_seen": 87185920,
      "step": 4050,
      "time_per_iteration": 2.488213300704956
    },
    {
      "auxiliary_loss_clip": 0.01172095,
      "auxiliary_loss_mlp": 0.01033539,
      "balance_loss_clip": 1.05360746,
      "balance_loss_mlp": 1.0255816,
      "epoch": 0.4871039499789575,
      "flos": 15121409483520.0,
      "grad_norm": 2.0672555592850683,
      "language_loss": 0.79591835,
      "learning_rate": 2.180475589073227e-06,
      "loss": 0.81797469,
      "num_input_tokens_seen": 87203620,
      "step": 4051,
      "time_per_iteration": 2.5021560192108154
    },
    {
      "auxiliary_loss_clip": 0.01153007,
      "auxiliary_loss_mlp": 0.01023362,
      "balance_loss_clip": 1.04803586,
      "balance_loss_mlp": 1.01601255,
      "epoch": 0.48722419286959656,
      "flos": 26174066808960.0,
      "grad_norm": 1.7298100598468127,
      "language_loss": 0.73434186,
      "learning_rate": 2.1796997817676456e-06,
      "loss": 0.75610554,
      "num_input_tokens_seen": 87224630,
      "step": 4052,
      "time_per_iteration": 2.5463407039642334
    },
    {
      "auxiliary_loss_clip": 0.01165575,
      "auxiliary_loss_mlp": 0.00760735,
      "balance_loss_clip": 1.05238986,
      "balance_loss_mlp": 1.00019836,
      "epoch": 0.4873444357602357,
      "flos": 24026142349440.0,
      "grad_norm": 1.4914326274598346,
      "language_loss": 0.67723584,
      "learning_rate": 2.1789239472017494e-06,
      "loss": 0.69649893,
      "num_input_tokens_seen": 87246280,
      "step": 4053,
      "time_per_iteration": 2.580043315887451
    },
    {
      "auxiliary_loss_clip": 0.01133971,
      "auxiliary_loss_mlp": 0.01029185,
      "balance_loss_clip": 1.0473249,
      "balance_loss_mlp": 1.02091217,
      "epoch": 0.4874646786508748,
      "flos": 22820441500800.0,
      "grad_norm": 3.161445088248908,
      "language_loss": 0.73164916,
      "learning_rate": 2.1781480854932326e-06,
      "loss": 0.7532807,
      "num_input_tokens_seen": 87266045,
      "step": 4054,
      "time_per_iteration": 2.5829977989196777
    },
    {
      "auxiliary_loss_clip": 0.011225,
      "auxiliary_loss_mlp": 0.01034758,
      "balance_loss_clip": 1.04971468,
      "balance_loss_mlp": 1.02728975,
      "epoch": 0.48758492154151384,
      "flos": 21287594557440.0,
      "grad_norm": 3.171770037107228,
      "language_loss": 0.79121721,
      "learning_rate": 2.1773721967597933e-06,
      "loss": 0.8127898,
      "num_input_tokens_seen": 87284495,
      "step": 4055,
      "time_per_iteration": 2.612440824508667
    },
    {
      "auxiliary_loss_clip": 0.01046018,
      "auxiliary_loss_mlp": 0.01005522,
      "balance_loss_clip": 1.01710224,
      "balance_loss_mlp": 1.00432444,
      "epoch": 0.48770516443215295,
      "flos": 62244109180800.0,
      "grad_norm": 0.8445793526804394,
      "language_loss": 0.57393217,
      "learning_rate": 2.1765962811191322e-06,
      "loss": 0.59444761,
      "num_input_tokens_seen": 87338960,
      "step": 4056,
      "time_per_iteration": 3.1014516353607178
    },
    {
      "auxiliary_loss_clip": 0.01027347,
      "auxiliary_loss_mlp": 0.01006564,
      "balance_loss_clip": 1.02022934,
      "balance_loss_mlp": 1.00532424,
      "epoch": 0.48782540732279206,
      "flos": 66133451882880.0,
      "grad_norm": 0.824701256285342,
      "language_loss": 0.62040156,
      "learning_rate": 2.1758203386889566e-06,
      "loss": 0.64074063,
      "num_input_tokens_seen": 87401730,
      "step": 4057,
      "time_per_iteration": 3.2202255725860596
    },
    {
      "auxiliary_loss_clip": 0.01136105,
      "auxiliary_loss_mlp": 0.00761818,
      "balance_loss_clip": 1.04919529,
      "balance_loss_mlp": 1.00019717,
      "epoch": 0.4879456502134311,
      "flos": 14607922608000.0,
      "grad_norm": 3.1523159821456974,
      "language_loss": 0.84610164,
      "learning_rate": 2.1750443695869746e-06,
      "loss": 0.86508095,
      "num_input_tokens_seen": 87417300,
      "step": 4058,
      "time_per_iteration": 2.578232765197754
    },
    {
      "auxiliary_loss_clip": 0.01164965,
      "auxiliary_loss_mlp": 0.01026022,
      "balance_loss_clip": 1.04996061,
      "balance_loss_mlp": 1.01817846,
      "epoch": 0.4880658931040702,
      "flos": 19500464257920.0,
      "grad_norm": 1.711206553629646,
      "language_loss": 0.86037219,
      "learning_rate": 2.174268373930901e-06,
      "loss": 0.88228208,
      "num_input_tokens_seen": 87434815,
      "step": 4059,
      "time_per_iteration": 2.5200185775756836
    },
    {
      "auxiliary_loss_clip": 0.01131583,
      "auxiliary_loss_mlp": 0.00761604,
      "balance_loss_clip": 1.05027783,
      "balance_loss_mlp": 1.00018251,
      "epoch": 0.48818613599470934,
      "flos": 16723060928640.0,
      "grad_norm": 2.5952456997432423,
      "language_loss": 0.80133331,
      "learning_rate": 2.1734923518384537e-06,
      "loss": 0.82026529,
      "num_input_tokens_seen": 87451420,
      "step": 4060,
      "time_per_iteration": 2.5693461894989014
    },
    {
      "auxiliary_loss_clip": 0.01122102,
      "auxiliary_loss_mlp": 0.01028246,
      "balance_loss_clip": 1.0484271,
      "balance_loss_mlp": 1.02061093,
      "epoch": 0.4883063788853484,
      "flos": 26756932803840.0,
      "grad_norm": 1.6926019891895128,
      "language_loss": 0.82130969,
      "learning_rate": 2.1727163034273547e-06,
      "loss": 0.84281313,
      "num_input_tokens_seen": 87469585,
      "step": 4061,
      "time_per_iteration": 2.647096633911133
    },
    {
      "auxiliary_loss_clip": 0.01163208,
      "auxiliary_loss_mlp": 0.01030062,
      "balance_loss_clip": 1.04881108,
      "balance_loss_mlp": 1.02252197,
      "epoch": 0.4884266217759875,
      "flos": 16763388923520.0,
      "grad_norm": 2.0396234510720466,
      "language_loss": 0.78861111,
      "learning_rate": 2.17194022881533e-06,
      "loss": 0.81054378,
      "num_input_tokens_seen": 87485675,
      "step": 4062,
      "time_per_iteration": 2.48291015625
    },
    {
      "auxiliary_loss_clip": 0.0115475,
      "auxiliary_loss_mlp": 0.01030487,
      "balance_loss_clip": 1.05089664,
      "balance_loss_mlp": 1.02176666,
      "epoch": 0.4885468646666266,
      "flos": 24207132003840.0,
      "grad_norm": 1.6680653289736693,
      "language_loss": 0.67816854,
      "learning_rate": 2.1711641281201092e-06,
      "loss": 0.70002091,
      "num_input_tokens_seen": 87505605,
      "step": 4063,
      "time_per_iteration": 3.4254682064056396
    },
    {
      "auxiliary_loss_clip": 0.0116629,
      "auxiliary_loss_mlp": 0.01027764,
      "balance_loss_clip": 1.05490041,
      "balance_loss_mlp": 1.0202719,
      "epoch": 0.48866710755726567,
      "flos": 14610795696000.0,
      "grad_norm": 2.046741381279861,
      "language_loss": 0.79160511,
      "learning_rate": 2.1703880014594264e-06,
      "loss": 0.8135457,
      "num_input_tokens_seen": 87523195,
      "step": 4064,
      "time_per_iteration": 2.5121071338653564
    },
    {
      "auxiliary_loss_clip": 0.01113182,
      "auxiliary_loss_mlp": 0.01029896,
      "balance_loss_clip": 1.04750955,
      "balance_loss_mlp": 1.02278185,
      "epoch": 0.4887873504479048,
      "flos": 28804451771520.0,
      "grad_norm": 1.6029144081947768,
      "language_loss": 0.73953438,
      "learning_rate": 2.1696118489510182e-06,
      "loss": 0.76096517,
      "num_input_tokens_seen": 87544125,
      "step": 4065,
      "time_per_iteration": 2.6652181148529053
    },
    {
      "auxiliary_loss_clip": 0.01142949,
      "auxiliary_loss_mlp": 0.00761637,
      "balance_loss_clip": 1.04977119,
      "balance_loss_mlp": 1.00019574,
      "epoch": 0.48890759333854383,
      "flos": 22784387224320.0,
      "grad_norm": 2.386706233509167,
      "language_loss": 0.72898239,
      "learning_rate": 2.1688356707126286e-06,
      "loss": 0.74802822,
      "num_input_tokens_seen": 87563745,
      "step": 4066,
      "time_per_iteration": 2.5987792015075684
    },
    {
      "auxiliary_loss_clip": 0.01133116,
      "auxiliary_loss_mlp": 0.01031154,
      "balance_loss_clip": 1.05069959,
      "balance_loss_mlp": 1.02292848,
      "epoch": 0.48902783622918294,
      "flos": 17786088956160.0,
      "grad_norm": 1.9193284996521307,
      "language_loss": 0.70219517,
      "learning_rate": 2.168059466862001e-06,
      "loss": 0.72383785,
      "num_input_tokens_seen": 87581895,
      "step": 4067,
      "time_per_iteration": 3.328705072402954
    },
    {
      "auxiliary_loss_clip": 0.01150955,
      "auxiliary_loss_mlp": 0.01029577,
      "balance_loss_clip": 1.04695225,
      "balance_loss_mlp": 1.02201295,
      "epoch": 0.48914807911982205,
      "flos": 22310294590080.0,
      "grad_norm": 3.5982188518597464,
      "language_loss": 0.8180114,
      "learning_rate": 2.167283237516887e-06,
      "loss": 0.83981681,
      "num_input_tokens_seen": 87600170,
      "step": 4068,
      "time_per_iteration": 2.5494580268859863
    },
    {
      "auxiliary_loss_clip": 0.01154082,
      "auxiliary_loss_mlp": 0.01028456,
      "balance_loss_clip": 1.0510788,
      "balance_loss_mlp": 1.02071309,
      "epoch": 0.4892683220104611,
      "flos": 16363020954240.0,
      "grad_norm": 1.7084735013652907,
      "language_loss": 0.74400461,
      "learning_rate": 2.1665069827950383e-06,
      "loss": 0.76583004,
      "num_input_tokens_seen": 87617455,
      "step": 4069,
      "time_per_iteration": 3.2794036865234375
    },
    {
      "auxiliary_loss_clip": 0.01146666,
      "auxiliary_loss_mlp": 0.01031064,
      "balance_loss_clip": 1.04964399,
      "balance_loss_mlp": 1.02381277,
      "epoch": 0.4893885649011002,
      "flos": 15739144606080.0,
      "grad_norm": 1.6451703315750443,
      "language_loss": 0.86808717,
      "learning_rate": 2.1657307028142126e-06,
      "loss": 0.88986444,
      "num_input_tokens_seen": 87634995,
      "step": 4070,
      "time_per_iteration": 2.5219714641571045
    },
    {
      "auxiliary_loss_clip": 0.01151952,
      "auxiliary_loss_mlp": 0.01033389,
      "balance_loss_clip": 1.05118108,
      "balance_loss_mlp": 1.02517605,
      "epoch": 0.48950880779173933,
      "flos": 28581984887040.0,
      "grad_norm": 2.0168145259010464,
      "language_loss": 0.66825962,
      "learning_rate": 2.164954397692171e-06,
      "loss": 0.69011307,
      "num_input_tokens_seen": 87654420,
      "step": 4071,
      "time_per_iteration": 3.439969301223755
    },
    {
      "auxiliary_loss_clip": 0.0105634,
      "auxiliary_loss_mlp": 0.01002605,
      "balance_loss_clip": 1.02097774,
      "balance_loss_mlp": 1.00130582,
      "epoch": 0.4896290506823784,
      "flos": 66186310746240.0,
      "grad_norm": 1.1364450998683144,
      "language_loss": 0.77349752,
      "learning_rate": 2.164178067546678e-06,
      "loss": 0.79408693,
      "num_input_tokens_seen": 87713585,
      "step": 4072,
      "time_per_iteration": 3.199843168258667
    },
    {
      "auxiliary_loss_clip": 0.0115469,
      "auxiliary_loss_mlp": 0.01024791,
      "balance_loss_clip": 1.04896259,
      "balance_loss_mlp": 1.01704502,
      "epoch": 0.4897492935730175,
      "flos": 12531065207040.0,
      "grad_norm": 1.739137066606048,
      "language_loss": 0.90505004,
      "learning_rate": 2.163401712495504e-06,
      "loss": 0.92684484,
      "num_input_tokens_seen": 87731280,
      "step": 4073,
      "time_per_iteration": 2.5418434143066406
    },
    {
      "auxiliary_loss_clip": 0.01119971,
      "auxiliary_loss_mlp": 0.010329,
      "balance_loss_clip": 1.04709065,
      "balance_loss_mlp": 1.02531266,
      "epoch": 0.4898695364636566,
      "flos": 23476816679040.0,
      "grad_norm": 2.1651824183629316,
      "language_loss": 0.79402113,
      "learning_rate": 2.1626253326564194e-06,
      "loss": 0.81554985,
      "num_input_tokens_seen": 87750230,
      "step": 4074,
      "time_per_iteration": 2.621030569076538
    },
    {
      "auxiliary_loss_clip": 0.01147775,
      "auxiliary_loss_mlp": 0.01031786,
      "balance_loss_clip": 1.04741073,
      "balance_loss_mlp": 1.02379942,
      "epoch": 0.48998977935429566,
      "flos": 27160209774720.0,
      "grad_norm": 1.6489024674091168,
      "language_loss": 0.76898026,
      "learning_rate": 2.161848928147201e-06,
      "loss": 0.7907759,
      "num_input_tokens_seen": 87770500,
      "step": 4075,
      "time_per_iteration": 2.6019067764282227
    },
    {
      "auxiliary_loss_clip": 0.01162701,
      "auxiliary_loss_mlp": 0.01027237,
      "balance_loss_clip": 1.05317283,
      "balance_loss_mlp": 1.01938105,
      "epoch": 0.4901100222449348,
      "flos": 20339588856960.0,
      "grad_norm": 2.089728574454659,
      "language_loss": 0.80730718,
      "learning_rate": 2.161072499085629e-06,
      "loss": 0.82920659,
      "num_input_tokens_seen": 87789495,
      "step": 4076,
      "time_per_iteration": 2.5156238079071045
    },
    {
      "auxiliary_loss_clip": 0.01140122,
      "auxiliary_loss_mlp": 0.01031719,
      "balance_loss_clip": 1.0511198,
      "balance_loss_mlp": 1.02402377,
      "epoch": 0.4902302651355739,
      "flos": 30446359384320.0,
      "grad_norm": 1.5373179364288327,
      "language_loss": 0.82935333,
      "learning_rate": 2.160296045589487e-06,
      "loss": 0.85107172,
      "num_input_tokens_seen": 87812955,
      "step": 4077,
      "time_per_iteration": 2.663034677505493
    },
    {
      "auxiliary_loss_clip": 0.01161073,
      "auxiliary_loss_mlp": 0.01027839,
      "balance_loss_clip": 1.0515933,
      "balance_loss_mlp": 1.01981616,
      "epoch": 0.49035050802621294,
      "flos": 19174180089600.0,
      "grad_norm": 1.7379484858740346,
      "language_loss": 0.69796538,
      "learning_rate": 2.159519567776562e-06,
      "loss": 0.71985447,
      "num_input_tokens_seen": 87832605,
      "step": 4078,
      "time_per_iteration": 2.5077149868011475
    },
    {
      "auxiliary_loss_clip": 0.01122129,
      "auxiliary_loss_mlp": 0.01031386,
      "balance_loss_clip": 1.04398012,
      "balance_loss_mlp": 1.02318454,
      "epoch": 0.49047075091685205,
      "flos": 22228489365120.0,
      "grad_norm": 2.4493201324949943,
      "language_loss": 0.71230358,
      "learning_rate": 2.1587430657646463e-06,
      "loss": 0.73383868,
      "num_input_tokens_seen": 87846040,
      "step": 4079,
      "time_per_iteration": 2.6034135818481445
    },
    {
      "auxiliary_loss_clip": 0.01150514,
      "auxiliary_loss_mlp": 0.01032826,
      "balance_loss_clip": 1.05409706,
      "balance_loss_mlp": 1.02527952,
      "epoch": 0.4905909938074911,
      "flos": 20156516213760.0,
      "grad_norm": 1.790849935282541,
      "language_loss": 0.78108692,
      "learning_rate": 2.157966539671533e-06,
      "loss": 0.80292034,
      "num_input_tokens_seen": 87865680,
      "step": 4080,
      "time_per_iteration": 2.5437145233154297
    },
    {
      "auxiliary_loss_clip": 0.01139825,
      "auxiliary_loss_mlp": 0.01029113,
      "balance_loss_clip": 1.05032432,
      "balance_loss_mlp": 1.022053,
      "epoch": 0.4907112366981302,
      "flos": 17202217380480.0,
      "grad_norm": 1.8825246193916172,
      "language_loss": 0.67319083,
      "learning_rate": 2.157189989615021e-06,
      "loss": 0.69488025,
      "num_input_tokens_seen": 87884270,
      "step": 4081,
      "time_per_iteration": 2.5717525482177734
    },
    {
      "auxiliary_loss_clip": 0.01163903,
      "auxiliary_loss_mlp": 0.00762225,
      "balance_loss_clip": 1.05011427,
      "balance_loss_mlp": 1.00015759,
      "epoch": 0.4908314795887693,
      "flos": 21688968107520.0,
      "grad_norm": 2.9090867705678547,
      "language_loss": 0.75403357,
      "learning_rate": 2.156413415712913e-06,
      "loss": 0.77329481,
      "num_input_tokens_seen": 87906320,
      "step": 4082,
      "time_per_iteration": 2.5772855281829834
    },
    {
      "auxiliary_loss_clip": 0.01153938,
      "auxiliary_loss_mlp": 0.00761434,
      "balance_loss_clip": 1.05221987,
      "balance_loss_mlp": 1.00016856,
      "epoch": 0.4909517224794084,
      "flos": 26213676531840.0,
      "grad_norm": 1.6183906080018549,
      "language_loss": 0.78697866,
      "learning_rate": 2.155636818083014e-06,
      "loss": 0.80613238,
      "num_input_tokens_seen": 87927690,
      "step": 4083,
      "time_per_iteration": 2.588271379470825
    },
    {
      "auxiliary_loss_clip": 0.01146856,
      "auxiliary_loss_mlp": 0.01032617,
      "balance_loss_clip": 1.05075681,
      "balance_loss_mlp": 1.02532744,
      "epoch": 0.4910719653700475,
      "flos": 23148377694720.0,
      "grad_norm": 1.8048707165874864,
      "language_loss": 0.8396436,
      "learning_rate": 2.154860196843134e-06,
      "loss": 0.86143839,
      "num_input_tokens_seen": 87946885,
      "step": 4084,
      "time_per_iteration": 2.597961664199829
    },
    {
      "auxiliary_loss_clip": 0.01177569,
      "auxiliary_loss_mlp": 0.01028547,
      "balance_loss_clip": 1.05234957,
      "balance_loss_mlp": 1.0207386,
      "epoch": 0.4911922082606866,
      "flos": 23331845387520.0,
      "grad_norm": 2.056942888142575,
      "language_loss": 0.76894057,
      "learning_rate": 2.154083552111085e-06,
      "loss": 0.7910018,
      "num_input_tokens_seen": 87966055,
      "step": 4085,
      "time_per_iteration": 2.51446795463562
    },
    {
      "auxiliary_loss_clip": 0.01178908,
      "auxiliary_loss_mlp": 0.01027775,
      "balance_loss_clip": 1.05178881,
      "balance_loss_mlp": 1.01985383,
      "epoch": 0.49131245115132566,
      "flos": 29203239542400.0,
      "grad_norm": 1.6273680593012725,
      "language_loss": 0.81904781,
      "learning_rate": 2.1533068840046834e-06,
      "loss": 0.84111464,
      "num_input_tokens_seen": 87986320,
      "step": 4086,
      "time_per_iteration": 2.578711748123169
    },
    {
      "auxiliary_loss_clip": 0.01145908,
      "auxiliary_loss_mlp": 0.0076171,
      "balance_loss_clip": 1.05105031,
      "balance_loss_mlp": 1.00017917,
      "epoch": 0.49143269404196477,
      "flos": 20147465986560.0,
      "grad_norm": 3.237740827907876,
      "language_loss": 0.61697578,
      "learning_rate": 2.152530192641749e-06,
      "loss": 0.63605201,
      "num_input_tokens_seen": 88001230,
      "step": 4087,
      "time_per_iteration": 2.5416808128356934
    },
    {
      "auxiliary_loss_clip": 0.01165883,
      "auxiliary_loss_mlp": 0.01026605,
      "balance_loss_clip": 1.05228579,
      "balance_loss_mlp": 1.01923156,
      "epoch": 0.4915529369326039,
      "flos": 24389809597440.0,
      "grad_norm": 1.8998714739131946,
      "language_loss": 0.7273941,
      "learning_rate": 2.1517534781401068e-06,
      "loss": 0.74931896,
      "num_input_tokens_seen": 88019110,
      "step": 4088,
      "time_per_iteration": 2.5542490482330322
    },
    {
      "auxiliary_loss_clip": 0.01163153,
      "auxiliary_loss_mlp": 0.01028,
      "balance_loss_clip": 1.05113935,
      "balance_loss_mlp": 1.0206027,
      "epoch": 0.49167317982324293,
      "flos": 10524305197440.0,
      "grad_norm": 2.151768297455284,
      "language_loss": 0.69418269,
      "learning_rate": 2.150976740617581e-06,
      "loss": 0.71609426,
      "num_input_tokens_seen": 88035670,
      "step": 4089,
      "time_per_iteration": 3.2695202827453613
    },
    {
      "auxiliary_loss_clip": 0.01156421,
      "auxiliary_loss_mlp": 0.01032266,
      "balance_loss_clip": 1.0532012,
      "balance_loss_mlp": 1.02450514,
      "epoch": 0.49179342271388204,
      "flos": 25593427457280.0,
      "grad_norm": 1.9997206100011011,
      "language_loss": 0.71126676,
      "learning_rate": 2.150199980192006e-06,
      "loss": 0.73315358,
      "num_input_tokens_seen": 88054790,
      "step": 4090,
      "time_per_iteration": 2.5988078117370605
    },
    {
      "auxiliary_loss_clip": 0.01142877,
      "auxiliary_loss_mlp": 0.01026548,
      "balance_loss_clip": 1.04854751,
      "balance_loss_mlp": 1.01868653,
      "epoch": 0.49191366560452116,
      "flos": 21102043875840.0,
      "grad_norm": 1.5863816777045285,
      "language_loss": 0.80836058,
      "learning_rate": 2.1494231969812114e-06,
      "loss": 0.83005488,
      "num_input_tokens_seen": 88073780,
      "step": 4091,
      "time_per_iteration": 2.5230631828308105
    },
    {
      "auxiliary_loss_clip": 0.01140259,
      "auxiliary_loss_mlp": 0.01032069,
      "balance_loss_clip": 1.05032337,
      "balance_loss_mlp": 1.02472615,
      "epoch": 0.4920339084951602,
      "flos": 26067520091520.0,
      "grad_norm": 1.982135663228463,
      "language_loss": 0.80597854,
      "learning_rate": 2.1486463911030372e-06,
      "loss": 0.82770187,
      "num_input_tokens_seen": 88094430,
      "step": 4092,
      "time_per_iteration": 2.6100234985351562
    },
    {
      "auxiliary_loss_clip": 0.01147312,
      "auxiliary_loss_mlp": 0.01026658,
      "balance_loss_clip": 1.04796052,
      "balance_loss_mlp": 1.01917171,
      "epoch": 0.4921541513857993,
      "flos": 25081269384960.0,
      "grad_norm": 1.7634554366553603,
      "language_loss": 0.74462116,
      "learning_rate": 2.147869562675324e-06,
      "loss": 0.76636088,
      "num_input_tokens_seen": 88113400,
      "step": 4093,
      "time_per_iteration": 3.3397929668426514
    },
    {
      "auxiliary_loss_clip": 0.01166813,
      "auxiliary_loss_mlp": 0.0103683,
      "balance_loss_clip": 1.05411267,
      "balance_loss_mlp": 1.02794313,
      "epoch": 0.49227439427643843,
      "flos": 24389809597440.0,
      "grad_norm": 1.6509237990113101,
      "language_loss": 0.72252804,
      "learning_rate": 2.147092711815915e-06,
      "loss": 0.74456453,
      "num_input_tokens_seen": 88132750,
      "step": 4094,
      "time_per_iteration": 2.569129467010498
    },
    {
      "auxiliary_loss_clip": 0.01135893,
      "auxiliary_loss_mlp": 0.01031058,
      "balance_loss_clip": 1.05151331,
      "balance_loss_mlp": 1.02325535,
      "epoch": 0.4923946371670775,
      "flos": 11363753018880.0,
      "grad_norm": 2.78757949260212,
      "language_loss": 0.8650744,
      "learning_rate": 2.1463158386426593e-06,
      "loss": 0.8867439,
      "num_input_tokens_seen": 88150560,
      "step": 4095,
      "time_per_iteration": 3.2723093032836914
    },
    {
      "auxiliary_loss_clip": 0.01157327,
      "auxiliary_loss_mlp": 0.01027577,
      "balance_loss_clip": 1.05343342,
      "balance_loss_mlp": 1.01941156,
      "epoch": 0.4925148800577166,
      "flos": 30445964334720.0,
      "grad_norm": 2.6044202747462513,
      "language_loss": 0.77162236,
      "learning_rate": 2.145538943273407e-06,
      "loss": 0.7934714,
      "num_input_tokens_seen": 88170835,
      "step": 4096,
      "time_per_iteration": 3.4488463401794434
    },
    {
      "auxiliary_loss_clip": 0.01179348,
      "auxiliary_loss_mlp": 0.01032384,
      "balance_loss_clip": 1.05343246,
      "balance_loss_mlp": 1.02437329,
      "epoch": 0.49263512294835565,
      "flos": 20850454039680.0,
      "grad_norm": 2.177459688660364,
      "language_loss": 0.71779943,
      "learning_rate": 2.144762025826013e-06,
      "loss": 0.7399168,
      "num_input_tokens_seen": 88189925,
      "step": 4097,
      "time_per_iteration": 2.49027943611145
    },
    {
      "auxiliary_loss_clip": 0.01168669,
      "auxiliary_loss_mlp": 0.01031814,
      "balance_loss_clip": 1.05326521,
      "balance_loss_mlp": 1.02355874,
      "epoch": 0.49275536583899476,
      "flos": 23767477534080.0,
      "grad_norm": 2.918457134827686,
      "language_loss": 0.87204623,
      "learning_rate": 2.143985086418334e-06,
      "loss": 0.89405107,
      "num_input_tokens_seen": 88205105,
      "step": 4098,
      "time_per_iteration": 2.5130274295806885
    },
    {
      "auxiliary_loss_clip": 0.01152885,
      "auxiliary_loss_mlp": 0.01025323,
      "balance_loss_clip": 1.05128205,
      "balance_loss_mlp": 1.01814044,
      "epoch": 0.4928756087296339,
      "flos": 22273522041600.0,
      "grad_norm": 1.3349778110086583,
      "language_loss": 0.7657333,
      "learning_rate": 2.1432081251682324e-06,
      "loss": 0.7875154,
      "num_input_tokens_seen": 88225475,
      "step": 4099,
      "time_per_iteration": 2.5365042686462402
    },
    {
      "auxiliary_loss_clip": 0.01164189,
      "auxiliary_loss_mlp": 0.01027312,
      "balance_loss_clip": 1.05451417,
      "balance_loss_mlp": 1.01922369,
      "epoch": 0.49299585162027293,
      "flos": 19645471463040.0,
      "grad_norm": 1.7018370934955276,
      "language_loss": 0.86880267,
      "learning_rate": 2.142431142193572e-06,
      "loss": 0.89071763,
      "num_input_tokens_seen": 88243255,
      "step": 4100,
      "time_per_iteration": 2.497908353805542
    },
    {
      "auxiliary_loss_clip": 0.0117803,
      "auxiliary_loss_mlp": 0.0103358,
      "balance_loss_clip": 1.05447102,
      "balance_loss_mlp": 1.02643013,
      "epoch": 0.49311609451091204,
      "flos": 38837138497920.0,
      "grad_norm": 2.206713364879448,
      "language_loss": 0.71828365,
      "learning_rate": 2.1416541376122207e-06,
      "loss": 0.74039972,
      "num_input_tokens_seen": 88263435,
      "step": 4101,
      "time_per_iteration": 2.6191036701202393
    },
    {
      "auxiliary_loss_clip": 0.011789,
      "auxiliary_loss_mlp": 0.01032025,
      "balance_loss_clip": 1.05333352,
      "balance_loss_mlp": 1.02393055,
      "epoch": 0.49323633740155115,
      "flos": 28329102161280.0,
      "grad_norm": 1.6721328351008704,
      "language_loss": 0.72892427,
      "learning_rate": 2.1408771115420496e-06,
      "loss": 0.75103354,
      "num_input_tokens_seen": 88283295,
      "step": 4102,
      "time_per_iteration": 2.5364179611206055
    },
    {
      "auxiliary_loss_clip": 0.01128346,
      "auxiliary_loss_mlp": 0.01033032,
      "balance_loss_clip": 1.05378699,
      "balance_loss_mlp": 1.02552164,
      "epoch": 0.4933565802921902,
      "flos": 21135584200320.0,
      "grad_norm": 1.9374610797091183,
      "language_loss": 0.6483233,
      "learning_rate": 2.140100064100932e-06,
      "loss": 0.66993707,
      "num_input_tokens_seen": 88299270,
      "step": 4103,
      "time_per_iteration": 2.589509963989258
    },
    {
      "auxiliary_loss_clip": 0.01162108,
      "auxiliary_loss_mlp": 0.01029301,
      "balance_loss_clip": 1.0524627,
      "balance_loss_mlp": 1.0214808,
      "epoch": 0.4934768231828293,
      "flos": 18039007595520.0,
      "grad_norm": 2.0203283255523456,
      "language_loss": 0.75762922,
      "learning_rate": 2.139322995406746e-06,
      "loss": 0.77954328,
      "num_input_tokens_seen": 88316905,
      "step": 4104,
      "time_per_iteration": 2.462756633758545
    },
    {
      "auxiliary_loss_clip": 0.01182066,
      "auxiliary_loss_mlp": 0.01037552,
      "balance_loss_clip": 1.05649364,
      "balance_loss_mlp": 1.02942824,
      "epoch": 0.4935970660734684,
      "flos": 23469957181440.0,
      "grad_norm": 2.424513943100918,
      "language_loss": 0.79401016,
      "learning_rate": 2.1385459055773727e-06,
      "loss": 0.81620646,
      "num_input_tokens_seen": 88335095,
      "step": 4105,
      "time_per_iteration": 2.492323398590088
    },
    {
      "auxiliary_loss_clip": 0.01109823,
      "auxiliary_loss_mlp": 0.00760932,
      "balance_loss_clip": 1.04347062,
      "balance_loss_mlp": 1.00018728,
      "epoch": 0.4937173089641075,
      "flos": 64479258840960.0,
      "grad_norm": 2.075186886873627,
      "language_loss": 0.7409814,
      "learning_rate": 2.137768794730696e-06,
      "loss": 0.75968897,
      "num_input_tokens_seen": 88358545,
      "step": 4106,
      "time_per_iteration": 2.998983383178711
    },
    {
      "auxiliary_loss_clip": 0.01157358,
      "auxiliary_loss_mlp": 0.01035041,
      "balance_loss_clip": 1.05429769,
      "balance_loss_mlp": 1.02685118,
      "epoch": 0.4938375518547466,
      "flos": 22346025644160.0,
      "grad_norm": 1.748475360599871,
      "language_loss": 0.80381751,
      "learning_rate": 2.1369916629846026e-06,
      "loss": 0.82574147,
      "num_input_tokens_seen": 88378295,
      "step": 4107,
      "time_per_iteration": 2.5438485145568848
    },
    {
      "auxiliary_loss_clip": 0.01151975,
      "auxiliary_loss_mlp": 0.01027265,
      "balance_loss_clip": 1.05047667,
      "balance_loss_mlp": 1.01983571,
      "epoch": 0.4939577947453857,
      "flos": 17858700299520.0,
      "grad_norm": 3.0102353892912883,
      "language_loss": 0.74903703,
      "learning_rate": 2.136214510456983e-06,
      "loss": 0.77082944,
      "num_input_tokens_seen": 88396750,
      "step": 4108,
      "time_per_iteration": 2.499143123626709
    },
    {
      "auxiliary_loss_clip": 0.01035946,
      "auxiliary_loss_mlp": 0.00751502,
      "balance_loss_clip": 1.01938808,
      "balance_loss_mlp": 1.00006902,
      "epoch": 0.49407803763602476,
      "flos": 70066746875520.0,
      "grad_norm": 0.902609334423674,
      "language_loss": 0.63166678,
      "learning_rate": 2.1354373372657296e-06,
      "loss": 0.64954126,
      "num_input_tokens_seen": 88455190,
      "step": 4109,
      "time_per_iteration": 3.204637050628662
    },
    {
      "auxiliary_loss_clip": 0.01178091,
      "auxiliary_loss_mlp": 0.0103436,
      "balance_loss_clip": 1.05567837,
      "balance_loss_mlp": 1.02618814,
      "epoch": 0.49419828052666387,
      "flos": 24317485562880.0,
      "grad_norm": 1.4524144752389598,
      "language_loss": 0.71105587,
      "learning_rate": 2.1346601435287404e-06,
      "loss": 0.73318046,
      "num_input_tokens_seen": 88477460,
      "step": 4110,
      "time_per_iteration": 2.4971580505371094
    },
    {
      "auxiliary_loss_clip": 0.01147599,
      "auxiliary_loss_mlp": 0.01037003,
      "balance_loss_clip": 1.04760861,
      "balance_loss_mlp": 1.02963555,
      "epoch": 0.494318523417303,
      "flos": 29386060790400.0,
      "grad_norm": 1.7133734042410407,
      "language_loss": 0.80106771,
      "learning_rate": 2.1338829293639144e-06,
      "loss": 0.82291371,
      "num_input_tokens_seen": 88497820,
      "step": 4111,
      "time_per_iteration": 2.599439859390259
    },
    {
      "auxiliary_loss_clip": 0.01118812,
      "auxiliary_loss_mlp": 0.0103087,
      "balance_loss_clip": 1.04597521,
      "balance_loss_mlp": 1.02329993,
      "epoch": 0.49443876630794203,
      "flos": 15268284195840.0,
      "grad_norm": 2.5259661442667674,
      "language_loss": 0.83020836,
      "learning_rate": 2.1331056948891547e-06,
      "loss": 0.85170519,
      "num_input_tokens_seen": 88514920,
      "step": 4112,
      "time_per_iteration": 2.5853421688079834
    },
    {
      "auxiliary_loss_clip": 0.01147277,
      "auxiliary_loss_mlp": 0.01025498,
      "balance_loss_clip": 1.05090284,
      "balance_loss_mlp": 1.01732659,
      "epoch": 0.49455900919858115,
      "flos": 12347453859840.0,
      "grad_norm": 2.1411361041668733,
      "language_loss": 0.7600106,
      "learning_rate": 2.1323284402223666e-06,
      "loss": 0.7817384,
      "num_input_tokens_seen": 88530910,
      "step": 4113,
      "time_per_iteration": 2.508910655975342
    },
    {
      "auxiliary_loss_clip": 0.0117679,
      "auxiliary_loss_mlp": 0.00760224,
      "balance_loss_clip": 1.05670714,
      "balance_loss_mlp": 1.00017619,
      "epoch": 0.4946792520892202,
      "flos": 22779610715520.0,
      "grad_norm": 1.7994431446400758,
      "language_loss": 0.88147527,
      "learning_rate": 2.1315511654814597e-06,
      "loss": 0.90084541,
      "num_input_tokens_seen": 88549320,
      "step": 4114,
      "time_per_iteration": 2.508526563644409
    },
    {
      "auxiliary_loss_clip": 0.01145899,
      "auxiliary_loss_mlp": 0.01032088,
      "balance_loss_clip": 1.05214357,
      "balance_loss_mlp": 1.02542126,
      "epoch": 0.4947994949798593,
      "flos": 23148126299520.0,
      "grad_norm": 1.7532034367466658,
      "language_loss": 0.78349084,
      "learning_rate": 2.1307738707843456e-06,
      "loss": 0.80527073,
      "num_input_tokens_seen": 88568985,
      "step": 4115,
      "time_per_iteration": 3.3329477310180664
    },
    {
      "auxiliary_loss_clip": 0.01171405,
      "auxiliary_loss_mlp": 0.0103895,
      "balance_loss_clip": 1.05506158,
      "balance_loss_mlp": 1.03023577,
      "epoch": 0.4949197378704984,
      "flos": 23659997063040.0,
      "grad_norm": 1.945609180396013,
      "language_loss": 0.6934334,
      "learning_rate": 2.1299965562489385e-06,
      "loss": 0.71553689,
      "num_input_tokens_seen": 88588790,
      "step": 4116,
      "time_per_iteration": 2.540602922439575
    },
    {
      "auxiliary_loss_clip": 0.01161585,
      "auxiliary_loss_mlp": 0.01032441,
      "balance_loss_clip": 1.05047154,
      "balance_loss_mlp": 1.02516675,
      "epoch": 0.4950399807611375,
      "flos": 26911493026560.0,
      "grad_norm": 1.3258579220085758,
      "language_loss": 0.78690434,
      "learning_rate": 2.129219221993158e-06,
      "loss": 0.80884463,
      "num_input_tokens_seen": 88613575,
      "step": 4117,
      "time_per_iteration": 2.5905632972717285
    },
    {
      "auxiliary_loss_clip": 0.01040612,
      "auxiliary_loss_mlp": 0.01004339,
      "balance_loss_clip": 1.01929367,
      "balance_loss_mlp": 1.00320625,
      "epoch": 0.4951602236517766,
      "flos": 67315270187520.0,
      "grad_norm": 0.7896809960680853,
      "language_loss": 0.59993327,
      "learning_rate": 2.128441868134924e-06,
      "loss": 0.62038279,
      "num_input_tokens_seen": 88675510,
      "step": 4118,
      "time_per_iteration": 3.188976764678955
    },
    {
      "auxiliary_loss_clip": 0.01137411,
      "auxiliary_loss_mlp": 0.01031323,
      "balance_loss_clip": 1.04847789,
      "balance_loss_mlp": 1.0239048,
      "epoch": 0.4952804665424157,
      "flos": 19901442758400.0,
      "grad_norm": 2.252894202480421,
      "language_loss": 0.82677627,
      "learning_rate": 2.1276644947921606e-06,
      "loss": 0.84846354,
      "num_input_tokens_seen": 88694425,
      "step": 4119,
      "time_per_iteration": 3.3306000232696533
    },
    {
      "auxiliary_loss_clip": 0.01161603,
      "auxiliary_loss_mlp": 0.01029105,
      "balance_loss_clip": 1.05065155,
      "balance_loss_mlp": 1.02076101,
      "epoch": 0.49540070943305475,
      "flos": 18806813740800.0,
      "grad_norm": 1.7770651860566589,
      "language_loss": 0.82257318,
      "learning_rate": 2.126887102082795e-06,
      "loss": 0.84448028,
      "num_input_tokens_seen": 88714450,
      "step": 4120,
      "time_per_iteration": 2.487952947616577
    },
    {
      "auxiliary_loss_clip": 0.01135261,
      "auxiliary_loss_mlp": 0.01031313,
      "balance_loss_clip": 1.04627776,
      "balance_loss_mlp": 1.02407742,
      "epoch": 0.49552095232369386,
      "flos": 24934179191040.0,
      "grad_norm": 1.729039384768213,
      "language_loss": 0.7031076,
      "learning_rate": 2.126109690124757e-06,
      "loss": 0.72477335,
      "num_input_tokens_seen": 88735265,
      "step": 4121,
      "time_per_iteration": 3.3127381801605225
    },
    {
      "auxiliary_loss_clip": 0.01125476,
      "auxiliary_loss_mlp": 0.01026951,
      "balance_loss_clip": 1.04902315,
      "balance_loss_mlp": 1.01947045,
      "epoch": 0.495641195214333,
      "flos": 22857249962880.0,
      "grad_norm": 1.5732529297067925,
      "language_loss": 0.71537805,
      "learning_rate": 2.1253322590359786e-06,
      "loss": 0.7369023,
      "num_input_tokens_seen": 88754600,
      "step": 4122,
      "time_per_iteration": 2.592883348464966
    },
    {
      "auxiliary_loss_clip": 0.01163046,
      "auxiliary_loss_mlp": 0.01034176,
      "balance_loss_clip": 1.05138791,
      "balance_loss_mlp": 1.02658844,
      "epoch": 0.49576143810497203,
      "flos": 25769748343680.0,
      "grad_norm": 1.575374183826372,
      "language_loss": 0.74081182,
      "learning_rate": 2.124554808934397e-06,
      "loss": 0.762784,
      "num_input_tokens_seen": 88775180,
      "step": 4123,
      "time_per_iteration": 3.348723888397217
    },
    {
      "auxiliary_loss_clip": 0.01113796,
      "auxiliary_loss_mlp": 0.01031677,
      "balance_loss_clip": 1.04220986,
      "balance_loss_mlp": 1.02414298,
      "epoch": 0.49588168099561114,
      "flos": 22128838058880.0,
      "grad_norm": 1.7083068582222076,
      "language_loss": 0.72967529,
      "learning_rate": 2.1237773399379496e-06,
      "loss": 0.75112998,
      "num_input_tokens_seen": 88796145,
      "step": 4124,
      "time_per_iteration": 2.6150074005126953
    },
    {
      "auxiliary_loss_clip": 0.01152756,
      "auxiliary_loss_mlp": 0.01028975,
      "balance_loss_clip": 1.04819787,
      "balance_loss_mlp": 1.02094579,
      "epoch": 0.49600192388625025,
      "flos": 24387331559040.0,
      "grad_norm": 1.657994744657819,
      "language_loss": 0.87036741,
      "learning_rate": 2.122999852164578e-06,
      "loss": 0.89218473,
      "num_input_tokens_seen": 88816765,
      "step": 4125,
      "time_per_iteration": 2.556964635848999
    },
    {
      "auxiliary_loss_clip": 0.01120369,
      "auxiliary_loss_mlp": 0.01028896,
      "balance_loss_clip": 1.04913497,
      "balance_loss_mlp": 1.02105212,
      "epoch": 0.4961221667768893,
      "flos": 22857429530880.0,
      "grad_norm": 2.262996378218739,
      "language_loss": 0.58849156,
      "learning_rate": 2.122222345732227e-06,
      "loss": 0.60998416,
      "num_input_tokens_seen": 88836680,
      "step": 4126,
      "time_per_iteration": 2.6030972003936768
    },
    {
      "auxiliary_loss_clip": 0.01136732,
      "auxiliary_loss_mlp": 0.01030732,
      "balance_loss_clip": 1.04809618,
      "balance_loss_mlp": 1.0228492,
      "epoch": 0.4962424096675284,
      "flos": 17858089768320.0,
      "grad_norm": 1.6555735875777922,
      "language_loss": 0.82764208,
      "learning_rate": 2.121444820758843e-06,
      "loss": 0.84931672,
      "num_input_tokens_seen": 88855320,
      "step": 4127,
      "time_per_iteration": 2.554067373275757
    },
    {
      "auxiliary_loss_clip": 0.01119911,
      "auxiliary_loss_mlp": 0.01032993,
      "balance_loss_clip": 1.04862988,
      "balance_loss_mlp": 1.02497625,
      "epoch": 0.49636265255816747,
      "flos": 21793611404160.0,
      "grad_norm": 1.967197102579569,
      "language_loss": 0.78690612,
      "learning_rate": 2.120667277362376e-06,
      "loss": 0.8084352,
      "num_input_tokens_seen": 88874035,
      "step": 4128,
      "time_per_iteration": 2.594515800476074
    },
    {
      "auxiliary_loss_clip": 0.0118236,
      "auxiliary_loss_mlp": 0.010268,
      "balance_loss_clip": 1.0568099,
      "balance_loss_mlp": 1.01864648,
      "epoch": 0.4964828954488066,
      "flos": 16358603581440.0,
      "grad_norm": 1.8957120822679856,
      "language_loss": 0.85178983,
      "learning_rate": 2.1198897156607796e-06,
      "loss": 0.8738814,
      "num_input_tokens_seen": 88891390,
      "step": 4129,
      "time_per_iteration": 2.4333302974700928
    },
    {
      "auxiliary_loss_clip": 0.01167511,
      "auxiliary_loss_mlp": 0.01033022,
      "balance_loss_clip": 1.05245018,
      "balance_loss_mlp": 1.02526188,
      "epoch": 0.4966031383394457,
      "flos": 24711101775360.0,
      "grad_norm": 2.4210931040895005,
      "language_loss": 0.73310387,
      "learning_rate": 2.1191121357720085e-06,
      "loss": 0.75510925,
      "num_input_tokens_seen": 88909450,
      "step": 4130,
      "time_per_iteration": 2.5157558917999268
    },
    {
      "auxiliary_loss_clip": 0.01115445,
      "auxiliary_loss_mlp": 0.0103212,
      "balance_loss_clip": 1.04705024,
      "balance_loss_mlp": 1.02454412,
      "epoch": 0.49672338123008475,
      "flos": 22930615491840.0,
      "grad_norm": 1.8234506545270768,
      "language_loss": 0.74377966,
      "learning_rate": 2.1183345378140206e-06,
      "loss": 0.76525533,
      "num_input_tokens_seen": 88929195,
      "step": 4131,
      "time_per_iteration": 2.6002607345581055
    },
    {
      "auxiliary_loss_clip": 0.01061406,
      "auxiliary_loss_mlp": 0.01010184,
      "balance_loss_clip": 1.01664901,
      "balance_loss_mlp": 1.00892639,
      "epoch": 0.49684362412072386,
      "flos": 65976736844160.0,
      "grad_norm": 0.8519891515345855,
      "language_loss": 0.61991847,
      "learning_rate": 2.1175569219047783e-06,
      "loss": 0.64063442,
      "num_input_tokens_seen": 88990635,
      "step": 4132,
      "time_per_iteration": 3.2120490074157715
    },
    {
      "auxiliary_loss_clip": 0.01177488,
      "auxiliary_loss_mlp": 0.01033308,
      "balance_loss_clip": 1.05357111,
      "balance_loss_mlp": 1.02585125,
      "epoch": 0.49696386701136297,
      "flos": 19971288754560.0,
      "grad_norm": 1.9133935785879337,
      "language_loss": 0.73331118,
      "learning_rate": 2.1167792881622437e-06,
      "loss": 0.75541914,
      "num_input_tokens_seen": 89009655,
      "step": 4133,
      "time_per_iteration": 2.479883909225464
    },
    {
      "auxiliary_loss_clip": 0.0114861,
      "auxiliary_loss_mlp": 0.01027374,
      "balance_loss_clip": 1.05444872,
      "balance_loss_mlp": 1.01962256,
      "epoch": 0.497084109902002,
      "flos": 24750819239040.0,
      "grad_norm": 1.7145511489382614,
      "language_loss": 0.81107306,
      "learning_rate": 2.116001636704384e-06,
      "loss": 0.83283293,
      "num_input_tokens_seen": 89030040,
      "step": 4134,
      "time_per_iteration": 2.554598331451416
    },
    {
      "auxiliary_loss_clip": 0.0113029,
      "auxiliary_loss_mlp": 0.01027209,
      "balance_loss_clip": 1.04912591,
      "balance_loss_mlp": 1.01917386,
      "epoch": 0.49720435279264114,
      "flos": 21871825269120.0,
      "grad_norm": 4.057534061135621,
      "language_loss": 0.80483896,
      "learning_rate": 2.1152239676491685e-06,
      "loss": 0.82641399,
      "num_input_tokens_seen": 89048145,
      "step": 4135,
      "time_per_iteration": 2.5975704193115234
    },
    {
      "auxiliary_loss_clip": 0.0115528,
      "auxiliary_loss_mlp": 0.01026097,
      "balance_loss_clip": 1.0500195,
      "balance_loss_mlp": 1.01847339,
      "epoch": 0.49732459568328025,
      "flos": 23805794367360.0,
      "grad_norm": 1.7468151034275665,
      "language_loss": 0.73347038,
      "learning_rate": 2.114446281114569e-06,
      "loss": 0.75528419,
      "num_input_tokens_seen": 89067165,
      "step": 4136,
      "time_per_iteration": 2.5571212768554688
    },
    {
      "auxiliary_loss_clip": 0.01143517,
      "auxiliary_loss_mlp": 0.01042098,
      "balance_loss_clip": 1.05055523,
      "balance_loss_mlp": 1.0345397,
      "epoch": 0.4974448385739193,
      "flos": 20047742853120.0,
      "grad_norm": 1.9585269528026878,
      "language_loss": 0.75991654,
      "learning_rate": 2.1136685772185587e-06,
      "loss": 0.78177273,
      "num_input_tokens_seen": 89086190,
      "step": 4137,
      "time_per_iteration": 2.5242931842803955
    },
    {
      "auxiliary_loss_clip": 0.01149596,
      "auxiliary_loss_mlp": 0.00761956,
      "balance_loss_clip": 1.04540873,
      "balance_loss_mlp": 1.00017703,
      "epoch": 0.4975650814645584,
      "flos": 24821347593600.0,
      "grad_norm": 1.5997846234397637,
      "language_loss": 0.77929437,
      "learning_rate": 2.1128908560791163e-06,
      "loss": 0.79840994,
      "num_input_tokens_seen": 89106020,
      "step": 4138,
      "time_per_iteration": 2.573556661605835
    },
    {
      "auxiliary_loss_clip": 0.01179039,
      "auxiliary_loss_mlp": 0.0103448,
      "balance_loss_clip": 1.05534947,
      "balance_loss_mlp": 1.02626097,
      "epoch": 0.4976853243551975,
      "flos": 19829477859840.0,
      "grad_norm": 1.8638726751558847,
      "language_loss": 0.78572941,
      "learning_rate": 2.1121131178142203e-06,
      "loss": 0.80786467,
      "num_input_tokens_seen": 89125385,
      "step": 4139,
      "time_per_iteration": 2.4593639373779297
    },
    {
      "auxiliary_loss_clip": 0.01148047,
      "auxiliary_loss_mlp": 0.0102675,
      "balance_loss_clip": 1.0474416,
      "balance_loss_mlp": 1.01906705,
      "epoch": 0.4978055672458366,
      "flos": 23142990654720.0,
      "grad_norm": 1.4542402657673086,
      "language_loss": 0.82282186,
      "learning_rate": 2.1113353625418544e-06,
      "loss": 0.84456986,
      "num_input_tokens_seen": 89143935,
      "step": 4140,
      "time_per_iteration": 2.5740013122558594
    },
    {
      "auxiliary_loss_clip": 0.01165494,
      "auxiliary_loss_mlp": 0.01029668,
      "balance_loss_clip": 1.05731058,
      "balance_loss_mlp": 1.02224171,
      "epoch": 0.4979258101364757,
      "flos": 15559914718080.0,
      "grad_norm": 1.5943686835394164,
      "language_loss": 0.78957248,
      "learning_rate": 2.1105575903800017e-06,
      "loss": 0.81152409,
      "num_input_tokens_seen": 89162655,
      "step": 4141,
      "time_per_iteration": 3.30733060836792
    },
    {
      "auxiliary_loss_clip": 0.01168467,
      "auxiliary_loss_mlp": 0.01039388,
      "balance_loss_clip": 1.0530616,
      "balance_loss_mlp": 1.03160715,
      "epoch": 0.4980460530271148,
      "flos": 26356169784960.0,
      "grad_norm": 1.749433591963683,
      "language_loss": 0.8504833,
      "learning_rate": 2.1097798014466502e-06,
      "loss": 0.87256181,
      "num_input_tokens_seen": 89182255,
      "step": 4142,
      "time_per_iteration": 2.5389459133148193
    },
    {
      "auxiliary_loss_clip": 0.01168126,
      "auxiliary_loss_mlp": 0.01034352,
      "balance_loss_clip": 1.05336189,
      "balance_loss_mlp": 1.02632356,
      "epoch": 0.49816629591775385,
      "flos": 17274541415040.0,
      "grad_norm": 2.4086696858656453,
      "language_loss": 0.59181774,
      "learning_rate": 2.109001995859791e-06,
      "loss": 0.61384249,
      "num_input_tokens_seen": 89201155,
      "step": 4143,
      "time_per_iteration": 2.478395700454712
    },
    {
      "auxiliary_loss_clip": 0.0104691,
      "auxiliary_loss_mlp": 0.01002331,
      "balance_loss_clip": 1.01585317,
      "balance_loss_mlp": 1.00116909,
      "epoch": 0.49828653880839296,
      "flos": 64930947344640.0,
      "grad_norm": 0.7933570114983841,
      "language_loss": 0.60061121,
      "learning_rate": 2.108224173737415e-06,
      "loss": 0.62110353,
      "num_input_tokens_seen": 89264455,
      "step": 4144,
      "time_per_iteration": 3.161641836166382
    },
    {
      "auxiliary_loss_clip": 0.01145748,
      "auxiliary_loss_mlp": 0.01031089,
      "balance_loss_clip": 1.04768622,
      "balance_loss_mlp": 1.02308702,
      "epoch": 0.498406781699032,
      "flos": 27484806003840.0,
      "grad_norm": 1.769731746884637,
      "language_loss": 0.76225603,
      "learning_rate": 2.1074463351975183e-06,
      "loss": 0.78402436,
      "num_input_tokens_seen": 89283340,
      "step": 4145,
      "time_per_iteration": 3.3626086711883545
    },
    {
      "auxiliary_loss_clip": 0.01139496,
      "auxiliary_loss_mlp": 0.01032986,
      "balance_loss_clip": 1.05050921,
      "balance_loss_mlp": 1.02559257,
      "epoch": 0.49852702458967113,
      "flos": 31499870307840.0,
      "grad_norm": 1.7761163191300797,
      "language_loss": 0.71566433,
      "learning_rate": 2.106668480358098e-06,
      "loss": 0.73738915,
      "num_input_tokens_seen": 89303565,
      "step": 4146,
      "time_per_iteration": 2.63297963142395
    },
    {
      "auxiliary_loss_clip": 0.01144735,
      "auxiliary_loss_mlp": 0.01024828,
      "balance_loss_clip": 1.04910302,
      "balance_loss_mlp": 1.01653135,
      "epoch": 0.49864726748031024,
      "flos": 22852868503680.0,
      "grad_norm": 1.7322204005742854,
      "language_loss": 0.70784891,
      "learning_rate": 2.105890609337154e-06,
      "loss": 0.72954452,
      "num_input_tokens_seen": 89322080,
      "step": 4147,
      "time_per_iteration": 3.311931848526001
    },
    {
      "auxiliary_loss_clip": 0.01070423,
      "auxiliary_loss_mlp": 0.0100047,
      "balance_loss_clip": 1.01766753,
      "balance_loss_mlp": 0.99924815,
      "epoch": 0.4987675103709493,
      "flos": 70405708544640.0,
      "grad_norm": 0.6892486820123155,
      "language_loss": 0.63864636,
      "learning_rate": 2.1051127222526883e-06,
      "loss": 0.65935528,
      "num_input_tokens_seen": 89394195,
      "step": 4148,
      "time_per_iteration": 3.1817545890808105
    },
    {
      "auxiliary_loss_clip": 0.01165057,
      "auxiliary_loss_mlp": 0.01024578,
      "balance_loss_clip": 1.05593157,
      "balance_loss_mlp": 1.01727295,
      "epoch": 0.4988877532615884,
      "flos": 28767571482240.0,
      "grad_norm": 1.6004983534681403,
      "language_loss": 0.80643386,
      "learning_rate": 2.1043348192227067e-06,
      "loss": 0.82833016,
      "num_input_tokens_seen": 89414565,
      "step": 4149,
      "time_per_iteration": 3.3491454124450684
    },
    {
      "auxiliary_loss_clip": 0.01127805,
      "auxiliary_loss_mlp": 0.01033131,
      "balance_loss_clip": 1.04977751,
      "balance_loss_mlp": 1.02617192,
      "epoch": 0.4990079961522275,
      "flos": 16872700988160.0,
      "grad_norm": 1.8169041872467049,
      "language_loss": 0.61740887,
      "learning_rate": 2.1035569003652156e-06,
      "loss": 0.6390183,
      "num_input_tokens_seen": 89433195,
      "step": 4150,
      "time_per_iteration": 2.575763702392578
    },
    {
      "auxiliary_loss_clip": 0.0112155,
      "auxiliary_loss_mlp": 0.01033038,
      "balance_loss_clip": 1.04833043,
      "balance_loss_mlp": 1.02451396,
      "epoch": 0.4991282390428666,
      "flos": 13291042187520.0,
      "grad_norm": 1.9879549648538788,
      "language_loss": 0.81356823,
      "learning_rate": 2.1027789657982255e-06,
      "loss": 0.83511412,
      "num_input_tokens_seen": 89447410,
      "step": 4151,
      "time_per_iteration": 2.5380499362945557
    },
    {
      "auxiliary_loss_clip": 0.01124009,
      "auxiliary_loss_mlp": 0.01025955,
      "balance_loss_clip": 1.05113745,
      "balance_loss_mlp": 1.01809919,
      "epoch": 0.4992484819335057,
      "flos": 21537496454400.0,
      "grad_norm": 1.866633289057137,
      "language_loss": 0.77052605,
      "learning_rate": 2.1020010156397482e-06,
      "loss": 0.79202569,
      "num_input_tokens_seen": 89464630,
      "step": 4152,
      "time_per_iteration": 2.60009765625
    },
    {
      "auxiliary_loss_clip": 0.01164788,
      "auxiliary_loss_mlp": 0.01030254,
      "balance_loss_clip": 1.05229187,
      "balance_loss_mlp": 1.02248788,
      "epoch": 0.4993687248241448,
      "flos": 24860095390080.0,
      "grad_norm": 1.536899164926935,
      "language_loss": 0.77268744,
      "learning_rate": 2.101223050007797e-06,
      "loss": 0.79463792,
      "num_input_tokens_seen": 89483180,
      "step": 4153,
      "time_per_iteration": 2.5271964073181152
    },
    {
      "auxiliary_loss_clip": 0.01069785,
      "auxiliary_loss_mlp": 0.01000766,
      "balance_loss_clip": 1.01721942,
      "balance_loss_mlp": 0.99955654,
      "epoch": 0.49948896771478385,
      "flos": 62941602453120.0,
      "grad_norm": 0.8192399197389844,
      "language_loss": 0.5385918,
      "learning_rate": 2.1004450690203904e-06,
      "loss": 0.55929732,
      "num_input_tokens_seen": 89539260,
      "step": 4154,
      "time_per_iteration": 3.1213138103485107
    },
    {
      "auxiliary_loss_clip": 0.01069197,
      "auxiliary_loss_mlp": 0.01000537,
      "balance_loss_clip": 1.01655769,
      "balance_loss_mlp": 0.99930328,
      "epoch": 0.49960921060542296,
      "flos": 68284213516800.0,
      "grad_norm": 0.8520163787265298,
      "language_loss": 0.63357306,
      "learning_rate": 2.099667072795546e-06,
      "loss": 0.65427035,
      "num_input_tokens_seen": 89601380,
      "step": 4155,
      "time_per_iteration": 3.1359729766845703
    },
    {
      "auxiliary_loss_clip": 0.01162262,
      "auxiliary_loss_mlp": 0.01025622,
      "balance_loss_clip": 1.051368,
      "balance_loss_mlp": 1.01848745,
      "epoch": 0.49972945349606207,
      "flos": 23659350618240.0,
      "grad_norm": 1.8489233767116153,
      "language_loss": 0.79844034,
      "learning_rate": 2.0988890614512864e-06,
      "loss": 0.82031918,
      "num_input_tokens_seen": 89621270,
      "step": 4156,
      "time_per_iteration": 2.5088822841644287
    },
    {
      "auxiliary_loss_clip": 0.01155043,
      "auxiliary_loss_mlp": 0.01028185,
      "balance_loss_clip": 1.05414093,
      "balance_loss_mlp": 1.02052021,
      "epoch": 0.4998496963867011,
      "flos": 19755825022080.0,
      "grad_norm": 1.751985588006529,
      "language_loss": 0.84343684,
      "learning_rate": 2.098111035105635e-06,
      "loss": 0.86526912,
      "num_input_tokens_seen": 89639695,
      "step": 4157,
      "time_per_iteration": 2.5491511821746826
    },
    {
      "auxiliary_loss_clip": 0.01121976,
      "auxiliary_loss_mlp": 0.01033299,
      "balance_loss_clip": 1.0530138,
      "balance_loss_mlp": 1.02546096,
      "epoch": 0.49996993927734024,
      "flos": 22265728790400.0,
      "grad_norm": 1.6143187047978216,
      "language_loss": 0.72947335,
      "learning_rate": 2.0973329938766176e-06,
      "loss": 0.75102609,
      "num_input_tokens_seen": 89657125,
      "step": 4158,
      "time_per_iteration": 2.569042205810547
    },
    {
      "auxiliary_loss_clip": 0.01170771,
      "auxiliary_loss_mlp": 0.01033381,
      "balance_loss_clip": 1.05442297,
      "balance_loss_mlp": 1.02498305,
      "epoch": 0.5000901821679793,
      "flos": 23327212533120.0,
      "grad_norm": 2.55334245985013,
      "language_loss": 0.78650689,
      "learning_rate": 2.0965549378822618e-06,
      "loss": 0.80854839,
      "num_input_tokens_seen": 89678415,
      "step": 4159,
      "time_per_iteration": 2.5390241146087646
    },
    {
      "auxiliary_loss_clip": 0.01075835,
      "auxiliary_loss_mlp": 0.01032195,
      "balance_loss_clip": 1.04328394,
      "balance_loss_mlp": 1.02424121,
      "epoch": 0.5002104250586185,
      "flos": 20339014239360.0,
      "grad_norm": 1.8919775288650955,
      "language_loss": 0.83957708,
      "learning_rate": 2.095776867240599e-06,
      "loss": 0.86065739,
      "num_input_tokens_seen": 89695405,
      "step": 4160,
      "time_per_iteration": 2.668036937713623
    },
    {
      "auxiliary_loss_clip": 0.01132489,
      "auxiliary_loss_mlp": 0.01033993,
      "balance_loss_clip": 1.04832447,
      "balance_loss_mlp": 1.02636647,
      "epoch": 0.5003306679492575,
      "flos": 13991372634240.0,
      "grad_norm": 2.0132264537897973,
      "language_loss": 0.82610387,
      "learning_rate": 2.094998782069661e-06,
      "loss": 0.84776866,
      "num_input_tokens_seen": 89713110,
      "step": 4161,
      "time_per_iteration": 2.554746627807617
    },
    {
      "auxiliary_loss_clip": 0.01176486,
      "auxiliary_loss_mlp": 0.0102803,
      "balance_loss_clip": 1.05281913,
      "balance_loss_mlp": 1.01985216,
      "epoch": 0.5004509108398966,
      "flos": 27672762896640.0,
      "grad_norm": 1.743975916764495,
      "language_loss": 0.75405288,
      "learning_rate": 2.0942206824874845e-06,
      "loss": 0.77609807,
      "num_input_tokens_seen": 89735885,
      "step": 4162,
      "time_per_iteration": 2.5166373252868652
    },
    {
      "auxiliary_loss_clip": 0.0116186,
      "auxiliary_loss_mlp": 0.01025988,
      "balance_loss_clip": 1.05199146,
      "balance_loss_mlp": 1.01800096,
      "epoch": 0.5005711537305357,
      "flos": 14976186796800.0,
      "grad_norm": 2.0158587188373773,
      "language_loss": 0.78963685,
      "learning_rate": 2.093442568612105e-06,
      "loss": 0.81151527,
      "num_input_tokens_seen": 89753690,
      "step": 4163,
      "time_per_iteration": 2.4764583110809326
    },
    {
      "auxiliary_loss_clip": 0.01178324,
      "auxiliary_loss_mlp": 0.01024323,
      "balance_loss_clip": 1.05359054,
      "balance_loss_mlp": 1.01664543,
      "epoch": 0.5006913966211748,
      "flos": 26503259978880.0,
      "grad_norm": 1.5289662338513503,
      "language_loss": 0.85118699,
      "learning_rate": 2.0926644405615613e-06,
      "loss": 0.87321347,
      "num_input_tokens_seen": 89774590,
      "step": 4164,
      "time_per_iteration": 2.509922504425049
    },
    {
      "auxiliary_loss_clip": 0.01133091,
      "auxiliary_loss_mlp": 0.01032747,
      "balance_loss_clip": 1.05059099,
      "balance_loss_mlp": 1.02536833,
      "epoch": 0.5008116395118138,
      "flos": 20449295971200.0,
      "grad_norm": 4.205800256962008,
      "language_loss": 0.81262183,
      "learning_rate": 2.091886298453897e-06,
      "loss": 0.83428025,
      "num_input_tokens_seen": 89792775,
      "step": 4165,
      "time_per_iteration": 2.5400145053863525
    },
    {
      "auxiliary_loss_clip": 0.01159769,
      "auxiliary_loss_mlp": 0.01038075,
      "balance_loss_clip": 1.04971814,
      "balance_loss_mlp": 1.03066969,
      "epoch": 0.500931882402453,
      "flos": 21579871524480.0,
      "grad_norm": 1.8706031384962958,
      "language_loss": 0.73170936,
      "learning_rate": 2.091108142407153e-06,
      "loss": 0.75368786,
      "num_input_tokens_seen": 89811515,
      "step": 4166,
      "time_per_iteration": 3.262004852294922
    },
    {
      "auxiliary_loss_clip": 0.01045975,
      "auxiliary_loss_mlp": 0.01006986,
      "balance_loss_clip": 1.01785231,
      "balance_loss_mlp": 1.00579441,
      "epoch": 0.5010521252930921,
      "flos": 57785011925760.0,
      "grad_norm": 0.8324467400861327,
      "language_loss": 0.62391603,
      "learning_rate": 2.090329972539377e-06,
      "loss": 0.64444566,
      "num_input_tokens_seen": 89870080,
      "step": 4167,
      "time_per_iteration": 3.163869619369507
    },
    {
      "auxiliary_loss_clip": 0.0107453,
      "auxiliary_loss_mlp": 0.01020824,
      "balance_loss_clip": 1.04371798,
      "balance_loss_mlp": 1.01347506,
      "epoch": 0.5011723681837311,
      "flos": 18625500864000.0,
      "grad_norm": 16.11228876097598,
      "language_loss": 0.68567574,
      "learning_rate": 2.089551788968616e-06,
      "loss": 0.70662934,
      "num_input_tokens_seen": 89888045,
      "step": 4168,
      "time_per_iteration": 2.7228903770446777
    },
    {
      "auxiliary_loss_clip": 0.01067506,
      "auxiliary_loss_mlp": 0.01003896,
      "balance_loss_clip": 1.01480865,
      "balance_loss_mlp": 1.0026921,
      "epoch": 0.5012926110743702,
      "flos": 55883146608000.0,
      "grad_norm": 0.8329862534729657,
      "language_loss": 0.60787725,
      "learning_rate": 2.08877359181292e-06,
      "loss": 0.6285913,
      "num_input_tokens_seen": 89944610,
      "step": 4169,
      "time_per_iteration": 3.203059196472168
    },
    {
      "auxiliary_loss_clip": 0.01137493,
      "auxiliary_loss_mlp": 0.01026594,
      "balance_loss_clip": 1.04540968,
      "balance_loss_mlp": 1.01876783,
      "epoch": 0.5014128539650093,
      "flos": 24238266117120.0,
      "grad_norm": 2.0616467086050387,
      "language_loss": 0.85589921,
      "learning_rate": 2.0879953811903396e-06,
      "loss": 0.87754011,
      "num_input_tokens_seen": 89959495,
      "step": 4170,
      "time_per_iteration": 2.5822701454162598
    },
    {
      "auxiliary_loss_clip": 0.01164882,
      "auxiliary_loss_mlp": 0.01027253,
      "balance_loss_clip": 1.05312705,
      "balance_loss_mlp": 1.01868761,
      "epoch": 0.5015330968556484,
      "flos": 27527468382720.0,
      "grad_norm": 1.7806684351801745,
      "language_loss": 0.78732824,
      "learning_rate": 2.08721715721893e-06,
      "loss": 0.80924964,
      "num_input_tokens_seen": 89978820,
      "step": 4171,
      "time_per_iteration": 3.353858232498169
    },
    {
      "auxiliary_loss_clip": 0.01162609,
      "auxiliary_loss_mlp": 0.01028919,
      "balance_loss_clip": 1.05244851,
      "balance_loss_mlp": 1.02115297,
      "epoch": 0.5016533397462875,
      "flos": 23800802376960.0,
      "grad_norm": 1.9706806156744732,
      "language_loss": 0.76989758,
      "learning_rate": 2.0864389200167477e-06,
      "loss": 0.7918129,
      "num_input_tokens_seen": 89997075,
      "step": 4172,
      "time_per_iteration": 2.5338642597198486
    },
    {
      "auxiliary_loss_clip": 0.01167983,
      "auxiliary_loss_mlp": 0.00760892,
      "balance_loss_clip": 1.05456293,
      "balance_loss_mlp": 1.00017202,
      "epoch": 0.5017735826369266,
      "flos": 25295009264640.0,
      "grad_norm": 1.8617837910706252,
      "language_loss": 0.79108065,
      "learning_rate": 2.0856606697018504e-06,
      "loss": 0.81036937,
      "num_input_tokens_seen": 90015085,
      "step": 4173,
      "time_per_iteration": 2.5470848083496094
    },
    {
      "auxiliary_loss_clip": 0.01145988,
      "auxiliary_loss_mlp": 0.01034758,
      "balance_loss_clip": 1.04871273,
      "balance_loss_mlp": 1.02656507,
      "epoch": 0.5018938255275657,
      "flos": 16873203778560.0,
      "grad_norm": 2.411441451079812,
      "language_loss": 0.73592293,
      "learning_rate": 2.084882406392297e-06,
      "loss": 0.75773036,
      "num_input_tokens_seen": 90033045,
      "step": 4174,
      "time_per_iteration": 4.00055456161499
    },
    {
      "auxiliary_loss_clip": 0.01163824,
      "auxiliary_loss_mlp": 0.01028393,
      "balance_loss_clip": 1.05199301,
      "balance_loss_mlp": 1.02073693,
      "epoch": 0.5020140684182047,
      "flos": 25515429073920.0,
      "grad_norm": 2.9864552961503423,
      "language_loss": 0.70971072,
      "learning_rate": 2.0841041302061496e-06,
      "loss": 0.73163283,
      "num_input_tokens_seen": 90052505,
      "step": 4175,
      "time_per_iteration": 3.3809430599212646
    },
    {
      "auxiliary_loss_clip": 0.0113667,
      "auxiliary_loss_mlp": 0.01026583,
      "balance_loss_clip": 1.04608202,
      "balance_loss_mlp": 1.01864111,
      "epoch": 0.5021343113088439,
      "flos": 23659278791040.0,
      "grad_norm": 1.73420903101188,
      "language_loss": 0.75613773,
      "learning_rate": 2.083325841261473e-06,
      "loss": 0.77777028,
      "num_input_tokens_seen": 90071565,
      "step": 4176,
      "time_per_iteration": 2.5321435928344727
    },
    {
      "auxiliary_loss_clip": 0.01142154,
      "auxiliary_loss_mlp": 0.01032576,
      "balance_loss_clip": 1.04664278,
      "balance_loss_mlp": 1.02505958,
      "epoch": 0.502254554199483,
      "flos": 24534673148160.0,
      "grad_norm": 2.0346113726414394,
      "language_loss": 0.66083282,
      "learning_rate": 2.0825475396763322e-06,
      "loss": 0.68258011,
      "num_input_tokens_seen": 90092215,
      "step": 4177,
      "time_per_iteration": 2.5607407093048096
    },
    {
      "auxiliary_loss_clip": 0.01070165,
      "auxiliary_loss_mlp": 0.01033874,
      "balance_loss_clip": 1.04065919,
      "balance_loss_mlp": 1.02608347,
      "epoch": 0.502374797090122,
      "flos": 34240285607040.0,
      "grad_norm": 1.3726682677620918,
      "language_loss": 0.65577304,
      "learning_rate": 2.081769225568796e-06,
      "loss": 0.67681342,
      "num_input_tokens_seen": 90114665,
      "step": 4178,
      "time_per_iteration": 2.8261711597442627
    },
    {
      "auxiliary_loss_clip": 0.01166424,
      "auxiliary_loss_mlp": 0.01028864,
      "balance_loss_clip": 1.05168939,
      "balance_loss_mlp": 1.0213958,
      "epoch": 0.5024950399807612,
      "flos": 26031106679040.0,
      "grad_norm": 1.411452304758719,
      "language_loss": 0.76094055,
      "learning_rate": 2.0809908990569327e-06,
      "loss": 0.78289342,
      "num_input_tokens_seen": 90136445,
      "step": 4179,
      "time_per_iteration": 2.540700674057007
    },
    {
      "auxiliary_loss_clip": 0.01147171,
      "auxiliary_loss_mlp": 0.01027369,
      "balance_loss_clip": 1.04948068,
      "balance_loss_mlp": 1.01996565,
      "epoch": 0.5026152828714002,
      "flos": 21252438120960.0,
      "grad_norm": 1.6776774641586172,
      "language_loss": 0.79082036,
      "learning_rate": 2.0802125602588146e-06,
      "loss": 0.81256568,
      "num_input_tokens_seen": 90155710,
      "step": 4180,
      "time_per_iteration": 2.5347554683685303
    },
    {
      "auxiliary_loss_clip": 0.01177463,
      "auxiliary_loss_mlp": 0.01031387,
      "balance_loss_clip": 1.05333328,
      "balance_loss_mlp": 1.02363241,
      "epoch": 0.5027355257620393,
      "flos": 30956111245440.0,
      "grad_norm": 1.8901339210564587,
      "language_loss": 0.66836363,
      "learning_rate": 2.0794342092925146e-06,
      "loss": 0.6904521,
      "num_input_tokens_seen": 90176845,
      "step": 4181,
      "time_per_iteration": 2.5426881313323975
    },
    {
      "auxiliary_loss_clip": 0.01167483,
      "auxiliary_loss_mlp": 0.0103927,
      "balance_loss_clip": 1.05474925,
      "balance_loss_mlp": 1.03202558,
      "epoch": 0.5028557686526784,
      "flos": 24791147233920.0,
      "grad_norm": 1.932340151779029,
      "language_loss": 0.68094885,
      "learning_rate": 2.078655846276108e-06,
      "loss": 0.7030164,
      "num_input_tokens_seen": 90197175,
      "step": 4182,
      "time_per_iteration": 2.5261037349700928
    },
    {
      "auxiliary_loss_clip": 0.0114584,
      "auxiliary_loss_mlp": 0.01031238,
      "balance_loss_clip": 1.04974163,
      "balance_loss_mlp": 1.02323878,
      "epoch": 0.5029760115433175,
      "flos": 22966992990720.0,
      "grad_norm": 2.0505844719069826,
      "language_loss": 0.68900019,
      "learning_rate": 2.0778774713276727e-06,
      "loss": 0.71077096,
      "num_input_tokens_seen": 90216650,
      "step": 4183,
      "time_per_iteration": 2.5530202388763428
    },
    {
      "auxiliary_loss_clip": 0.01159387,
      "auxiliary_loss_mlp": 0.01032715,
      "balance_loss_clip": 1.04730916,
      "balance_loss_mlp": 1.02454281,
      "epoch": 0.5030962544339566,
      "flos": 15305164485120.0,
      "grad_norm": 1.9509235515434742,
      "language_loss": 0.67494833,
      "learning_rate": 2.077099084565287e-06,
      "loss": 0.69686937,
      "num_input_tokens_seen": 90234055,
      "step": 4184,
      "time_per_iteration": 2.4468019008636475
    },
    {
      "auxiliary_loss_clip": 0.01141259,
      "auxiliary_loss_mlp": 0.0102971,
      "balance_loss_clip": 1.04552436,
      "balance_loss_mlp": 1.02204478,
      "epoch": 0.5032164973245957,
      "flos": 24494847943680.0,
      "grad_norm": 2.1553172988686926,
      "language_loss": 0.65419692,
      "learning_rate": 2.0763206861070313e-06,
      "loss": 0.6759066,
      "num_input_tokens_seen": 90253115,
      "step": 4185,
      "time_per_iteration": 2.5473978519439697
    },
    {
      "auxiliary_loss_clip": 0.01178872,
      "auxiliary_loss_mlp": 0.0102991,
      "balance_loss_clip": 1.05412555,
      "balance_loss_mlp": 1.02271605,
      "epoch": 0.5033367402152348,
      "flos": 16213452721920.0,
      "grad_norm": 1.9242689369314694,
      "language_loss": 0.75485933,
      "learning_rate": 2.0755422760709876e-06,
      "loss": 0.77694714,
      "num_input_tokens_seen": 90270515,
      "step": 4186,
      "time_per_iteration": 2.420877456665039
    },
    {
      "auxiliary_loss_clip": 0.01113528,
      "auxiliary_loss_mlp": 0.01031559,
      "balance_loss_clip": 1.04585934,
      "balance_loss_mlp": 1.023417,
      "epoch": 0.5034569831058738,
      "flos": 21391375927680.0,
      "grad_norm": 1.9830960000054427,
      "language_loss": 0.76889509,
      "learning_rate": 2.0747638545752417e-06,
      "loss": 0.79034597,
      "num_input_tokens_seen": 90289075,
      "step": 4187,
      "time_per_iteration": 2.578738212585449
    },
    {
      "auxiliary_loss_clip": 0.01146718,
      "auxiliary_loss_mlp": 0.01029116,
      "balance_loss_clip": 1.0529995,
      "balance_loss_mlp": 1.02124166,
      "epoch": 0.503577225996513,
      "flos": 20558751690240.0,
      "grad_norm": 1.9440099360491072,
      "language_loss": 0.83215129,
      "learning_rate": 2.073985421737878e-06,
      "loss": 0.85390961,
      "num_input_tokens_seen": 90306385,
      "step": 4188,
      "time_per_iteration": 2.5155348777770996
    },
    {
      "auxiliary_loss_clip": 0.01167038,
      "auxiliary_loss_mlp": 0.01024971,
      "balance_loss_clip": 1.05470324,
      "balance_loss_mlp": 1.01726389,
      "epoch": 0.5036974688871521,
      "flos": 27229157930880.0,
      "grad_norm": 2.4746969504253884,
      "language_loss": 0.74212736,
      "learning_rate": 2.0732069776769844e-06,
      "loss": 0.76404744,
      "num_input_tokens_seen": 90323795,
      "step": 4189,
      "time_per_iteration": 2.5896644592285156
    },
    {
      "auxiliary_loss_clip": 0.0117994,
      "auxiliary_loss_mlp": 0.01030963,
      "balance_loss_clip": 1.055933,
      "balance_loss_mlp": 1.02317238,
      "epoch": 0.5038177117777911,
      "flos": 20412164286720.0,
      "grad_norm": 2.985659233832435,
      "language_loss": 0.73377407,
      "learning_rate": 2.072428522510651e-06,
      "loss": 0.7558831,
      "num_input_tokens_seen": 90340360,
      "step": 4190,
      "time_per_iteration": 2.4469399452209473
    },
    {
      "auxiliary_loss_clip": 0.01127845,
      "auxiliary_loss_mlp": 0.01032591,
      "balance_loss_clip": 1.04685116,
      "balance_loss_mlp": 1.02476239,
      "epoch": 0.5039379546684303,
      "flos": 21907987286400.0,
      "grad_norm": 2.1850802944588024,
      "language_loss": 0.76329792,
      "learning_rate": 2.071650056356968e-06,
      "loss": 0.78490233,
      "num_input_tokens_seen": 90357900,
      "step": 4191,
      "time_per_iteration": 2.55739688873291
    },
    {
      "auxiliary_loss_clip": 0.01178018,
      "auxiliary_loss_mlp": 0.01032061,
      "balance_loss_clip": 1.05429316,
      "balance_loss_mlp": 1.02379405,
      "epoch": 0.5040581975590693,
      "flos": 20010718909440.0,
      "grad_norm": 1.9055121657039031,
      "language_loss": 0.79658091,
      "learning_rate": 2.070871579334028e-06,
      "loss": 0.81868172,
      "num_input_tokens_seen": 90377010,
      "step": 4192,
      "time_per_iteration": 3.2054247856140137
    },
    {
      "auxiliary_loss_clip": 0.01177887,
      "auxiliary_loss_mlp": 0.01026652,
      "balance_loss_clip": 1.05453849,
      "balance_loss_mlp": 1.01885605,
      "epoch": 0.5041784404497084,
      "flos": 20959837931520.0,
      "grad_norm": 1.8798911773675568,
      "language_loss": 0.7166698,
      "learning_rate": 2.0700930915599264e-06,
      "loss": 0.73871517,
      "num_input_tokens_seen": 90396740,
      "step": 4193,
      "time_per_iteration": 2.490858793258667
    },
    {
      "auxiliary_loss_clip": 0.01177327,
      "auxiliary_loss_mlp": 0.01030954,
      "balance_loss_clip": 1.05407524,
      "balance_loss_mlp": 1.02321136,
      "epoch": 0.5042986833403476,
      "flos": 12495082757760.0,
      "grad_norm": 1.980026942077062,
      "language_loss": 0.78054607,
      "learning_rate": 2.0693145931527583e-06,
      "loss": 0.80262893,
      "num_input_tokens_seen": 90413220,
      "step": 4194,
      "time_per_iteration": 2.4330241680145264
    },
    {
      "auxiliary_loss_clip": 0.01143162,
      "auxiliary_loss_mlp": 0.01026054,
      "balance_loss_clip": 1.04919732,
      "balance_loss_mlp": 1.01858592,
      "epoch": 0.5044189262309866,
      "flos": 29202305788800.0,
      "grad_norm": 1.6372760965022268,
      "language_loss": 0.78126812,
      "learning_rate": 2.068536084230622e-06,
      "loss": 0.80296028,
      "num_input_tokens_seen": 90435085,
      "step": 4195,
      "time_per_iteration": 2.5685012340545654
    },
    {
      "auxiliary_loss_clip": 0.0116416,
      "auxiliary_loss_mlp": 0.01030857,
      "balance_loss_clip": 1.05200505,
      "balance_loss_mlp": 1.02332282,
      "epoch": 0.5045391691216257,
      "flos": 23873198238720.0,
      "grad_norm": 1.91953398232203,
      "language_loss": 0.88615626,
      "learning_rate": 2.067757564911616e-06,
      "loss": 0.90810645,
      "num_input_tokens_seen": 90453660,
      "step": 4196,
      "time_per_iteration": 2.51774001121521
    },
    {
      "auxiliary_loss_clip": 0.01155845,
      "auxiliary_loss_mlp": 0.00761585,
      "balance_loss_clip": 1.05011117,
      "balance_loss_mlp": 1.00019908,
      "epoch": 0.5046594120122648,
      "flos": 24644990793600.0,
      "grad_norm": 1.8569619877672905,
      "language_loss": 0.92448008,
      "learning_rate": 2.0669790353138407e-06,
      "loss": 0.9436543,
      "num_input_tokens_seen": 90472625,
      "step": 4197,
      "time_per_iteration": 3.5277116298675537
    },
    {
      "auxiliary_loss_clip": 0.01132486,
      "auxiliary_loss_mlp": 0.00761489,
      "balance_loss_clip": 1.05234206,
      "balance_loss_mlp": 1.00017345,
      "epoch": 0.5047796549029039,
      "flos": 23362835846400.0,
      "grad_norm": 2.2883263087813543,
      "language_loss": 0.72756457,
      "learning_rate": 2.0662004955553995e-06,
      "loss": 0.74650431,
      "num_input_tokens_seen": 90492325,
      "step": 4198,
      "time_per_iteration": 2.577946186065674
    },
    {
      "auxiliary_loss_clip": 0.01144672,
      "auxiliary_loss_mlp": 0.01027847,
      "balance_loss_clip": 1.04984498,
      "balance_loss_mlp": 1.02067614,
      "epoch": 0.5048998977935429,
      "flos": 17304095329920.0,
      "grad_norm": 2.242887396870967,
      "language_loss": 0.76986921,
      "learning_rate": 2.065421945754395e-06,
      "loss": 0.79159439,
      "num_input_tokens_seen": 90510055,
      "step": 4199,
      "time_per_iteration": 3.5041067600250244
    },
    {
      "auxiliary_loss_clip": 0.01124371,
      "auxiliary_loss_mlp": 0.01034844,
      "balance_loss_clip": 1.05040789,
      "balance_loss_mlp": 1.02784395,
      "epoch": 0.505020140684182,
      "flos": 34856979235200.0,
      "grad_norm": 1.581844721386092,
      "language_loss": 0.78115481,
      "learning_rate": 2.0646433860289344e-06,
      "loss": 0.80274695,
      "num_input_tokens_seen": 90528980,
      "step": 4200,
      "time_per_iteration": 2.7079176902770996
    },
    {
      "auxiliary_loss_clip": 0.01167235,
      "auxiliary_loss_mlp": 0.00761784,
      "balance_loss_clip": 1.05257678,
      "balance_loss_mlp": 1.00021267,
      "epoch": 0.5051403835748212,
      "flos": 24863974058880.0,
      "grad_norm": 1.898663818054111,
      "language_loss": 0.82666916,
      "learning_rate": 2.0638648164971233e-06,
      "loss": 0.84595931,
      "num_input_tokens_seen": 90547445,
      "step": 4201,
      "time_per_iteration": 3.3593008518218994
    },
    {
      "auxiliary_loss_clip": 0.01146937,
      "auxiliary_loss_mlp": 0.0103006,
      "balance_loss_clip": 1.05138469,
      "balance_loss_mlp": 1.02278876,
      "epoch": 0.5052606264654602,
      "flos": 20959694277120.0,
      "grad_norm": 1.7610956015174006,
      "language_loss": 0.88204002,
      "learning_rate": 2.06308623727707e-06,
      "loss": 0.90381002,
      "num_input_tokens_seen": 90567545,
      "step": 4202,
      "time_per_iteration": 2.5143041610717773
    },
    {
      "auxiliary_loss_clip": 0.01158006,
      "auxiliary_loss_mlp": 0.01030068,
      "balance_loss_clip": 1.05105853,
      "balance_loss_mlp": 1.0223012,
      "epoch": 0.5053808693560993,
      "flos": 19642382893440.0,
      "grad_norm": 3.414404056917229,
      "language_loss": 0.76468009,
      "learning_rate": 2.0623076484868846e-06,
      "loss": 0.78656077,
      "num_input_tokens_seen": 90585000,
      "step": 4203,
      "time_per_iteration": 2.4805896282196045
    },
    {
      "auxiliary_loss_clip": 0.0104732,
      "auxiliary_loss_mlp": 0.01001328,
      "balance_loss_clip": 1.01873136,
      "balance_loss_mlp": 1.00019574,
      "epoch": 0.5055011122467384,
      "flos": 67504915019520.0,
      "grad_norm": 0.8355337152611019,
      "language_loss": 0.60724926,
      "learning_rate": 2.061529050244679e-06,
      "loss": 0.62773573,
      "num_input_tokens_seen": 90644745,
      "step": 4204,
      "time_per_iteration": 3.085674285888672
    },
    {
      "auxiliary_loss_clip": 0.01139251,
      "auxiliary_loss_mlp": 0.01028543,
      "balance_loss_clip": 1.04781306,
      "balance_loss_mlp": 1.02096403,
      "epoch": 0.5056213551373775,
      "flos": 16872952383360.0,
      "grad_norm": 1.8714808254221986,
      "language_loss": 0.74201095,
      "learning_rate": 2.060750442668565e-06,
      "loss": 0.76368892,
      "num_input_tokens_seen": 90662500,
      "step": 4205,
      "time_per_iteration": 2.56261944770813
    },
    {
      "auxiliary_loss_clip": 0.01166494,
      "auxiliary_loss_mlp": 0.01031116,
      "balance_loss_clip": 1.05485272,
      "balance_loss_mlp": 1.02362943,
      "epoch": 0.5057415980280165,
      "flos": 15334179696000.0,
      "grad_norm": 2.0871415031534095,
      "language_loss": 0.63285404,
      "learning_rate": 2.059971825876657e-06,
      "loss": 0.65483022,
      "num_input_tokens_seen": 90677010,
      "step": 4206,
      "time_per_iteration": 2.4507975578308105
    },
    {
      "auxiliary_loss_clip": 0.01165913,
      "auxiliary_loss_mlp": 0.01024603,
      "balance_loss_clip": 1.05157995,
      "balance_loss_mlp": 1.01730704,
      "epoch": 0.5058618409186557,
      "flos": 19025976574080.0,
      "grad_norm": 1.8202379029697286,
      "language_loss": 0.76183939,
      "learning_rate": 2.0591931999870713e-06,
      "loss": 0.78374457,
      "num_input_tokens_seen": 90695935,
      "step": 4207,
      "time_per_iteration": 2.4887349605560303
    },
    {
      "auxiliary_loss_clip": 0.01054631,
      "auxiliary_loss_mlp": 0.01001131,
      "balance_loss_clip": 1.01687884,
      "balance_loss_mlp": 0.99997443,
      "epoch": 0.5059820838092948,
      "flos": 63453114080640.0,
      "grad_norm": 0.8154018563046301,
      "language_loss": 0.57601523,
      "learning_rate": 2.0584145651179234e-06,
      "loss": 0.59657288,
      "num_input_tokens_seen": 90751645,
      "step": 4208,
      "time_per_iteration": 3.111002206802368
    },
    {
      "auxiliary_loss_clip": 0.01150409,
      "auxiliary_loss_mlp": 0.00761132,
      "balance_loss_clip": 1.05227113,
      "balance_loss_mlp": 1.000175,
      "epoch": 0.5061023266999338,
      "flos": 15441803821440.0,
      "grad_norm": 2.293162442815295,
      "language_loss": 0.79760212,
      "learning_rate": 2.0576359213873327e-06,
      "loss": 0.81671751,
      "num_input_tokens_seen": 90766795,
      "step": 4209,
      "time_per_iteration": 2.491255283355713
    },
    {
      "auxiliary_loss_clip": 0.01155021,
      "auxiliary_loss_mlp": 0.01030972,
      "balance_loss_clip": 1.0485692,
      "balance_loss_mlp": 1.02280045,
      "epoch": 0.506222569590573,
      "flos": 22451063990400.0,
      "grad_norm": 3.1146726348033598,
      "language_loss": 0.69539279,
      "learning_rate": 2.056857268913419e-06,
      "loss": 0.71725273,
      "num_input_tokens_seen": 90786845,
      "step": 4210,
      "time_per_iteration": 2.5574803352355957
    },
    {
      "auxiliary_loss_clip": 0.01162657,
      "auxiliary_loss_mlp": 0.01032823,
      "balance_loss_clip": 1.05253291,
      "balance_loss_mlp": 1.02526522,
      "epoch": 0.506342812481212,
      "flos": 17558665994880.0,
      "grad_norm": 2.4971189791046338,
      "language_loss": 0.84000343,
      "learning_rate": 2.056078607814303e-06,
      "loss": 0.86195827,
      "num_input_tokens_seen": 90802630,
      "step": 4211,
      "time_per_iteration": 2.4595162868499756
    },
    {
      "auxiliary_loss_clip": 0.01161974,
      "auxiliary_loss_mlp": 0.01024487,
      "balance_loss_clip": 1.05292678,
      "balance_loss_mlp": 1.01710176,
      "epoch": 0.5064630553718511,
      "flos": 23402050519680.0,
      "grad_norm": 1.6345426627252317,
      "language_loss": 0.78312564,
      "learning_rate": 2.055299938208106e-06,
      "loss": 0.80499023,
      "num_input_tokens_seen": 90823620,
      "step": 4212,
      "time_per_iteration": 2.5733213424682617
    },
    {
      "auxiliary_loss_clip": 0.01168702,
      "auxiliary_loss_mlp": 0.01036878,
      "balance_loss_clip": 1.05311191,
      "balance_loss_mlp": 1.0290103,
      "epoch": 0.5065832982624903,
      "flos": 23987035416960.0,
      "grad_norm": 1.6112922213156586,
      "language_loss": 0.86340767,
      "learning_rate": 2.0545212602129526e-06,
      "loss": 0.88546348,
      "num_input_tokens_seen": 90843475,
      "step": 4213,
      "time_per_iteration": 2.5165650844573975
    },
    {
      "auxiliary_loss_clip": 0.01141536,
      "auxiliary_loss_mlp": 0.01032005,
      "balance_loss_clip": 1.04848361,
      "balance_loss_mlp": 1.02408361,
      "epoch": 0.5067035411531293,
      "flos": 21503058289920.0,
      "grad_norm": 1.9953579905253076,
      "language_loss": 0.65967679,
      "learning_rate": 2.0537425739469673e-06,
      "loss": 0.68141222,
      "num_input_tokens_seen": 90862410,
      "step": 4214,
      "time_per_iteration": 2.5255584716796875
    },
    {
      "auxiliary_loss_clip": 0.01062264,
      "auxiliary_loss_mlp": 0.01003015,
      "balance_loss_clip": 1.01865435,
      "balance_loss_mlp": 1.0017271,
      "epoch": 0.5068237840437684,
      "flos": 65934397687680.0,
      "grad_norm": 0.8400665669410669,
      "language_loss": 0.59557259,
      "learning_rate": 2.052963879528276e-06,
      "loss": 0.61622536,
      "num_input_tokens_seen": 90922280,
      "step": 4215,
      "time_per_iteration": 3.067805767059326
    },
    {
      "auxiliary_loss_clip": 0.01164331,
      "auxiliary_loss_mlp": 0.01028943,
      "balance_loss_clip": 1.05378652,
      "balance_loss_mlp": 1.0215131,
      "epoch": 0.5069440269344075,
      "flos": 27264206626560.0,
      "grad_norm": 2.0978684890715185,
      "language_loss": 0.76611459,
      "learning_rate": 2.052185177075007e-06,
      "loss": 0.78804737,
      "num_input_tokens_seen": 90941850,
      "step": 4216,
      "time_per_iteration": 2.56473708152771
    },
    {
      "auxiliary_loss_clip": 0.01165612,
      "auxiliary_loss_mlp": 0.01029182,
      "balance_loss_clip": 1.05159724,
      "balance_loss_mlp": 1.02119756,
      "epoch": 0.5070642698250466,
      "flos": 23366319465600.0,
      "grad_norm": 2.0021178232316807,
      "language_loss": 0.82545358,
      "learning_rate": 2.051406466705288e-06,
      "loss": 0.8474015,
      "num_input_tokens_seen": 90961390,
      "step": 4217,
      "time_per_iteration": 2.530457019805908
    },
    {
      "auxiliary_loss_clip": 0.01176707,
      "auxiliary_loss_mlp": 0.01029921,
      "balance_loss_clip": 1.05155206,
      "balance_loss_mlp": 1.02279198,
      "epoch": 0.5071845127156857,
      "flos": 20340127560960.0,
      "grad_norm": 2.86171989589709,
      "language_loss": 0.81645536,
      "learning_rate": 2.0506277485372486e-06,
      "loss": 0.8385216,
      "num_input_tokens_seen": 90980215,
      "step": 4218,
      "time_per_iteration": 3.2556400299072266
    },
    {
      "auxiliary_loss_clip": 0.0115695,
      "auxiliary_loss_mlp": 0.01026047,
      "balance_loss_clip": 1.05057788,
      "balance_loss_mlp": 1.01767874,
      "epoch": 0.5073047556063248,
      "flos": 12092955022080.0,
      "grad_norm": 1.7778781648639799,
      "language_loss": 0.66863912,
      "learning_rate": 2.04984902268902e-06,
      "loss": 0.69046909,
      "num_input_tokens_seen": 90997415,
      "step": 4219,
      "time_per_iteration": 2.4564056396484375
    },
    {
      "auxiliary_loss_clip": 0.01172983,
      "auxiliary_loss_mlp": 0.01031169,
      "balance_loss_clip": 1.05459666,
      "balance_loss_mlp": 1.0229528,
      "epoch": 0.5074249984969639,
      "flos": 19682854542720.0,
      "grad_norm": 2.506483656681027,
      "language_loss": 0.75785333,
      "learning_rate": 2.0490702892787345e-06,
      "loss": 0.77989483,
      "num_input_tokens_seen": 91016475,
      "step": 4220,
      "time_per_iteration": 2.5158536434173584
    },
    {
      "auxiliary_loss_clip": 0.01153525,
      "auxiliary_loss_mlp": 0.01029406,
      "balance_loss_clip": 1.04746556,
      "balance_loss_mlp": 1.02189565,
      "epoch": 0.5075452413876029,
      "flos": 28765703975040.0,
      "grad_norm": 1.7437900176018322,
      "language_loss": 0.62417096,
      "learning_rate": 2.0482915484245246e-06,
      "loss": 0.64600027,
      "num_input_tokens_seen": 91038095,
      "step": 4221,
      "time_per_iteration": 2.5436689853668213
    },
    {
      "auxiliary_loss_clip": 0.01113281,
      "auxiliary_loss_mlp": 0.01035679,
      "balance_loss_clip": 1.04616141,
      "balance_loss_mlp": 1.02775705,
      "epoch": 0.5076654842782421,
      "flos": 20339445202560.0,
      "grad_norm": 2.3523043350310457,
      "language_loss": 0.84171426,
      "learning_rate": 2.047512800244526e-06,
      "loss": 0.86320382,
      "num_input_tokens_seen": 91053360,
      "step": 4222,
      "time_per_iteration": 2.561302900314331
    },
    {
      "auxiliary_loss_clip": 0.01164682,
      "auxiliary_loss_mlp": 0.01028807,
      "balance_loss_clip": 1.05386817,
      "balance_loss_mlp": 1.02110004,
      "epoch": 0.5077857271688812,
      "flos": 26359653404160.0,
      "grad_norm": 1.8421534975061489,
      "language_loss": 0.78822172,
      "learning_rate": 2.046734044856873e-06,
      "loss": 0.81015658,
      "num_input_tokens_seen": 91072770,
      "step": 4223,
      "time_per_iteration": 3.2972469329833984
    },
    {
      "auxiliary_loss_clip": 0.01167099,
      "auxiliary_loss_mlp": 0.01026357,
      "balance_loss_clip": 1.05560184,
      "balance_loss_mlp": 1.01896,
      "epoch": 0.5079059700595202,
      "flos": 21798962530560.0,
      "grad_norm": 2.0372088377184627,
      "language_loss": 0.81090522,
      "learning_rate": 2.045955282379702e-06,
      "loss": 0.83283979,
      "num_input_tokens_seen": 91091430,
      "step": 4224,
      "time_per_iteration": 2.5045578479766846
    },
    {
      "auxiliary_loss_clip": 0.01161316,
      "auxiliary_loss_mlp": 0.01034365,
      "balance_loss_clip": 1.04988265,
      "balance_loss_mlp": 1.02654791,
      "epoch": 0.5080262129501594,
      "flos": 13187943175680.0,
      "grad_norm": 2.473930552007521,
      "language_loss": 0.7569347,
      "learning_rate": 2.045176512931152e-06,
      "loss": 0.7788915,
      "num_input_tokens_seen": 91106060,
      "step": 4225,
      "time_per_iteration": 3.4761996269226074
    },
    {
      "auxiliary_loss_clip": 0.011372,
      "auxiliary_loss_mlp": 0.01024587,
      "balance_loss_clip": 1.04983008,
      "balance_loss_mlp": 1.01657045,
      "epoch": 0.5081464558407984,
      "flos": 25301473712640.0,
      "grad_norm": 1.893992660779313,
      "language_loss": 0.75820231,
      "learning_rate": 2.0443977366293604e-06,
      "loss": 0.77982014,
      "num_input_tokens_seen": 91124100,
      "step": 4226,
      "time_per_iteration": 3.448227643966675
    },
    {
      "auxiliary_loss_clip": 0.01107081,
      "auxiliary_loss_mlp": 0.01025583,
      "balance_loss_clip": 1.04487848,
      "balance_loss_mlp": 1.01736355,
      "epoch": 0.5082666987314375,
      "flos": 30951226995840.0,
      "grad_norm": 1.584140773506295,
      "language_loss": 0.77131808,
      "learning_rate": 2.043618953592468e-06,
      "loss": 0.7926448,
      "num_input_tokens_seen": 91146555,
      "step": 4227,
      "time_per_iteration": 2.7680504322052
    },
    {
      "auxiliary_loss_clip": 0.01151506,
      "auxiliary_loss_mlp": 0.01038454,
      "balance_loss_clip": 1.0524646,
      "balance_loss_mlp": 1.03036571,
      "epoch": 0.5083869416220766,
      "flos": 19682495406720.0,
      "grad_norm": 2.235483727179507,
      "language_loss": 0.81278104,
      "learning_rate": 2.0428401639386144e-06,
      "loss": 0.83468068,
      "num_input_tokens_seen": 91167120,
      "step": 4228,
      "time_per_iteration": 2.8543386459350586
    },
    {
      "auxiliary_loss_clip": 0.01043192,
      "auxiliary_loss_mlp": 0.00999851,
      "balance_loss_clip": 1.01409721,
      "balance_loss_mlp": 0.9987008,
      "epoch": 0.5085071845127157,
      "flos": 71817535589760.0,
      "grad_norm": 0.8260345526442083,
      "language_loss": 0.58249795,
      "learning_rate": 2.042061367785943e-06,
      "loss": 0.6029284,
      "num_input_tokens_seen": 91220260,
      "step": 4229,
      "time_per_iteration": 3.0936594009399414
    },
    {
      "auxiliary_loss_clip": 0.01135906,
      "auxiliary_loss_mlp": 0.01031493,
      "balance_loss_clip": 1.04609931,
      "balance_loss_mlp": 1.02375674,
      "epoch": 0.5086274274033548,
      "flos": 35951608252800.0,
      "grad_norm": 2.0048920953621163,
      "language_loss": 0.75004053,
      "learning_rate": 2.041282565252594e-06,
      "loss": 0.77171457,
      "num_input_tokens_seen": 91240425,
      "step": 4230,
      "time_per_iteration": 2.747380256652832
    },
    {
      "auxiliary_loss_clip": 0.01130465,
      "auxiliary_loss_mlp": 0.01027626,
      "balance_loss_clip": 1.04377162,
      "balance_loss_mlp": 1.02036047,
      "epoch": 0.5087476702939938,
      "flos": 23513732881920.0,
      "grad_norm": 1.634317689359367,
      "language_loss": 0.77097267,
      "learning_rate": 2.040503756456714e-06,
      "loss": 0.79255354,
      "num_input_tokens_seen": 91259635,
      "step": 4231,
      "time_per_iteration": 2.571500301361084
    },
    {
      "auxiliary_loss_clip": 0.01154676,
      "auxiliary_loss_mlp": 0.01035077,
      "balance_loss_clip": 1.04760027,
      "balance_loss_mlp": 1.02784085,
      "epoch": 0.508867913184633,
      "flos": 15122091841920.0,
      "grad_norm": 1.8055752356351231,
      "language_loss": 0.78736484,
      "learning_rate": 2.0397249415164456e-06,
      "loss": 0.80926234,
      "num_input_tokens_seen": 91276990,
      "step": 4232,
      "time_per_iteration": 2.471558094024658
    },
    {
      "auxiliary_loss_clip": 0.01137794,
      "auxiliary_loss_mlp": 0.01034042,
      "balance_loss_clip": 1.04456997,
      "balance_loss_mlp": 1.0260371,
      "epoch": 0.508988156075272,
      "flos": 25885309374720.0,
      "grad_norm": 1.894749687970946,
      "language_loss": 0.80238324,
      "learning_rate": 2.0389461205499354e-06,
      "loss": 0.82410163,
      "num_input_tokens_seen": 91296125,
      "step": 4233,
      "time_per_iteration": 2.559316635131836
    },
    {
      "auxiliary_loss_clip": 0.01137836,
      "auxiliary_loss_mlp": 0.01032798,
      "balance_loss_clip": 1.04759526,
      "balance_loss_mlp": 1.02542472,
      "epoch": 0.5091083989659111,
      "flos": 13844857057920.0,
      "grad_norm": 2.121975015988607,
      "language_loss": 0.72800338,
      "learning_rate": 2.03816729367533e-06,
      "loss": 0.74970978,
      "num_input_tokens_seen": 91314280,
      "step": 4234,
      "time_per_iteration": 2.535372495651245
    },
    {
      "auxiliary_loss_clip": 0.01149938,
      "auxiliary_loss_mlp": 0.01032886,
      "balance_loss_clip": 1.05074239,
      "balance_loss_mlp": 1.0247556,
      "epoch": 0.5092286418565503,
      "flos": 21104881050240.0,
      "grad_norm": 1.9161355351015,
      "language_loss": 0.71780252,
      "learning_rate": 2.0373884610107765e-06,
      "loss": 0.73963082,
      "num_input_tokens_seen": 91334595,
      "step": 4235,
      "time_per_iteration": 2.5335328578948975
    },
    {
      "auxiliary_loss_clip": 0.01166432,
      "auxiliary_loss_mlp": 0.01025482,
      "balance_loss_clip": 1.05104828,
      "balance_loss_mlp": 1.01783466,
      "epoch": 0.5093488847471893,
      "flos": 18621298972800.0,
      "grad_norm": 2.110168761226361,
      "language_loss": 0.69410467,
      "learning_rate": 2.0366096226744225e-06,
      "loss": 0.7160238,
      "num_input_tokens_seen": 91349790,
      "step": 4236,
      "time_per_iteration": 2.453207492828369
    },
    {
      "auxiliary_loss_clip": 0.01154961,
      "auxiliary_loss_mlp": 0.01032731,
      "balance_loss_clip": 1.05097771,
      "balance_loss_mlp": 1.02486944,
      "epoch": 0.5094691276378284,
      "flos": 23803783205760.0,
      "grad_norm": 1.6303412867079519,
      "language_loss": 0.76964748,
      "learning_rate": 2.035830778784418e-06,
      "loss": 0.79152441,
      "num_input_tokens_seen": 91370465,
      "step": 4237,
      "time_per_iteration": 2.525562047958374
    },
    {
      "auxiliary_loss_clip": 0.01152983,
      "auxiliary_loss_mlp": 0.01033823,
      "balance_loss_clip": 1.05602157,
      "balance_loss_mlp": 1.02584243,
      "epoch": 0.5095893705284675,
      "flos": 17420410546560.0,
      "grad_norm": 1.8525517279484773,
      "language_loss": 0.79613757,
      "learning_rate": 2.0350519294589134e-06,
      "loss": 0.81800562,
      "num_input_tokens_seen": 91388505,
      "step": 4238,
      "time_per_iteration": 2.5369467735290527
    },
    {
      "auxiliary_loss_clip": 0.0111399,
      "auxiliary_loss_mlp": 0.01030052,
      "balance_loss_clip": 1.04315567,
      "balance_loss_mlp": 1.02143908,
      "epoch": 0.5097096134191066,
      "flos": 25849362839040.0,
      "grad_norm": 1.7301522862318621,
      "language_loss": 0.82847822,
      "learning_rate": 2.0342730748160588e-06,
      "loss": 0.8499186,
      "num_input_tokens_seen": 91408970,
      "step": 4239,
      "time_per_iteration": 2.6345903873443604
    },
    {
      "auxiliary_loss_clip": 0.01147611,
      "auxiliary_loss_mlp": 0.01027437,
      "balance_loss_clip": 1.04890871,
      "balance_loss_mlp": 1.0197655,
      "epoch": 0.5098298563097456,
      "flos": 27745122844800.0,
      "grad_norm": 2.524601991108417,
      "language_loss": 0.70918834,
      "learning_rate": 2.033494214974006e-06,
      "loss": 0.73093879,
      "num_input_tokens_seen": 91430115,
      "step": 4240,
      "time_per_iteration": 2.5926387310028076
    },
    {
      "auxiliary_loss_clip": 0.01138618,
      "auxiliary_loss_mlp": 0.01032592,
      "balance_loss_clip": 1.04994321,
      "balance_loss_mlp": 1.02515292,
      "epoch": 0.5099500992003848,
      "flos": 21358913011200.0,
      "grad_norm": 1.64939661716861,
      "language_loss": 0.84049702,
      "learning_rate": 2.0327153500509067e-06,
      "loss": 0.86220908,
      "num_input_tokens_seen": 91449140,
      "step": 4241,
      "time_per_iteration": 2.5476481914520264
    },
    {
      "auxiliary_loss_clip": 0.01149327,
      "auxiliary_loss_mlp": 0.01026815,
      "balance_loss_clip": 1.05007505,
      "balance_loss_mlp": 1.01906633,
      "epoch": 0.5100703420910239,
      "flos": 19865999013120.0,
      "grad_norm": 1.9343243634454035,
      "language_loss": 0.84635299,
      "learning_rate": 2.031936480164916e-06,
      "loss": 0.86811441,
      "num_input_tokens_seen": 91466880,
      "step": 4242,
      "time_per_iteration": 2.5282375812530518
    },
    {
      "auxiliary_loss_clip": 0.01145956,
      "auxiliary_loss_mlp": 0.010362,
      "balance_loss_clip": 1.05205619,
      "balance_loss_mlp": 1.028916,
      "epoch": 0.5101905849816629,
      "flos": 24648797635200.0,
      "grad_norm": 19.148422893939586,
      "language_loss": 0.80279952,
      "learning_rate": 2.0311576054341857e-06,
      "loss": 0.82462108,
      "num_input_tokens_seen": 91487495,
      "step": 4243,
      "time_per_iteration": 2.5755820274353027
    },
    {
      "auxiliary_loss_clip": 0.01177899,
      "auxiliary_loss_mlp": 0.01023635,
      "balance_loss_clip": 1.05500984,
      "balance_loss_mlp": 1.01592207,
      "epoch": 0.5103108278723021,
      "flos": 22930076787840.0,
      "grad_norm": 1.7619289755129703,
      "language_loss": 0.62463969,
      "learning_rate": 2.0303787259768715e-06,
      "loss": 0.64665496,
      "num_input_tokens_seen": 91508395,
      "step": 4244,
      "time_per_iteration": 3.1928088665008545
    },
    {
      "auxiliary_loss_clip": 0.01151236,
      "auxiliary_loss_mlp": 0.0103564,
      "balance_loss_clip": 1.05264115,
      "balance_loss_mlp": 1.02756989,
      "epoch": 0.5104310707629411,
      "flos": 21506613736320.0,
      "grad_norm": 5.623791405200984,
      "language_loss": 0.70001751,
      "learning_rate": 2.0295998419111294e-06,
      "loss": 0.72188628,
      "num_input_tokens_seen": 91525685,
      "step": 4245,
      "time_per_iteration": 2.5347659587860107
    },
    {
      "auxiliary_loss_clip": 0.01105989,
      "auxiliary_loss_mlp": 0.01028578,
      "balance_loss_clip": 1.04291689,
      "balance_loss_mlp": 1.02051091,
      "epoch": 0.5105513136535802,
      "flos": 14903180403840.0,
      "grad_norm": 3.3553847771161256,
      "language_loss": 0.73540884,
      "learning_rate": 2.028820953355115e-06,
      "loss": 0.75675452,
      "num_input_tokens_seen": 91543785,
      "step": 4246,
      "time_per_iteration": 2.616319417953491
    },
    {
      "auxiliary_loss_clip": 0.01154637,
      "auxiliary_loss_mlp": 0.01027665,
      "balance_loss_clip": 1.04963875,
      "balance_loss_mlp": 1.02023208,
      "epoch": 0.5106715565442194,
      "flos": 22602212421120.0,
      "grad_norm": 1.9120691037856727,
      "language_loss": 0.78607589,
      "learning_rate": 2.0280420604269834e-06,
      "loss": 0.80789888,
      "num_input_tokens_seen": 91563325,
      "step": 4247,
      "time_per_iteration": 2.538588047027588
    },
    {
      "auxiliary_loss_clip": 0.01058061,
      "auxiliary_loss_mlp": 0.01001748,
      "balance_loss_clip": 1.01679182,
      "balance_loss_mlp": 1.00053239,
      "epoch": 0.5107917994348584,
      "flos": 71027645558400.0,
      "grad_norm": 0.7076487985884422,
      "language_loss": 0.5895884,
      "learning_rate": 2.027263163244895e-06,
      "loss": 0.61018646,
      "num_input_tokens_seen": 91632450,
      "step": 4248,
      "time_per_iteration": 4.110201597213745
    },
    {
      "auxiliary_loss_clip": 0.01161138,
      "auxiliary_loss_mlp": 0.01029502,
      "balance_loss_clip": 1.05235934,
      "balance_loss_mlp": 1.02197981,
      "epoch": 0.5109120423254975,
      "flos": 24827416992000.0,
      "grad_norm": 1.5232248889729476,
      "language_loss": 0.74421763,
      "learning_rate": 2.026484261927005e-06,
      "loss": 0.76612401,
      "num_input_tokens_seen": 91651945,
      "step": 4249,
      "time_per_iteration": 2.5455281734466553
    },
    {
      "auxiliary_loss_clip": 0.01171682,
      "auxiliary_loss_mlp": 0.01033018,
      "balance_loss_clip": 1.05535161,
      "balance_loss_mlp": 1.02519751,
      "epoch": 0.5110322852161366,
      "flos": 21247661612160.0,
      "grad_norm": 2.3133633147219066,
      "language_loss": 0.74179596,
      "learning_rate": 2.025705356591475e-06,
      "loss": 0.76384294,
      "num_input_tokens_seen": 91669635,
      "step": 4250,
      "time_per_iteration": 3.2100374698638916
    },
    {
      "auxiliary_loss_clip": 0.01034022,
      "auxiliary_loss_mlp": 0.00751192,
      "balance_loss_clip": 1.0140388,
      "balance_loss_mlp": 1.00013685,
      "epoch": 0.5111525281067757,
      "flos": 66457114358400.0,
      "grad_norm": 0.7620301251467885,
      "language_loss": 0.57974541,
      "learning_rate": 2.024926447356462e-06,
      "loss": 0.5975976,
      "num_input_tokens_seen": 91731920,
      "step": 4251,
      "time_per_iteration": 3.9212234020233154
    },
    {
      "auxiliary_loss_clip": 0.01160513,
      "auxiliary_loss_mlp": 0.01029625,
      "balance_loss_clip": 1.05122852,
      "balance_loss_mlp": 1.02154827,
      "epoch": 0.5112727709974147,
      "flos": 14866731077760.0,
      "grad_norm": 1.832463590767697,
      "language_loss": 0.78780591,
      "learning_rate": 2.024147534340127e-06,
      "loss": 0.80970722,
      "num_input_tokens_seen": 91749780,
      "step": 4252,
      "time_per_iteration": 2.4754793643951416
    },
    {
      "auxiliary_loss_clip": 0.01143618,
      "auxiliary_loss_mlp": 0.01027299,
      "balance_loss_clip": 1.04650879,
      "balance_loss_mlp": 1.01939499,
      "epoch": 0.5113930138880539,
      "flos": 21177600134400.0,
      "grad_norm": 1.622173986512249,
      "language_loss": 0.79724145,
      "learning_rate": 2.02336861766063e-06,
      "loss": 0.81895065,
      "num_input_tokens_seen": 91768840,
      "step": 4253,
      "time_per_iteration": 2.5390281677246094
    },
    {
      "auxiliary_loss_clip": 0.01170614,
      "auxiliary_loss_mlp": 0.01027267,
      "balance_loss_clip": 1.05396771,
      "balance_loss_mlp": 1.01858222,
      "epoch": 0.511513256778693,
      "flos": 20409111630720.0,
      "grad_norm": 1.8552641160080228,
      "language_loss": 0.79037726,
      "learning_rate": 2.0225896974361327e-06,
      "loss": 0.81235605,
      "num_input_tokens_seen": 91788945,
      "step": 4254,
      "time_per_iteration": 2.5272183418273926
    },
    {
      "auxiliary_loss_clip": 0.01037173,
      "auxiliary_loss_mlp": 0.01003101,
      "balance_loss_clip": 1.01716614,
      "balance_loss_mlp": 1.00199211,
      "epoch": 0.511633499669332,
      "flos": 69879975131520.0,
      "grad_norm": 1.0568224739637946,
      "language_loss": 0.59942567,
      "learning_rate": 2.0218107737847962e-06,
      "loss": 0.6198284,
      "num_input_tokens_seen": 91850990,
      "step": 4255,
      "time_per_iteration": 3.1861486434936523
    },
    {
      "auxiliary_loss_clip": 0.01178761,
      "auxiliary_loss_mlp": 0.01026259,
      "balance_loss_clip": 1.05452871,
      "balance_loss_mlp": 1.01867151,
      "epoch": 0.5117537425599712,
      "flos": 24097855852800.0,
      "grad_norm": 2.0071463806249676,
      "language_loss": 0.74596131,
      "learning_rate": 2.0210318468247826e-06,
      "loss": 0.76801145,
      "num_input_tokens_seen": 91869960,
      "step": 4256,
      "time_per_iteration": 2.4899258613586426
    },
    {
      "auxiliary_loss_clip": 0.01145946,
      "auxiliary_loss_mlp": 0.01028569,
      "balance_loss_clip": 1.0475688,
      "balance_loss_mlp": 1.02091575,
      "epoch": 0.5118739854506102,
      "flos": 20959550622720.0,
      "grad_norm": 1.833633324792196,
      "language_loss": 0.81499207,
      "learning_rate": 2.020252916674255e-06,
      "loss": 0.83673728,
      "num_input_tokens_seen": 91889075,
      "step": 4257,
      "time_per_iteration": 2.523074150085449
    },
    {
      "auxiliary_loss_clip": 0.0116426,
      "auxiliary_loss_mlp": 0.01031308,
      "balance_loss_clip": 1.05110884,
      "balance_loss_mlp": 1.02340436,
      "epoch": 0.5119942283412493,
      "flos": 17457326749440.0,
      "grad_norm": 3.2766195322447604,
      "language_loss": 0.81395841,
      "learning_rate": 2.019473983451375e-06,
      "loss": 0.83591408,
      "num_input_tokens_seen": 91907495,
      "step": 4258,
      "time_per_iteration": 2.4823482036590576
    },
    {
      "auxiliary_loss_clip": 0.01139747,
      "auxiliary_loss_mlp": 0.01034248,
      "balance_loss_clip": 1.04798651,
      "balance_loss_mlp": 1.02653563,
      "epoch": 0.5121144712318885,
      "flos": 21066743784960.0,
      "grad_norm": 1.718382687595878,
      "language_loss": 0.71629989,
      "learning_rate": 2.0186950472743076e-06,
      "loss": 0.73803985,
      "num_input_tokens_seen": 91927400,
      "step": 4259,
      "time_per_iteration": 2.5631825923919678
    },
    {
      "auxiliary_loss_clip": 0.01177569,
      "auxiliary_loss_mlp": 0.01029843,
      "balance_loss_clip": 1.05253041,
      "balance_loss_mlp": 1.02259779,
      "epoch": 0.5122347141225275,
      "flos": 19860791541120.0,
      "grad_norm": 1.5907694330539792,
      "language_loss": 0.73875964,
      "learning_rate": 2.0179161082612162e-06,
      "loss": 0.7608338,
      "num_input_tokens_seen": 91946790,
      "step": 4260,
      "time_per_iteration": 2.4622490406036377
    },
    {
      "auxiliary_loss_clip": 0.01142555,
      "auxiliary_loss_mlp": 0.01029346,
      "balance_loss_clip": 1.04540813,
      "balance_loss_mlp": 1.02103686,
      "epoch": 0.5123549570131666,
      "flos": 22528487756160.0,
      "grad_norm": 1.9066259043377503,
      "language_loss": 0.73033667,
      "learning_rate": 2.017137166530266e-06,
      "loss": 0.75205564,
      "num_input_tokens_seen": 91966325,
      "step": 4261,
      "time_per_iteration": 2.518465280532837
    },
    {
      "auxiliary_loss_clip": 0.01153596,
      "auxiliary_loss_mlp": 0.0102991,
      "balance_loss_clip": 1.05061865,
      "balance_loss_mlp": 1.02240241,
      "epoch": 0.5124751999038056,
      "flos": 20333375804160.0,
      "grad_norm": 2.419968695534637,
      "language_loss": 0.79922098,
      "learning_rate": 2.0163582221996213e-06,
      "loss": 0.82105601,
      "num_input_tokens_seen": 91984700,
      "step": 4262,
      "time_per_iteration": 2.526128053665161
    },
    {
      "auxiliary_loss_clip": 0.01149544,
      "auxiliary_loss_mlp": 0.01026539,
      "balance_loss_clip": 1.05141807,
      "balance_loss_mlp": 1.01846302,
      "epoch": 0.5125954427944448,
      "flos": 39785970211200.0,
      "grad_norm": 1.9274577912212143,
      "language_loss": 0.68044186,
      "learning_rate": 2.015579275387446e-06,
      "loss": 0.70220268,
      "num_input_tokens_seen": 92010020,
      "step": 4263,
      "time_per_iteration": 2.6817667484283447
    },
    {
      "auxiliary_loss_clip": 0.01141072,
      "auxiliary_loss_mlp": 0.01039116,
      "balance_loss_clip": 1.0507102,
      "balance_loss_mlp": 1.03150463,
      "epoch": 0.5127156856850839,
      "flos": 29205394358400.0,
      "grad_norm": 1.9466448444659963,
      "language_loss": 0.68668735,
      "learning_rate": 2.0148003262119085e-06,
      "loss": 0.70848924,
      "num_input_tokens_seen": 92030990,
      "step": 4264,
      "time_per_iteration": 2.5926387310028076
    },
    {
      "auxiliary_loss_clip": 0.01132762,
      "auxiliary_loss_mlp": 0.01030384,
      "balance_loss_clip": 1.04861522,
      "balance_loss_mlp": 1.02202141,
      "epoch": 0.5128359285757229,
      "flos": 13553693412480.0,
      "grad_norm": 1.9330508920982399,
      "language_loss": 0.76764655,
      "learning_rate": 2.0140213747911728e-06,
      "loss": 0.78927803,
      "num_input_tokens_seen": 92049525,
      "step": 4265,
      "time_per_iteration": 2.534339666366577
    },
    {
      "auxiliary_loss_clip": 0.01137208,
      "auxiliary_loss_mlp": 0.01027385,
      "balance_loss_clip": 1.05091655,
      "balance_loss_mlp": 1.01956451,
      "epoch": 0.5129561714663621,
      "flos": 25192089820800.0,
      "grad_norm": 2.2984195678513917,
      "language_loss": 0.80954206,
      "learning_rate": 2.013242421243406e-06,
      "loss": 0.83118796,
      "num_input_tokens_seen": 92068430,
      "step": 4266,
      "time_per_iteration": 2.6071014404296875
    },
    {
      "auxiliary_loss_clip": 0.01121758,
      "auxiliary_loss_mlp": 0.01034601,
      "balance_loss_clip": 1.04915237,
      "balance_loss_mlp": 1.02706146,
      "epoch": 0.5130764143570011,
      "flos": 18150223080960.0,
      "grad_norm": 1.7084537595462241,
      "language_loss": 0.79124749,
      "learning_rate": 2.012463465686774e-06,
      "loss": 0.81281102,
      "num_input_tokens_seen": 92088180,
      "step": 4267,
      "time_per_iteration": 2.6276049613952637
    },
    {
      "auxiliary_loss_clip": 0.01021289,
      "auxiliary_loss_mlp": 0.01004738,
      "balance_loss_clip": 1.01466119,
      "balance_loss_mlp": 1.00355172,
      "epoch": 0.5131966572476402,
      "flos": 59794896418560.0,
      "grad_norm": 0.7712023109704312,
      "language_loss": 0.54791576,
      "learning_rate": 2.0116845082394446e-06,
      "loss": 0.56817603,
      "num_input_tokens_seen": 92153015,
      "step": 4268,
      "time_per_iteration": 3.1926779747009277
    },
    {
      "auxiliary_loss_clip": 0.01162787,
      "auxiliary_loss_mlp": 0.01028965,
      "balance_loss_clip": 1.04818869,
      "balance_loss_mlp": 1.02118659,
      "epoch": 0.5133169001382794,
      "flos": 18515219132160.0,
      "grad_norm": 1.9862043950702042,
      "language_loss": 0.78559029,
      "learning_rate": 2.0109055490195836e-06,
      "loss": 0.80750775,
      "num_input_tokens_seen": 92171470,
      "step": 4269,
      "time_per_iteration": 2.490267038345337
    },
    {
      "auxiliary_loss_clip": 0.01106351,
      "auxiliary_loss_mlp": 0.01030836,
      "balance_loss_clip": 1.03785563,
      "balance_loss_mlp": 1.02353394,
      "epoch": 0.5134371430289184,
      "flos": 15523537219200.0,
      "grad_norm": 2.0399401364698275,
      "language_loss": 0.64504135,
      "learning_rate": 2.0101265881453605e-06,
      "loss": 0.66641319,
      "num_input_tokens_seen": 92189945,
      "step": 4270,
      "time_per_iteration": 3.34428334236145
    },
    {
      "auxiliary_loss_clip": 0.01143222,
      "auxiliary_loss_mlp": 0.01030272,
      "balance_loss_clip": 1.05055571,
      "balance_loss_mlp": 1.0232805,
      "epoch": 0.5135573859195575,
      "flos": 21433786911360.0,
      "grad_norm": 2.04778213618343,
      "language_loss": 0.78104997,
      "learning_rate": 2.009347625734941e-06,
      "loss": 0.80278492,
      "num_input_tokens_seen": 92209855,
      "step": 4271,
      "time_per_iteration": 2.5357444286346436
    },
    {
      "auxiliary_loss_clip": 0.01179778,
      "auxiliary_loss_mlp": 0.01039794,
      "balance_loss_clip": 1.05521417,
      "balance_loss_mlp": 1.03196752,
      "epoch": 0.5136776288101966,
      "flos": 17712651600000.0,
      "grad_norm": 2.2186477447461908,
      "language_loss": 0.75172687,
      "learning_rate": 2.0085686619064954e-06,
      "loss": 0.77392256,
      "num_input_tokens_seen": 92226295,
      "step": 4272,
      "time_per_iteration": 2.4501736164093018
    },
    {
      "auxiliary_loss_clip": 0.01166655,
      "auxiliary_loss_mlp": 0.0103537,
      "balance_loss_clip": 1.05247438,
      "balance_loss_mlp": 1.02787161,
      "epoch": 0.5137978717008357,
      "flos": 16581680997120.0,
      "grad_norm": 1.933968326521304,
      "language_loss": 0.82352865,
      "learning_rate": 2.00778969677819e-06,
      "loss": 0.84554887,
      "num_input_tokens_seen": 92243330,
      "step": 4273,
      "time_per_iteration": 2.486768960952759
    },
    {
      "auxiliary_loss_clip": 0.01147613,
      "auxiliary_loss_mlp": 0.0102764,
      "balance_loss_clip": 1.04891896,
      "balance_loss_mlp": 1.02061296,
      "epoch": 0.5139181145914747,
      "flos": 20668243322880.0,
      "grad_norm": 1.823184980667518,
      "language_loss": 0.63962734,
      "learning_rate": 2.0070107304681934e-06,
      "loss": 0.66137987,
      "num_input_tokens_seen": 92262285,
      "step": 4274,
      "time_per_iteration": 3.3091821670532227
    },
    {
      "auxiliary_loss_clip": 0.01135934,
      "auxiliary_loss_mlp": 0.01023819,
      "balance_loss_clip": 1.05141521,
      "balance_loss_mlp": 1.01679432,
      "epoch": 0.5140383574821139,
      "flos": 32926996546560.0,
      "grad_norm": 1.798578448272556,
      "language_loss": 0.77987027,
      "learning_rate": 2.006231763094675e-06,
      "loss": 0.8014679,
      "num_input_tokens_seen": 92283305,
      "step": 4275,
      "time_per_iteration": 2.6731138229370117
    },
    {
      "auxiliary_loss_clip": 0.01144445,
      "auxiliary_loss_mlp": 0.01031023,
      "balance_loss_clip": 1.05271268,
      "balance_loss_mlp": 1.02340508,
      "epoch": 0.514158600372753,
      "flos": 19537093152000.0,
      "grad_norm": 1.8416455320611747,
      "language_loss": 0.87550735,
      "learning_rate": 2.0054527947758027e-06,
      "loss": 0.89726204,
      "num_input_tokens_seen": 92302105,
      "step": 4276,
      "time_per_iteration": 3.227306604385376
    },
    {
      "auxiliary_loss_clip": 0.01055807,
      "auxiliary_loss_mlp": 0.0100296,
      "balance_loss_clip": 1.015576,
      "balance_loss_mlp": 1.00182724,
      "epoch": 0.514278843263392,
      "flos": 62523855279360.0,
      "grad_norm": 0.7327879555078525,
      "language_loss": 0.55964744,
      "learning_rate": 2.004673825629746e-06,
      "loss": 0.58023512,
      "num_input_tokens_seen": 92362885,
      "step": 4277,
      "time_per_iteration": 3.8583853244781494
    },
    {
      "auxiliary_loss_clip": 0.01138662,
      "auxiliary_loss_mlp": 0.01032363,
      "balance_loss_clip": 1.04494953,
      "balance_loss_mlp": 1.02448058,
      "epoch": 0.5143990861540312,
      "flos": 25882328545920.0,
      "grad_norm": 1.6022466719555193,
      "language_loss": 0.72359657,
      "learning_rate": 2.0038948557746744e-06,
      "loss": 0.74530685,
      "num_input_tokens_seen": 92384740,
      "step": 4278,
      "time_per_iteration": 2.579613208770752
    },
    {
      "auxiliary_loss_clip": 0.01161135,
      "auxiliary_loss_mlp": 0.01030187,
      "balance_loss_clip": 1.05219913,
      "balance_loss_mlp": 1.02268279,
      "epoch": 0.5145193290446702,
      "flos": 23330660238720.0,
      "grad_norm": 2.1340653561184886,
      "language_loss": 0.75109047,
      "learning_rate": 2.0031158853287558e-06,
      "loss": 0.7730037,
      "num_input_tokens_seen": 92405175,
      "step": 4279,
      "time_per_iteration": 2.5271520614624023
    },
    {
      "auxiliary_loss_clip": 0.01147987,
      "auxiliary_loss_mlp": 0.01028547,
      "balance_loss_clip": 1.05183756,
      "balance_loss_mlp": 1.0212965,
      "epoch": 0.5146395719353093,
      "flos": 22856603518080.0,
      "grad_norm": 2.451362351287426,
      "language_loss": 0.70303655,
      "learning_rate": 2.0023369144101593e-06,
      "loss": 0.7248019,
      "num_input_tokens_seen": 92423345,
      "step": 4280,
      "time_per_iteration": 2.5385520458221436
    },
    {
      "auxiliary_loss_clip": 0.01138382,
      "auxiliary_loss_mlp": 0.01029801,
      "balance_loss_clip": 1.04646277,
      "balance_loss_mlp": 1.02213025,
      "epoch": 0.5147598148259485,
      "flos": 26391577616640.0,
      "grad_norm": 1.5901620818437956,
      "language_loss": 0.76788986,
      "learning_rate": 2.0015579431370555e-06,
      "loss": 0.7895717,
      "num_input_tokens_seen": 92445025,
      "step": 4281,
      "time_per_iteration": 2.565688133239746
    },
    {
      "auxiliary_loss_clip": 0.01160391,
      "auxiliary_loss_mlp": 0.01028537,
      "balance_loss_clip": 1.05197489,
      "balance_loss_mlp": 1.02172971,
      "epoch": 0.5148800577165875,
      "flos": 29965694561280.0,
      "grad_norm": 2.465023786578449,
      "language_loss": 0.6962899,
      "learning_rate": 2.000778971627612e-06,
      "loss": 0.71817917,
      "num_input_tokens_seen": 92464490,
      "step": 4282,
      "time_per_iteration": 2.5728650093078613
    },
    {
      "auxiliary_loss_clip": 0.01138107,
      "auxiliary_loss_mlp": 0.01032584,
      "balance_loss_clip": 1.04622602,
      "balance_loss_mlp": 1.02537203,
      "epoch": 0.5150003006072266,
      "flos": 17931383470080.0,
      "grad_norm": 1.6752553638808674,
      "language_loss": 0.90569514,
      "learning_rate": 2e-06,
      "loss": 0.92740208,
      "num_input_tokens_seen": 92482085,
      "step": 4283,
      "time_per_iteration": 2.500642776489258
    },
    {
      "auxiliary_loss_clip": 0.01172812,
      "auxiliary_loss_mlp": 0.01022405,
      "balance_loss_clip": 1.05120277,
      "balance_loss_mlp": 1.01517498,
      "epoch": 0.5151205434978657,
      "flos": 18478733892480.0,
      "grad_norm": 1.6942927140333357,
      "language_loss": 0.85317343,
      "learning_rate": 1.9992210283723878e-06,
      "loss": 0.87512565,
      "num_input_tokens_seen": 92499325,
      "step": 4284,
      "time_per_iteration": 2.4344708919525146
    },
    {
      "auxiliary_loss_clip": 0.01170834,
      "auxiliary_loss_mlp": 0.01030659,
      "balance_loss_clip": 1.05148113,
      "balance_loss_mlp": 1.0231576,
      "epoch": 0.5152407863885048,
      "flos": 25341263003520.0,
      "grad_norm": 1.7003654974467028,
      "language_loss": 0.79792345,
      "learning_rate": 1.9984420568629448e-06,
      "loss": 0.81993842,
      "num_input_tokens_seen": 92522090,
      "step": 4285,
      "time_per_iteration": 2.5750534534454346
    },
    {
      "auxiliary_loss_clip": 0.01164562,
      "auxiliary_loss_mlp": 0.01026743,
      "balance_loss_clip": 1.05279434,
      "balance_loss_mlp": 1.01985002,
      "epoch": 0.5153610292791438,
      "flos": 18329740277760.0,
      "grad_norm": 2.044031772054289,
      "language_loss": 0.78132641,
      "learning_rate": 1.9976630855898405e-06,
      "loss": 0.80323946,
      "num_input_tokens_seen": 92539845,
      "step": 4286,
      "time_per_iteration": 2.48429536819458
    },
    {
      "auxiliary_loss_clip": 0.01137803,
      "auxiliary_loss_mlp": 0.0102639,
      "balance_loss_clip": 1.04178846,
      "balance_loss_mlp": 1.01907372,
      "epoch": 0.515481272169783,
      "flos": 30409945971840.0,
      "grad_norm": 2.724509720255688,
      "language_loss": 0.74418485,
      "learning_rate": 1.9968841146712445e-06,
      "loss": 0.76582682,
      "num_input_tokens_seen": 92559460,
      "step": 4287,
      "time_per_iteration": 2.591066598892212
    },
    {
      "auxiliary_loss_clip": 0.01104015,
      "auxiliary_loss_mlp": 0.00760992,
      "balance_loss_clip": 1.04432142,
      "balance_loss_mlp": 1.0002389,
      "epoch": 0.5156015150604221,
      "flos": 23037305863680.0,
      "grad_norm": 1.503322573046224,
      "language_loss": 0.71489048,
      "learning_rate": 1.996105144225326e-06,
      "loss": 0.73354059,
      "num_input_tokens_seen": 92579695,
      "step": 4288,
      "time_per_iteration": 2.635430335998535
    },
    {
      "auxiliary_loss_clip": 0.01156694,
      "auxiliary_loss_mlp": 0.01032607,
      "balance_loss_clip": 1.04894257,
      "balance_loss_mlp": 1.02557361,
      "epoch": 0.5157217579510611,
      "flos": 17858556645120.0,
      "grad_norm": 1.9039009578743982,
      "language_loss": 0.78859711,
      "learning_rate": 1.995326174370254e-06,
      "loss": 0.81049013,
      "num_input_tokens_seen": 92598795,
      "step": 4289,
      "time_per_iteration": 2.489499092102051
    },
    {
      "auxiliary_loss_clip": 0.01157494,
      "auxiliary_loss_mlp": 0.00760406,
      "balance_loss_clip": 1.04804647,
      "balance_loss_mlp": 1.00020742,
      "epoch": 0.5158420008417003,
      "flos": 19171486569600.0,
      "grad_norm": 1.5435974286013598,
      "language_loss": 0.7285282,
      "learning_rate": 1.994547205224197e-06,
      "loss": 0.74770713,
      "num_input_tokens_seen": 92617700,
      "step": 4290,
      "time_per_iteration": 2.483842134475708
    },
    {
      "auxiliary_loss_clip": 0.01141864,
      "auxiliary_loss_mlp": 0.01027229,
      "balance_loss_clip": 1.04927921,
      "balance_loss_mlp": 1.02005291,
      "epoch": 0.5159622437323393,
      "flos": 22419534827520.0,
      "grad_norm": 1.9175684838329485,
      "language_loss": 0.67363322,
      "learning_rate": 1.993768236905325e-06,
      "loss": 0.69532406,
      "num_input_tokens_seen": 92638370,
      "step": 4291,
      "time_per_iteration": 2.525869846343994
    },
    {
      "auxiliary_loss_clip": 0.01141301,
      "auxiliary_loss_mlp": 0.01027326,
      "balance_loss_clip": 1.0459851,
      "balance_loss_mlp": 1.0197922,
      "epoch": 0.5160824866229784,
      "flos": 24603010773120.0,
      "grad_norm": 3.0731533151732795,
      "language_loss": 0.65719104,
      "learning_rate": 1.992989269531807e-06,
      "loss": 0.67887735,
      "num_input_tokens_seen": 92657180,
      "step": 4292,
      "time_per_iteration": 2.538604497909546
    },
    {
      "auxiliary_loss_clip": 0.01140609,
      "auxiliary_loss_mlp": 0.01022867,
      "balance_loss_clip": 1.04499817,
      "balance_loss_mlp": 1.0155828,
      "epoch": 0.5162027295136175,
      "flos": 18002737837440.0,
      "grad_norm": 2.9723148686576883,
      "language_loss": 0.67778778,
      "learning_rate": 1.99221030322181e-06,
      "loss": 0.69942248,
      "num_input_tokens_seen": 92673985,
      "step": 4293,
      "time_per_iteration": 2.4929943084716797
    },
    {
      "auxiliary_loss_clip": 0.01147764,
      "auxiliary_loss_mlp": 0.01024946,
      "balance_loss_clip": 1.04824901,
      "balance_loss_mlp": 1.01815987,
      "epoch": 0.5163229724042566,
      "flos": 27344611221120.0,
      "grad_norm": 1.6434157297672074,
      "language_loss": 0.80803084,
      "learning_rate": 1.991431338093505e-06,
      "loss": 0.82975793,
      "num_input_tokens_seen": 92696340,
      "step": 4294,
      "time_per_iteration": 2.5777831077575684
    },
    {
      "auxiliary_loss_clip": 0.01146114,
      "auxiliary_loss_mlp": 0.01032726,
      "balance_loss_clip": 1.05198085,
      "balance_loss_mlp": 1.02578831,
      "epoch": 0.5164432152948957,
      "flos": 21762764599680.0,
      "grad_norm": 2.8270037749901387,
      "language_loss": 0.79189646,
      "learning_rate": 1.9906523742650587e-06,
      "loss": 0.81368488,
      "num_input_tokens_seen": 92715200,
      "step": 4295,
      "time_per_iteration": 2.5396621227264404
    },
    {
      "auxiliary_loss_clip": 0.01172799,
      "auxiliary_loss_mlp": 0.01023501,
      "balance_loss_clip": 1.0490067,
      "balance_loss_mlp": 1.01610959,
      "epoch": 0.5165634581855347,
      "flos": 25550334115200.0,
      "grad_norm": 1.9802139888884258,
      "language_loss": 0.77721947,
      "learning_rate": 1.9898734118546397e-06,
      "loss": 0.79918242,
      "num_input_tokens_seen": 92735150,
      "step": 4296,
      "time_per_iteration": 3.2893972396850586
    },
    {
      "auxiliary_loss_clip": 0.01092482,
      "auxiliary_loss_mlp": 0.01030336,
      "balance_loss_clip": 1.04241908,
      "balance_loss_mlp": 1.02294159,
      "epoch": 0.5166837010761739,
      "flos": 19901191363200.0,
      "grad_norm": 1.562076816308385,
      "language_loss": 0.80148435,
      "learning_rate": 1.989094450980416e-06,
      "loss": 0.82271254,
      "num_input_tokens_seen": 92755250,
      "step": 4297,
      "time_per_iteration": 2.7436141967773438
    },
    {
      "auxiliary_loss_clip": 0.01158957,
      "auxiliary_loss_mlp": 0.01023945,
      "balance_loss_clip": 1.05072916,
      "balance_loss_mlp": 1.016891,
      "epoch": 0.516803943966813,
      "flos": 26646076454400.0,
      "grad_norm": 1.8129591042748034,
      "language_loss": 0.76784611,
      "learning_rate": 1.9883154917605556e-06,
      "loss": 0.78967512,
      "num_input_tokens_seen": 92774460,
      "step": 4298,
      "time_per_iteration": 2.966700792312622
    },
    {
      "auxiliary_loss_clip": 0.01172466,
      "auxiliary_loss_mlp": 0.01028643,
      "balance_loss_clip": 1.05182445,
      "balance_loss_mlp": 1.02154064,
      "epoch": 0.516924186857452,
      "flos": 19682854542720.0,
      "grad_norm": 1.698789376289148,
      "language_loss": 0.83196694,
      "learning_rate": 1.9875365343132262e-06,
      "loss": 0.85397804,
      "num_input_tokens_seen": 92791580,
      "step": 4299,
      "time_per_iteration": 2.484851837158203
    },
    {
      "auxiliary_loss_clip": 0.01160496,
      "auxiliary_loss_mlp": 0.00760209,
      "balance_loss_clip": 1.05326211,
      "balance_loss_mlp": 1.00021482,
      "epoch": 0.5170444297480912,
      "flos": 15956583586560.0,
      "grad_norm": 2.3987613708300466,
      "language_loss": 0.85086077,
      "learning_rate": 1.9867575787565946e-06,
      "loss": 0.87006783,
      "num_input_tokens_seen": 92806240,
      "step": 4300,
      "time_per_iteration": 3.2455408573150635
    },
    {
      "auxiliary_loss_clip": 0.0116075,
      "auxiliary_loss_mlp": 0.01032031,
      "balance_loss_clip": 1.05208707,
      "balance_loss_mlp": 1.0238297,
      "epoch": 0.5171646726387302,
      "flos": 14174157968640.0,
      "grad_norm": 1.9726836595820358,
      "language_loss": 0.8623296,
      "learning_rate": 1.9859786252088275e-06,
      "loss": 0.88425744,
      "num_input_tokens_seen": 92823420,
      "step": 4301,
      "time_per_iteration": 2.489762783050537
    },
    {
      "auxiliary_loss_clip": 0.01136203,
      "auxiliary_loss_mlp": 0.01030202,
      "balance_loss_clip": 1.04831958,
      "balance_loss_mlp": 1.02297187,
      "epoch": 0.5172849155293693,
      "flos": 23578550974080.0,
      "grad_norm": 2.5937381900647836,
      "language_loss": 0.6657244,
      "learning_rate": 1.9851996737880914e-06,
      "loss": 0.68738848,
      "num_input_tokens_seen": 92838605,
      "step": 4302,
      "time_per_iteration": 2.5541136264801025
    },
    {
      "auxiliary_loss_clip": 0.01166765,
      "auxiliary_loss_mlp": 0.01027333,
      "balance_loss_clip": 1.05358434,
      "balance_loss_mlp": 1.01947117,
      "epoch": 0.5174051584200084,
      "flos": 14283541860480.0,
      "grad_norm": 2.4516718615943778,
      "language_loss": 0.74612677,
      "learning_rate": 1.9844207246125537e-06,
      "loss": 0.76806778,
      "num_input_tokens_seen": 92855185,
      "step": 4303,
      "time_per_iteration": 3.211686849594116
    },
    {
      "auxiliary_loss_clip": 0.011426,
      "auxiliary_loss_mlp": 0.01026196,
      "balance_loss_clip": 1.0474503,
      "balance_loss_mlp": 1.01904607,
      "epoch": 0.5175254013106475,
      "flos": 37889384192640.0,
      "grad_norm": 1.815214287715659,
      "language_loss": 0.68193781,
      "learning_rate": 1.983641777800379e-06,
      "loss": 0.7036258,
      "num_input_tokens_seen": 92877830,
      "step": 4304,
      "time_per_iteration": 3.4980385303497314
    },
    {
      "auxiliary_loss_clip": 0.01048729,
      "auxiliary_loss_mlp": 0.01006993,
      "balance_loss_clip": 1.01463199,
      "balance_loss_mlp": 1.00587273,
      "epoch": 0.5176456442012866,
      "flos": 68549737829760.0,
      "grad_norm": 0.7416688085340349,
      "language_loss": 0.5877192,
      "learning_rate": 1.9828628334697343e-06,
      "loss": 0.60827649,
      "num_input_tokens_seen": 92945040,
      "step": 4305,
      "time_per_iteration": 3.2807750701904297
    },
    {
      "auxiliary_loss_clip": 0.01049809,
      "auxiliary_loss_mlp": 0.01005186,
      "balance_loss_clip": 1.01679778,
      "balance_loss_mlp": 1.00392866,
      "epoch": 0.5177658870919257,
      "flos": 64084137235200.0,
      "grad_norm": 0.7628883832547182,
      "language_loss": 0.54724693,
      "learning_rate": 1.982083891738784e-06,
      "loss": 0.56779695,
      "num_input_tokens_seen": 93005910,
      "step": 4306,
      "time_per_iteration": 3.160919189453125
    },
    {
      "auxiliary_loss_clip": 0.01139996,
      "auxiliary_loss_mlp": 0.01022938,
      "balance_loss_clip": 1.05084968,
      "balance_loss_mlp": 1.01572907,
      "epoch": 0.5178861299825648,
      "flos": 26651248012800.0,
      "grad_norm": 1.4370681641474516,
      "language_loss": 0.82882369,
      "learning_rate": 1.9813049527256923e-06,
      "loss": 0.85045302,
      "num_input_tokens_seen": 93026305,
      "step": 4307,
      "time_per_iteration": 2.576843500137329
    },
    {
      "auxiliary_loss_clip": 0.01130005,
      "auxiliary_loss_mlp": 0.01028712,
      "balance_loss_clip": 1.04522121,
      "balance_loss_mlp": 1.0213685,
      "epoch": 0.5180063728732038,
      "flos": 17931886260480.0,
      "grad_norm": 2.3869369702616114,
      "language_loss": 0.8236984,
      "learning_rate": 1.9805260165486252e-06,
      "loss": 0.84528559,
      "num_input_tokens_seen": 93045675,
      "step": 4308,
      "time_per_iteration": 2.548743724822998
    },
    {
      "auxiliary_loss_clip": 0.01159085,
      "auxiliary_loss_mlp": 0.0102944,
      "balance_loss_clip": 1.05028069,
      "balance_loss_mlp": 1.02231109,
      "epoch": 0.518126615763843,
      "flos": 19500895221120.0,
      "grad_norm": 1.8344886278482482,
      "language_loss": 0.86412328,
      "learning_rate": 1.9797470833257457e-06,
      "loss": 0.8860085,
      "num_input_tokens_seen": 93065375,
      "step": 4309,
      "time_per_iteration": 2.4941225051879883
    },
    {
      "auxiliary_loss_clip": 0.01161356,
      "auxiliary_loss_mlp": 0.01026601,
      "balance_loss_clip": 1.05311108,
      "balance_loss_mlp": 1.01927006,
      "epoch": 0.5182468586544821,
      "flos": 20704082117760.0,
      "grad_norm": 1.9456945097354985,
      "language_loss": 0.77518833,
      "learning_rate": 1.9789681531752177e-06,
      "loss": 0.79706788,
      "num_input_tokens_seen": 93085595,
      "step": 4310,
      "time_per_iteration": 2.4898009300231934
    },
    {
      "auxiliary_loss_clip": 0.01117763,
      "auxiliary_loss_mlp": 0.01025721,
      "balance_loss_clip": 1.04811454,
      "balance_loss_mlp": 1.01907218,
      "epoch": 0.5183671015451211,
      "flos": 23112107936640.0,
      "grad_norm": 1.526410840470581,
      "language_loss": 0.72547281,
      "learning_rate": 1.978189226215204e-06,
      "loss": 0.74690759,
      "num_input_tokens_seen": 93106140,
      "step": 4311,
      "time_per_iteration": 2.6112945079803467
    },
    {
      "auxiliary_loss_clip": 0.01173083,
      "auxiliary_loss_mlp": 0.01028163,
      "balance_loss_clip": 1.05144417,
      "balance_loss_mlp": 1.02056897,
      "epoch": 0.5184873444357603,
      "flos": 17597090568960.0,
      "grad_norm": 2.1327617536273684,
      "language_loss": 0.77043498,
      "learning_rate": 1.9774103025638675e-06,
      "loss": 0.79244745,
      "num_input_tokens_seen": 93124265,
      "step": 4312,
      "time_per_iteration": 2.4537482261657715
    },
    {
      "auxiliary_loss_clip": 0.0112392,
      "auxiliary_loss_mlp": 0.01024466,
      "balance_loss_clip": 1.05333972,
      "balance_loss_mlp": 1.01717305,
      "epoch": 0.5186075873263993,
      "flos": 24936800883840.0,
      "grad_norm": 1.5101858690840215,
      "language_loss": 0.76432234,
      "learning_rate": 1.9766313823393696e-06,
      "loss": 0.78580618,
      "num_input_tokens_seen": 93145130,
      "step": 4313,
      "time_per_iteration": 2.6433751583099365
    },
    {
      "auxiliary_loss_clip": 0.01111494,
      "auxiliary_loss_mlp": 0.01024891,
      "balance_loss_clip": 1.04320371,
      "balance_loss_mlp": 1.01736319,
      "epoch": 0.5187278302170384,
      "flos": 15190106244480.0,
      "grad_norm": 1.9201772478816068,
      "language_loss": 0.69114101,
      "learning_rate": 1.975852465659873e-06,
      "loss": 0.71250486,
      "num_input_tokens_seen": 93161110,
      "step": 4314,
      "time_per_iteration": 2.5761466026306152
    },
    {
      "auxiliary_loss_clip": 0.01162248,
      "auxiliary_loss_mlp": 0.01028845,
      "balance_loss_clip": 1.05179429,
      "balance_loss_mlp": 1.02086973,
      "epoch": 0.5188480731076776,
      "flos": 25009412227200.0,
      "grad_norm": 2.5971185341169787,
      "language_loss": 0.70033312,
      "learning_rate": 1.9750735526435377e-06,
      "loss": 0.72224402,
      "num_input_tokens_seen": 93178055,
      "step": 4315,
      "time_per_iteration": 2.5180904865264893
    },
    {
      "auxiliary_loss_clip": 0.01144965,
      "auxiliary_loss_mlp": 0.01031149,
      "balance_loss_clip": 1.05094957,
      "balance_loss_mlp": 1.02378178,
      "epoch": 0.5189683159983166,
      "flos": 24790141653120.0,
      "grad_norm": 2.259954388130587,
      "language_loss": 0.79730552,
      "learning_rate": 1.974294643408525e-06,
      "loss": 0.81906664,
      "num_input_tokens_seen": 93195850,
      "step": 4316,
      "time_per_iteration": 2.561110496520996
    },
    {
      "auxiliary_loss_clip": 0.01161956,
      "auxiliary_loss_mlp": 0.01028147,
      "balance_loss_clip": 1.0490998,
      "balance_loss_mlp": 1.02073264,
      "epoch": 0.5190885588889557,
      "flos": 24754266944640.0,
      "grad_norm": 1.7838056486556368,
      "language_loss": 0.66797632,
      "learning_rate": 1.9735157380729947e-06,
      "loss": 0.68987727,
      "num_input_tokens_seen": 93216260,
      "step": 4317,
      "time_per_iteration": 2.542891025543213
    },
    {
      "auxiliary_loss_clip": 0.01145044,
      "auxiliary_loss_mlp": 0.01022218,
      "balance_loss_clip": 1.04825282,
      "balance_loss_mlp": 1.01531601,
      "epoch": 0.5192088017795948,
      "flos": 24712646060160.0,
      "grad_norm": 3.2412825707357977,
      "language_loss": 0.83921504,
      "learning_rate": 1.9727368367551053e-06,
      "loss": 0.86088765,
      "num_input_tokens_seen": 93234810,
      "step": 4318,
      "time_per_iteration": 2.585916519165039
    },
    {
      "auxiliary_loss_clip": 0.01133564,
      "auxiliary_loss_mlp": 0.01025845,
      "balance_loss_clip": 1.04657555,
      "balance_loss_mlp": 1.01831746,
      "epoch": 0.5193290446702339,
      "flos": 27229588894080.0,
      "grad_norm": 1.8749561341324799,
      "language_loss": 0.68556756,
      "learning_rate": 1.9719579395730164e-06,
      "loss": 0.70716166,
      "num_input_tokens_seen": 93254185,
      "step": 4319,
      "time_per_iteration": 2.6010890007019043
    },
    {
      "auxiliary_loss_clip": 0.01175793,
      "auxiliary_loss_mlp": 0.01031578,
      "balance_loss_clip": 1.05436301,
      "balance_loss_mlp": 1.02354276,
      "epoch": 0.5194492875608729,
      "flos": 11473352392320.0,
      "grad_norm": 1.9512024821535063,
      "language_loss": 0.93017763,
      "learning_rate": 1.9711790466448854e-06,
      "loss": 0.95225132,
      "num_input_tokens_seen": 93268205,
      "step": 4320,
      "time_per_iteration": 2.4232699871063232
    },
    {
      "auxiliary_loss_clip": 0.01121318,
      "auxiliary_loss_mlp": 0.0103391,
      "balance_loss_clip": 1.04561245,
      "balance_loss_mlp": 1.02583957,
      "epoch": 0.5195695304515121,
      "flos": 20338906498560.0,
      "grad_norm": 1.9815859668866573,
      "language_loss": 0.71214592,
      "learning_rate": 1.9704001580888704e-06,
      "loss": 0.73369813,
      "num_input_tokens_seen": 93286945,
      "step": 4321,
      "time_per_iteration": 2.618446111679077
    },
    {
      "auxiliary_loss_clip": 0.01140572,
      "auxiliary_loss_mlp": 0.00760529,
      "balance_loss_clip": 1.04553103,
      "balance_loss_mlp": 1.00026608,
      "epoch": 0.5196897733421512,
      "flos": 20048317470720.0,
      "grad_norm": 1.7791663164805345,
      "language_loss": 0.87203765,
      "learning_rate": 1.9696212740231283e-06,
      "loss": 0.89104867,
      "num_input_tokens_seen": 93305595,
      "step": 4322,
      "time_per_iteration": 3.2443859577178955
    },
    {
      "auxiliary_loss_clip": 0.011627,
      "auxiliary_loss_mlp": 0.01031559,
      "balance_loss_clip": 1.04720616,
      "balance_loss_mlp": 1.02347028,
      "epoch": 0.5198100162327902,
      "flos": 23805507058560.0,
      "grad_norm": 2.06420142559047,
      "language_loss": 0.82255387,
      "learning_rate": 1.9688423945658146e-06,
      "loss": 0.84449649,
      "num_input_tokens_seen": 93326460,
      "step": 4323,
      "time_per_iteration": 2.5548367500305176
    },
    {
      "auxiliary_loss_clip": 0.01106995,
      "auxiliary_loss_mlp": 0.01031549,
      "balance_loss_clip": 1.03936243,
      "balance_loss_mlp": 1.02356827,
      "epoch": 0.5199302591234293,
      "flos": 24023951619840.0,
      "grad_norm": 2.184852163598746,
      "language_loss": 0.7212401,
      "learning_rate": 1.9680635198350845e-06,
      "loss": 0.74262553,
      "num_input_tokens_seen": 93346170,
      "step": 4324,
      "time_per_iteration": 2.611891508102417
    },
    {
      "auxiliary_loss_clip": 0.01160757,
      "auxiliary_loss_mlp": 0.01029769,
      "balance_loss_clip": 1.04888391,
      "balance_loss_mlp": 1.02135921,
      "epoch": 0.5200505020140684,
      "flos": 26359366095360.0,
      "grad_norm": 2.086551594614165,
      "language_loss": 0.72409129,
      "learning_rate": 1.967284649949093e-06,
      "loss": 0.74599653,
      "num_input_tokens_seen": 93365380,
      "step": 4325,
      "time_per_iteration": 2.5255751609802246
    },
    {
      "auxiliary_loss_clip": 0.01125953,
      "auxiliary_loss_mlp": 0.01028322,
      "balance_loss_clip": 1.04318893,
      "balance_loss_mlp": 1.02036476,
      "epoch": 0.5201707449047075,
      "flos": 39604262284800.0,
      "grad_norm": 3.2213747430279027,
      "language_loss": 0.72593844,
      "learning_rate": 1.966505785025994e-06,
      "loss": 0.74748117,
      "num_input_tokens_seen": 93387285,
      "step": 4326,
      "time_per_iteration": 3.4943788051605225
    },
    {
      "auxiliary_loss_clip": 0.01129897,
      "auxiliary_loss_mlp": 0.01029594,
      "balance_loss_clip": 1.04825616,
      "balance_loss_mlp": 1.02190471,
      "epoch": 0.5202909877953465,
      "flos": 53682788292480.0,
      "grad_norm": 1.652744238698553,
      "language_loss": 0.7614094,
      "learning_rate": 1.965726925183941e-06,
      "loss": 0.7830044,
      "num_input_tokens_seen": 93410390,
      "step": 4327,
      "time_per_iteration": 2.824617624282837
    },
    {
      "auxiliary_loss_clip": 0.01177117,
      "auxiliary_loss_mlp": 0.01032278,
      "balance_loss_clip": 1.05462515,
      "balance_loss_mlp": 1.02514899,
      "epoch": 0.5204112306859857,
      "flos": 19537021324800.0,
      "grad_norm": 1.691198331155662,
      "language_loss": 0.84921753,
      "learning_rate": 1.964948070541087e-06,
      "loss": 0.87131149,
      "num_input_tokens_seen": 93429050,
      "step": 4328,
      "time_per_iteration": 2.449265480041504
    },
    {
      "auxiliary_loss_clip": 0.01149014,
      "auxiliary_loss_mlp": 0.01027824,
      "balance_loss_clip": 1.04669809,
      "balance_loss_mlp": 1.02057326,
      "epoch": 0.5205314735766248,
      "flos": 15304697608320.0,
      "grad_norm": 2.412532269474611,
      "language_loss": 0.69546652,
      "learning_rate": 1.9641692212155816e-06,
      "loss": 0.71723497,
      "num_input_tokens_seen": 93446815,
      "step": 4329,
      "time_per_iteration": 4.035771608352661
    },
    {
      "auxiliary_loss_clip": 0.01117321,
      "auxiliary_loss_mlp": 0.01033209,
      "balance_loss_clip": 1.04851675,
      "balance_loss_mlp": 1.02580285,
      "epoch": 0.5206517164672638,
      "flos": 59263701160320.0,
      "grad_norm": 1.7867410301765996,
      "language_loss": 0.72398216,
      "learning_rate": 1.9633903773255777e-06,
      "loss": 0.74548745,
      "num_input_tokens_seen": 93469130,
      "step": 4330,
      "time_per_iteration": 2.9070353507995605
    },
    {
      "auxiliary_loss_clip": 0.01171196,
      "auxiliary_loss_mlp": 0.01026661,
      "balance_loss_clip": 1.05031657,
      "balance_loss_mlp": 1.01892388,
      "epoch": 0.520771959357903,
      "flos": 26871129118080.0,
      "grad_norm": 1.6932988932973856,
      "language_loss": 0.74663693,
      "learning_rate": 1.9626115389892237e-06,
      "loss": 0.76861548,
      "num_input_tokens_seen": 93489920,
      "step": 4331,
      "time_per_iteration": 2.5091192722320557
    },
    {
      "auxiliary_loss_clip": 0.01138332,
      "auxiliary_loss_mlp": 0.01024844,
      "balance_loss_clip": 1.04996037,
      "balance_loss_mlp": 1.01762557,
      "epoch": 0.520892202248542,
      "flos": 26907075653760.0,
      "grad_norm": 1.9457100968877312,
      "language_loss": 0.85235435,
      "learning_rate": 1.96183270632467e-06,
      "loss": 0.87398612,
      "num_input_tokens_seen": 93509770,
      "step": 4332,
      "time_per_iteration": 2.580357551574707
    },
    {
      "auxiliary_loss_clip": 0.01123231,
      "auxiliary_loss_mlp": 0.00761652,
      "balance_loss_clip": 1.04533958,
      "balance_loss_mlp": 1.00020421,
      "epoch": 0.5210124451391811,
      "flos": 25849434666240.0,
      "grad_norm": 1.6158249132612017,
      "language_loss": 0.79357386,
      "learning_rate": 1.9610538794500644e-06,
      "loss": 0.81242263,
      "num_input_tokens_seen": 93529320,
      "step": 4333,
      "time_per_iteration": 2.5829505920410156
    },
    {
      "auxiliary_loss_clip": 0.01035336,
      "auxiliary_loss_mlp": 0.01001922,
      "balance_loss_clip": 1.01303053,
      "balance_loss_mlp": 1.00080705,
      "epoch": 0.5211326880298203,
      "flos": 70553804319360.0,
      "grad_norm": 0.7750401140738005,
      "language_loss": 0.59450758,
      "learning_rate": 1.9602750584835542e-06,
      "loss": 0.6148802,
      "num_input_tokens_seen": 93595255,
      "step": 4334,
      "time_per_iteration": 3.244784116744995
    },
    {
      "auxiliary_loss_clip": 0.01141891,
      "auxiliary_loss_mlp": 0.01025101,
      "balance_loss_clip": 1.04608369,
      "balance_loss_mlp": 1.01795387,
      "epoch": 0.5212529309204593,
      "flos": 15628898787840.0,
      "grad_norm": 2.1032570279648586,
      "language_loss": 0.82714176,
      "learning_rate": 1.959496243543286e-06,
      "loss": 0.84881163,
      "num_input_tokens_seen": 93613135,
      "step": 4335,
      "time_per_iteration": 2.501112222671509
    },
    {
      "auxiliary_loss_clip": 0.01169467,
      "auxiliary_loss_mlp": 0.01038869,
      "balance_loss_clip": 1.0576998,
      "balance_loss_mlp": 1.03119802,
      "epoch": 0.5213731738110984,
      "flos": 26242655829120.0,
      "grad_norm": 5.267384649767942,
      "language_loss": 0.7927568,
      "learning_rate": 1.9587174347474057e-06,
      "loss": 0.8148402,
      "num_input_tokens_seen": 93629645,
      "step": 4336,
      "time_per_iteration": 2.517946481704712
    },
    {
      "auxiliary_loss_clip": 0.01104374,
      "auxiliary_loss_mlp": 0.01030626,
      "balance_loss_clip": 1.04257238,
      "balance_loss_mlp": 1.02295518,
      "epoch": 0.5214934167017375,
      "flos": 19418407637760.0,
      "grad_norm": 2.897984696764316,
      "language_loss": 0.81706464,
      "learning_rate": 1.9579386322140574e-06,
      "loss": 0.83841467,
      "num_input_tokens_seen": 93645325,
      "step": 4337,
      "time_per_iteration": 2.5837342739105225
    },
    {
      "auxiliary_loss_clip": 0.01177382,
      "auxiliary_loss_mlp": 0.00761355,
      "balance_loss_clip": 1.05347574,
      "balance_loss_mlp": 1.00024486,
      "epoch": 0.5216136595923766,
      "flos": 30955788023040.0,
      "grad_norm": 1.7238683178322511,
      "language_loss": 0.80713773,
      "learning_rate": 1.9571598360613854e-06,
      "loss": 0.82652509,
      "num_input_tokens_seen": 93668200,
      "step": 4338,
      "time_per_iteration": 2.5425260066986084
    },
    {
      "auxiliary_loss_clip": 0.01132269,
      "auxiliary_loss_mlp": 0.0103278,
      "balance_loss_clip": 1.04461873,
      "balance_loss_mlp": 1.02483499,
      "epoch": 0.5217339024830157,
      "flos": 21945047143680.0,
      "grad_norm": 2.237095052034925,
      "language_loss": 0.6967032,
      "learning_rate": 1.956381046407532e-06,
      "loss": 0.71835375,
      "num_input_tokens_seen": 93688495,
      "step": 4339,
      "time_per_iteration": 2.5344159603118896
    },
    {
      "auxiliary_loss_clip": 0.01128862,
      "auxiliary_loss_mlp": 0.01024317,
      "balance_loss_clip": 1.04589725,
      "balance_loss_mlp": 1.01691985,
      "epoch": 0.5218541453736548,
      "flos": 20923209037440.0,
      "grad_norm": 2.2745408874591173,
      "language_loss": 0.86170483,
      "learning_rate": 1.9556022633706394e-06,
      "loss": 0.88323665,
      "num_input_tokens_seen": 93707285,
      "step": 4340,
      "time_per_iteration": 2.620112180709839
    },
    {
      "auxiliary_loss_clip": 0.01140042,
      "auxiliary_loss_mlp": 0.01024428,
      "balance_loss_clip": 1.04697442,
      "balance_loss_mlp": 1.01685774,
      "epoch": 0.5219743882642939,
      "flos": 23951663498880.0,
      "grad_norm": 1.7515197537141873,
      "language_loss": 0.79680735,
      "learning_rate": 1.954823487068848e-06,
      "loss": 0.818452,
      "num_input_tokens_seen": 93727495,
      "step": 4341,
      "time_per_iteration": 2.58849835395813
    },
    {
      "auxiliary_loss_clip": 0.011613,
      "auxiliary_loss_mlp": 0.01031828,
      "balance_loss_clip": 1.05462301,
      "balance_loss_mlp": 1.02478909,
      "epoch": 0.5220946311549329,
      "flos": 28799280213120.0,
      "grad_norm": 1.613758661386504,
      "language_loss": 0.80984467,
      "learning_rate": 1.9540447176202976e-06,
      "loss": 0.83177596,
      "num_input_tokens_seen": 93748740,
      "step": 4342,
      "time_per_iteration": 2.574246406555176
    },
    {
      "auxiliary_loss_clip": 0.01054609,
      "auxiliary_loss_mlp": 0.0100149,
      "balance_loss_clip": 1.01460981,
      "balance_loss_mlp": 1.00031018,
      "epoch": 0.5222148740455721,
      "flos": 67189369017600.0,
      "grad_norm": 0.9085845522109597,
      "language_loss": 0.60751379,
      "learning_rate": 1.9532659551431272e-06,
      "loss": 0.62807477,
      "num_input_tokens_seen": 93815770,
      "step": 4343,
      "time_per_iteration": 3.255690097808838
    },
    {
      "auxiliary_loss_clip": 0.01162333,
      "auxiliary_loss_mlp": 0.01029815,
      "balance_loss_clip": 1.05084801,
      "balance_loss_mlp": 1.02241802,
      "epoch": 0.5223351169362112,
      "flos": 61856164339200.0,
      "grad_norm": 1.6197307792173696,
      "language_loss": 0.67538464,
      "learning_rate": 1.9524871997554744e-06,
      "loss": 0.69730616,
      "num_input_tokens_seen": 93843530,
      "step": 4344,
      "time_per_iteration": 2.866302490234375
    },
    {
      "auxiliary_loss_clip": 0.01162189,
      "auxiliary_loss_mlp": 0.01027231,
      "balance_loss_clip": 1.05266571,
      "balance_loss_mlp": 1.01991725,
      "epoch": 0.5224553598268502,
      "flos": 14647388676480.0,
      "grad_norm": 2.1186597073799263,
      "language_loss": 0.8056972,
      "learning_rate": 1.951708451575475e-06,
      "loss": 0.82759142,
      "num_input_tokens_seen": 93860595,
      "step": 4345,
      "time_per_iteration": 2.459747791290283
    },
    {
      "auxiliary_loss_clip": 0.01139216,
      "auxiliary_loss_mlp": 0.01025429,
      "balance_loss_clip": 1.04612708,
      "balance_loss_mlp": 1.01786518,
      "epoch": 0.5225756027174894,
      "flos": 14826043946880.0,
      "grad_norm": 2.011513654271153,
      "language_loss": 0.82454473,
      "learning_rate": 1.9509297107212657e-06,
      "loss": 0.84619117,
      "num_input_tokens_seen": 93877365,
      "step": 4346,
      "time_per_iteration": 2.540949583053589
    },
    {
      "auxiliary_loss_clip": 0.0117392,
      "auxiliary_loss_mlp": 0.01025493,
      "balance_loss_clip": 1.05204129,
      "balance_loss_mlp": 1.0182631,
      "epoch": 0.5226958456081284,
      "flos": 23512009029120.0,
      "grad_norm": 1.6236361113761322,
      "language_loss": 0.79030716,
      "learning_rate": 1.95015097731098e-06,
      "loss": 0.81230128,
      "num_input_tokens_seen": 93896855,
      "step": 4347,
      "time_per_iteration": 2.4724838733673096
    },
    {
      "auxiliary_loss_clip": 0.01175885,
      "auxiliary_loss_mlp": 0.01025367,
      "balance_loss_clip": 1.05333185,
      "balance_loss_mlp": 1.01806557,
      "epoch": 0.5228160884987675,
      "flos": 19062928690560.0,
      "grad_norm": 2.082196817356619,
      "language_loss": 0.81988221,
      "learning_rate": 1.949372251462751e-06,
      "loss": 0.84189463,
      "num_input_tokens_seen": 93914270,
      "step": 4348,
      "time_per_iteration": 3.157294511795044
    },
    {
      "auxiliary_loss_clip": 0.01133509,
      "auxiliary_loss_mlp": 0.0076044,
      "balance_loss_clip": 1.04779506,
      "balance_loss_mlp": 1.00024152,
      "epoch": 0.5229363313894067,
      "flos": 21063224252160.0,
      "grad_norm": 2.0394488985735437,
      "language_loss": 0.82965827,
      "learning_rate": 1.9485935332947124e-06,
      "loss": 0.84859776,
      "num_input_tokens_seen": 93932180,
      "step": 4349,
      "time_per_iteration": 2.5740952491760254
    },
    {
      "auxiliary_loss_clip": 0.01140629,
      "auxiliary_loss_mlp": 0.01028169,
      "balance_loss_clip": 1.04833508,
      "balance_loss_mlp": 1.0213027,
      "epoch": 0.5230565742800457,
      "flos": 14830389492480.0,
      "grad_norm": 2.5631441943906954,
      "language_loss": 0.83755541,
      "learning_rate": 1.947814822924993e-06,
      "loss": 0.85924333,
      "num_input_tokens_seen": 93949690,
      "step": 4350,
      "time_per_iteration": 2.5039548873901367
    },
    {
      "auxiliary_loss_clip": 0.01176177,
      "auxiliary_loss_mlp": 0.01027432,
      "balance_loss_clip": 1.05493808,
      "balance_loss_mlp": 1.02038121,
      "epoch": 0.5231768171706848,
      "flos": 25813021253760.0,
      "grad_norm": 1.9205349783857655,
      "language_loss": 0.82680279,
      "learning_rate": 1.9470361204717236e-06,
      "loss": 0.84883887,
      "num_input_tokens_seen": 93968830,
      "step": 4351,
      "time_per_iteration": 2.501235246658325
    },
    {
      "auxiliary_loss_clip": 0.01134154,
      "auxiliary_loss_mlp": 0.00760982,
      "balance_loss_clip": 1.04556167,
      "balance_loss_mlp": 1.0002116,
      "epoch": 0.5232970600613239,
      "flos": 22743807834240.0,
      "grad_norm": 1.5978927249794943,
      "language_loss": 0.80635512,
      "learning_rate": 1.9462574260530326e-06,
      "loss": 0.82530648,
      "num_input_tokens_seen": 93989110,
      "step": 4352,
      "time_per_iteration": 3.3562674522399902
    },
    {
      "auxiliary_loss_clip": 0.01150271,
      "auxiliary_loss_mlp": 0.01021718,
      "balance_loss_clip": 1.0486815,
      "balance_loss_mlp": 1.01419556,
      "epoch": 0.523417302951963,
      "flos": 17310703432320.0,
      "grad_norm": 1.7879170412746461,
      "language_loss": 0.81106943,
      "learning_rate": 1.9454787397870472e-06,
      "loss": 0.8327893,
      "num_input_tokens_seen": 94006430,
      "step": 4353,
      "time_per_iteration": 2.487812042236328
    },
    {
      "auxiliary_loss_clip": 0.01092645,
      "auxiliary_loss_mlp": 0.01026017,
      "balance_loss_clip": 1.04470205,
      "balance_loss_mlp": 1.01873624,
      "epoch": 0.523537545842602,
      "flos": 18551740285440.0,
      "grad_norm": 1.946823981156344,
      "language_loss": 0.71794719,
      "learning_rate": 1.944700061791894e-06,
      "loss": 0.73913378,
      "num_input_tokens_seen": 94024825,
      "step": 4354,
      "time_per_iteration": 2.5891847610473633
    },
    {
      "auxiliary_loss_clip": 0.01159527,
      "auxiliary_loss_mlp": 0.01029187,
      "balance_loss_clip": 1.05085588,
      "balance_loss_mlp": 1.02244294,
      "epoch": 0.5236577887332411,
      "flos": 19719267955200.0,
      "grad_norm": 2.087160255965262,
      "language_loss": 0.65698278,
      "learning_rate": 1.943921392185698e-06,
      "loss": 0.67886996,
      "num_input_tokens_seen": 94043450,
      "step": 4355,
      "time_per_iteration": 3.348118543624878
    },
    {
      "auxiliary_loss_clip": 0.01145809,
      "auxiliary_loss_mlp": 0.01026692,
      "balance_loss_clip": 1.04789948,
      "balance_loss_mlp": 1.01986098,
      "epoch": 0.5237780316238803,
      "flos": 23550218121600.0,
      "grad_norm": 2.0378390808981814,
      "language_loss": 0.77018619,
      "learning_rate": 1.9431427310865814e-06,
      "loss": 0.79191124,
      "num_input_tokens_seen": 94063055,
      "step": 4356,
      "time_per_iteration": 2.5422146320343018
    },
    {
      "auxiliary_loss_clip": 0.01114096,
      "auxiliary_loss_mlp": 0.01029684,
      "balance_loss_clip": 1.0449661,
      "balance_loss_mlp": 1.02291024,
      "epoch": 0.5238982745145193,
      "flos": 22491894775680.0,
      "grad_norm": 1.6404692184753504,
      "language_loss": 0.78523147,
      "learning_rate": 1.942364078612667e-06,
      "loss": 0.80666935,
      "num_input_tokens_seen": 94081785,
      "step": 4357,
      "time_per_iteration": 2.5725324153900146
    },
    {
      "auxiliary_loss_clip": 0.01136276,
      "auxiliary_loss_mlp": 0.01028908,
      "balance_loss_clip": 1.04681921,
      "balance_loss_mlp": 1.0215354,
      "epoch": 0.5240185174051584,
      "flos": 27088927234560.0,
      "grad_norm": 1.8566216966224571,
      "language_loss": 0.75223601,
      "learning_rate": 1.9415854348820765e-06,
      "loss": 0.77388787,
      "num_input_tokens_seen": 94101635,
      "step": 4358,
      "time_per_iteration": 2.597478151321411
    },
    {
      "auxiliary_loss_clip": 0.0116421,
      "auxiliary_loss_mlp": 0.01025698,
      "balance_loss_clip": 1.05225587,
      "balance_loss_mlp": 1.01831877,
      "epoch": 0.5241387602957975,
      "flos": 22674680110080.0,
      "grad_norm": 2.7837447715986094,
      "language_loss": 0.68431997,
      "learning_rate": 1.940806800012929e-06,
      "loss": 0.70621914,
      "num_input_tokens_seen": 94121705,
      "step": 4359,
      "time_per_iteration": 2.50307035446167
    },
    {
      "auxiliary_loss_clip": 0.01113817,
      "auxiliary_loss_mlp": 0.00761515,
      "balance_loss_clip": 1.04727066,
      "balance_loss_mlp": 1.00024462,
      "epoch": 0.5242590031864366,
      "flos": 40553453134080.0,
      "grad_norm": 2.3060841951438498,
      "language_loss": 0.63183236,
      "learning_rate": 1.9400281741233432e-06,
      "loss": 0.65058565,
      "num_input_tokens_seen": 94146595,
      "step": 4360,
      "time_per_iteration": 2.774470090866089
    },
    {
      "auxiliary_loss_clip": 0.01031223,
      "auxiliary_loss_mlp": 0.01002283,
      "balance_loss_clip": 1.01648259,
      "balance_loss_mlp": 1.00117993,
      "epoch": 0.5243792460770756,
      "flos": 66676313105280.0,
      "grad_norm": 0.6597339566753475,
      "language_loss": 0.52570558,
      "learning_rate": 1.939249557331435e-06,
      "loss": 0.54604065,
      "num_input_tokens_seen": 94212410,
      "step": 4361,
      "time_per_iteration": 3.1993675231933594
    },
    {
      "auxiliary_loss_clip": 0.01141556,
      "auxiliary_loss_mlp": 0.0103273,
      "balance_loss_clip": 1.04957438,
      "balance_loss_mlp": 1.02518678,
      "epoch": 0.5244994889677148,
      "flos": 28183663992960.0,
      "grad_norm": 1.9905030258088594,
      "language_loss": 0.72913051,
      "learning_rate": 1.938470949755321e-06,
      "loss": 0.75087339,
      "num_input_tokens_seen": 94232290,
      "step": 4362,
      "time_per_iteration": 2.620419979095459
    },
    {
      "auxiliary_loss_clip": 0.01036174,
      "auxiliary_loss_mlp": 0.01004584,
      "balance_loss_clip": 1.01458609,
      "balance_loss_mlp": 1.00334454,
      "epoch": 0.5246197318583539,
      "flos": 65950379239680.0,
      "grad_norm": 0.8118250140638396,
      "language_loss": 0.55707443,
      "learning_rate": 1.937692351513115e-06,
      "loss": 0.57748204,
      "num_input_tokens_seen": 94291285,
      "step": 4363,
      "time_per_iteration": 3.1110446453094482
    },
    {
      "auxiliary_loss_clip": 0.01164913,
      "auxiliary_loss_mlp": 0.01023478,
      "balance_loss_clip": 1.05045283,
      "balance_loss_mlp": 1.01557696,
      "epoch": 0.5247399747489929,
      "flos": 21033490769280.0,
      "grad_norm": 1.6967105846823882,
      "language_loss": 0.80362284,
      "learning_rate": 1.9369137627229297e-06,
      "loss": 0.82550681,
      "num_input_tokens_seen": 94309685,
      "step": 4364,
      "time_per_iteration": 2.5113067626953125
    },
    {
      "auxiliary_loss_clip": 0.01158774,
      "auxiliary_loss_mlp": 0.01026807,
      "balance_loss_clip": 1.05170727,
      "balance_loss_mlp": 1.01894546,
      "epoch": 0.5248602176396321,
      "flos": 19025940660480.0,
      "grad_norm": 1.965114968892247,
      "language_loss": 0.88237119,
      "learning_rate": 1.936135183502877e-06,
      "loss": 0.90422702,
      "num_input_tokens_seen": 94326985,
      "step": 4365,
      "time_per_iteration": 2.477025032043457
    },
    {
      "auxiliary_loss_clip": 0.01134142,
      "auxiliary_loss_mlp": 0.01034315,
      "balance_loss_clip": 1.04695344,
      "balance_loss_mlp": 1.02637267,
      "epoch": 0.5249804605302711,
      "flos": 22200084685440.0,
      "grad_norm": 2.0786733367334893,
      "language_loss": 0.80465233,
      "learning_rate": 1.935356613971066e-06,
      "loss": 0.82633698,
      "num_input_tokens_seen": 94347645,
      "step": 4366,
      "time_per_iteration": 2.576385021209717
    },
    {
      "auxiliary_loss_clip": 0.01144672,
      "auxiliary_loss_mlp": 0.00761221,
      "balance_loss_clip": 1.04838586,
      "balance_loss_mlp": 1.0002265,
      "epoch": 0.5251007034209102,
      "flos": 23805686626560.0,
      "grad_norm": 1.7116912887748048,
      "language_loss": 0.76765323,
      "learning_rate": 1.9345780542456047e-06,
      "loss": 0.78671217,
      "num_input_tokens_seen": 94367020,
      "step": 4367,
      "time_per_iteration": 2.551572561264038
    },
    {
      "auxiliary_loss_clip": 0.01150562,
      "auxiliary_loss_mlp": 0.01031839,
      "balance_loss_clip": 1.05006611,
      "balance_loss_mlp": 1.02445138,
      "epoch": 0.5252209463115494,
      "flos": 23294605962240.0,
      "grad_norm": 1.9672417099438106,
      "language_loss": 0.71493721,
      "learning_rate": 1.9337995044446007e-06,
      "loss": 0.73676121,
      "num_input_tokens_seen": 94385860,
      "step": 4368,
      "time_per_iteration": 2.5049521923065186
    },
    {
      "auxiliary_loss_clip": 0.01165455,
      "auxiliary_loss_mlp": 0.01027592,
      "balance_loss_clip": 1.05131364,
      "balance_loss_mlp": 1.0201174,
      "epoch": 0.5253411892021884,
      "flos": 19828687760640.0,
      "grad_norm": 2.7320689642143923,
      "language_loss": 0.80467808,
      "learning_rate": 1.9330209646861596e-06,
      "loss": 0.82660854,
      "num_input_tokens_seen": 94405010,
      "step": 4369,
      "time_per_iteration": 2.481038808822632
    },
    {
      "auxiliary_loss_clip": 0.01144789,
      "auxiliary_loss_mlp": 0.01026782,
      "balance_loss_clip": 1.0503459,
      "balance_loss_mlp": 1.01958776,
      "epoch": 0.5254614320928275,
      "flos": 24133730561280.0,
      "grad_norm": 1.6060303564818765,
      "language_loss": 0.77574998,
      "learning_rate": 1.9322424350883843e-06,
      "loss": 0.79746568,
      "num_input_tokens_seen": 94426845,
      "step": 4370,
      "time_per_iteration": 2.5644166469573975
    },
    {
      "auxiliary_loss_clip": 0.01149222,
      "auxiliary_loss_mlp": 0.01022335,
      "balance_loss_clip": 1.04956937,
      "balance_loss_mlp": 1.01508391,
      "epoch": 0.5255816749834666,
      "flos": 24644954880000.0,
      "grad_norm": 1.6694470456080068,
      "language_loss": 0.78758436,
      "learning_rate": 1.931463915769379e-06,
      "loss": 0.80929989,
      "num_input_tokens_seen": 94446960,
      "step": 4371,
      "time_per_iteration": 2.5521018505096436
    },
    {
      "auxiliary_loss_clip": 0.01116411,
      "auxiliary_loss_mlp": 0.01026436,
      "balance_loss_clip": 1.04487014,
      "balance_loss_mlp": 1.01933098,
      "epoch": 0.5257019178741057,
      "flos": 14136595320960.0,
      "grad_norm": 2.1373168395041033,
      "language_loss": 0.74047267,
      "learning_rate": 1.930685406847242e-06,
      "loss": 0.76190114,
      "num_input_tokens_seen": 94461535,
      "step": 4372,
      "time_per_iteration": 2.564490556716919
    },
    {
      "auxiliary_loss_clip": 0.01142901,
      "auxiliary_loss_mlp": 0.01028419,
      "balance_loss_clip": 1.04792595,
      "balance_loss_mlp": 1.02143908,
      "epoch": 0.5258221607647448,
      "flos": 23548961145600.0,
      "grad_norm": 1.636842009503532,
      "language_loss": 0.81785357,
      "learning_rate": 1.9299069084400734e-06,
      "loss": 0.83956683,
      "num_input_tokens_seen": 94482395,
      "step": 4373,
      "time_per_iteration": 2.544419288635254
    },
    {
      "auxiliary_loss_clip": 0.01130968,
      "auxiliary_loss_mlp": 0.01029965,
      "balance_loss_clip": 1.04940557,
      "balance_loss_mlp": 1.02253175,
      "epoch": 0.5259424036553839,
      "flos": 24966103403520.0,
      "grad_norm": 2.4569699442096162,
      "language_loss": 0.69965005,
      "learning_rate": 1.9291284206659717e-06,
      "loss": 0.72125936,
      "num_input_tokens_seen": 94500580,
      "step": 4374,
      "time_per_iteration": 3.4046435356140137
    },
    {
      "auxiliary_loss_clip": 0.0117571,
      "auxiliary_loss_mlp": 0.01030943,
      "balance_loss_clip": 1.05367374,
      "balance_loss_mlp": 1.02387428,
      "epoch": 0.526062646546023,
      "flos": 28763908295040.0,
      "grad_norm": 2.1000201820796813,
      "language_loss": 0.71374583,
      "learning_rate": 1.928349943643032e-06,
      "loss": 0.73581237,
      "num_input_tokens_seen": 94519680,
      "step": 4375,
      "time_per_iteration": 2.524514675140381
    },
    {
      "auxiliary_loss_clip": 0.0115899,
      "auxiliary_loss_mlp": 0.01029784,
      "balance_loss_clip": 1.05470145,
      "balance_loss_mlp": 1.02239311,
      "epoch": 0.526182889436662,
      "flos": 22821375254400.0,
      "grad_norm": 1.6615569456083212,
      "language_loss": 0.81820965,
      "learning_rate": 1.9275714774893493e-06,
      "loss": 0.84009743,
      "num_input_tokens_seen": 94539135,
      "step": 4376,
      "time_per_iteration": 2.5265250205993652
    },
    {
      "auxiliary_loss_clip": 0.0112082,
      "auxiliary_loss_mlp": 0.01037741,
      "balance_loss_clip": 1.04477191,
      "balance_loss_mlp": 1.02982569,
      "epoch": 0.5263031323273012,
      "flos": 22929466256640.0,
      "grad_norm": 2.105748643431938,
      "language_loss": 0.72955751,
      "learning_rate": 1.9267930223230154e-06,
      "loss": 0.7511431,
      "num_input_tokens_seen": 94557610,
      "step": 4377,
      "time_per_iteration": 2.5526437759399414
    },
    {
      "auxiliary_loss_clip": 0.0114894,
      "auxiliary_loss_mlp": 0.01028753,
      "balance_loss_clip": 1.05027032,
      "balance_loss_mlp": 1.0212934,
      "epoch": 0.5264233752179402,
      "flos": 17748634049280.0,
      "grad_norm": 1.9434440504980581,
      "language_loss": 0.784428,
      "learning_rate": 1.9260145782621224e-06,
      "loss": 0.80620492,
      "num_input_tokens_seen": 94575390,
      "step": 4378,
      "time_per_iteration": 3.2847650051116943
    },
    {
      "auxiliary_loss_clip": 0.01144389,
      "auxiliary_loss_mlp": 0.01032803,
      "balance_loss_clip": 1.05181515,
      "balance_loss_mlp": 1.025352,
      "epoch": 0.5265436181085793,
      "flos": 24421626069120.0,
      "grad_norm": 1.7751260769894401,
      "language_loss": 0.87915868,
      "learning_rate": 1.925236145424758e-06,
      "loss": 0.90093058,
      "num_input_tokens_seen": 94594210,
      "step": 4379,
      "time_per_iteration": 2.6108434200286865
    },
    {
      "auxiliary_loss_clip": 0.01059582,
      "auxiliary_loss_mlp": 0.01001626,
      "balance_loss_clip": 1.01755476,
      "balance_loss_mlp": 1.00032067,
      "epoch": 0.5266638609992185,
      "flos": 69207298156800.0,
      "grad_norm": 0.7023055386252935,
      "language_loss": 0.57585275,
      "learning_rate": 1.924457723929012e-06,
      "loss": 0.59646475,
      "num_input_tokens_seen": 94665020,
      "step": 4380,
      "time_per_iteration": 3.2407994270324707
    },
    {
      "auxiliary_loss_clip": 0.01161836,
      "auxiliary_loss_mlp": 0.01025211,
      "balance_loss_clip": 1.05118442,
      "balance_loss_mlp": 1.0180403,
      "epoch": 0.5267841038898575,
      "flos": 20738699850240.0,
      "grad_norm": 1.5298503049488668,
      "language_loss": 0.82867837,
      "learning_rate": 1.9236793138929685e-06,
      "loss": 0.8505488,
      "num_input_tokens_seen": 94684290,
      "step": 4381,
      "time_per_iteration": 4.029268503189087
    },
    {
      "auxiliary_loss_clip": 0.01163627,
      "auxiliary_loss_mlp": 0.01027369,
      "balance_loss_clip": 1.05062282,
      "balance_loss_mlp": 1.02000141,
      "epoch": 0.5269043467804966,
      "flos": 17234392988160.0,
      "grad_norm": 1.9798652828125796,
      "language_loss": 0.81377637,
      "learning_rate": 1.9229009154347133e-06,
      "loss": 0.83568627,
      "num_input_tokens_seen": 94701880,
      "step": 4382,
      "time_per_iteration": 2.4519734382629395
    },
    {
      "auxiliary_loss_clip": 0.01104083,
      "auxiliary_loss_mlp": 0.00760668,
      "balance_loss_clip": 1.04409552,
      "balance_loss_mlp": 1.00022781,
      "epoch": 0.5270245896711357,
      "flos": 18223157646720.0,
      "grad_norm": 2.0605447722919195,
      "language_loss": 0.80200613,
      "learning_rate": 1.922122528672327e-06,
      "loss": 0.82065368,
      "num_input_tokens_seen": 94720545,
      "step": 4383,
      "time_per_iteration": 2.574641466140747
    },
    {
      "auxiliary_loss_clip": 0.01172337,
      "auxiliary_loss_mlp": 0.0103071,
      "balance_loss_clip": 1.05177867,
      "balance_loss_mlp": 1.02368295,
      "epoch": 0.5271448325617748,
      "flos": 21287558643840.0,
      "grad_norm": 2.1082110865424704,
      "language_loss": 0.78167868,
      "learning_rate": 1.9213441537238914e-06,
      "loss": 0.80370915,
      "num_input_tokens_seen": 94737420,
      "step": 4384,
      "time_per_iteration": 2.442627429962158
    },
    {
      "auxiliary_loss_clip": 0.01016101,
      "auxiliary_loss_mlp": 0.01003069,
      "balance_loss_clip": 1.01402915,
      "balance_loss_mlp": 1.00193644,
      "epoch": 0.5272650754524139,
      "flos": 65495497403520.0,
      "grad_norm": 0.8355061940010338,
      "language_loss": 0.57331359,
      "learning_rate": 1.920565790707485e-06,
      "loss": 0.59350526,
      "num_input_tokens_seen": 94802810,
      "step": 4385,
      "time_per_iteration": 3.329414129257202
    },
    {
      "auxiliary_loss_clip": 0.01123016,
      "auxiliary_loss_mlp": 0.01028499,
      "balance_loss_clip": 1.04630411,
      "balance_loss_mlp": 1.02088177,
      "epoch": 0.527385318343053,
      "flos": 19676426008320.0,
      "grad_norm": 2.152839378443088,
      "language_loss": 0.66104859,
      "learning_rate": 1.9197874397411853e-06,
      "loss": 0.68256372,
      "num_input_tokens_seen": 94819440,
      "step": 4386,
      "time_per_iteration": 2.585705041885376
    },
    {
      "auxiliary_loss_clip": 0.0112922,
      "auxiliary_loss_mlp": 0.0103161,
      "balance_loss_clip": 1.04308963,
      "balance_loss_mlp": 1.02345622,
      "epoch": 0.5275055612336921,
      "flos": 12712018947840.0,
      "grad_norm": 2.9900514082391,
      "language_loss": 0.6748929,
      "learning_rate": 1.919009100943067e-06,
      "loss": 0.69650126,
      "num_input_tokens_seen": 94835130,
      "step": 4387,
      "time_per_iteration": 2.522519826889038
    },
    {
      "auxiliary_loss_clip": 0.01128089,
      "auxiliary_loss_mlp": 0.01026851,
      "balance_loss_clip": 1.04836178,
      "balance_loss_mlp": 1.01924229,
      "epoch": 0.5276258041243311,
      "flos": 17749029098880.0,
      "grad_norm": 2.3868783146357826,
      "language_loss": 0.65661532,
      "learning_rate": 1.9182307744312043e-06,
      "loss": 0.67816472,
      "num_input_tokens_seen": 94852235,
      "step": 4388,
      "time_per_iteration": 2.5790281295776367
    },
    {
      "auxiliary_loss_clip": 0.01149578,
      "auxiliary_loss_mlp": 0.0103169,
      "balance_loss_clip": 1.04896045,
      "balance_loss_mlp": 1.02455878,
      "epoch": 0.5277460470149702,
      "flos": 22710447077760.0,
      "grad_norm": 1.8419393749197537,
      "language_loss": 0.76744133,
      "learning_rate": 1.9174524603236676e-06,
      "loss": 0.78925407,
      "num_input_tokens_seen": 94871185,
      "step": 4389,
      "time_per_iteration": 2.538311004638672
    },
    {
      "auxiliary_loss_clip": 0.01145256,
      "auxiliary_loss_mlp": 0.01030407,
      "balance_loss_clip": 1.04928732,
      "balance_loss_mlp": 1.02230656,
      "epoch": 0.5278662899056094,
      "flos": 19902699734400.0,
      "grad_norm": 3.2793837429932085,
      "language_loss": 0.76191056,
      "learning_rate": 1.916674158738527e-06,
      "loss": 0.78366721,
      "num_input_tokens_seen": 94890090,
      "step": 4390,
      "time_per_iteration": 2.5299510955810547
    },
    {
      "auxiliary_loss_clip": 0.01128404,
      "auxiliary_loss_mlp": 0.00761792,
      "balance_loss_clip": 1.05044699,
      "balance_loss_mlp": 1.00023127,
      "epoch": 0.5279865327962484,
      "flos": 18005215875840.0,
      "grad_norm": 1.7718163330304455,
      "language_loss": 0.60480559,
      "learning_rate": 1.9158958697938506e-06,
      "loss": 0.62370759,
      "num_input_tokens_seen": 94908470,
      "step": 4391,
      "time_per_iteration": 2.529927968978882
    },
    {
      "auxiliary_loss_clip": 0.01141715,
      "auxiliary_loss_mlp": 0.01028375,
      "balance_loss_clip": 1.04901958,
      "balance_loss_mlp": 1.02050996,
      "epoch": 0.5281067756868875,
      "flos": 15924443892480.0,
      "grad_norm": 2.3830035577691557,
      "language_loss": 0.85510594,
      "learning_rate": 1.9151175936077032e-06,
      "loss": 0.87680686,
      "num_input_tokens_seen": 94923440,
      "step": 4392,
      "time_per_iteration": 2.528831958770752
    },
    {
      "auxiliary_loss_clip": 0.01154841,
      "auxiliary_loss_mlp": 0.01029287,
      "balance_loss_clip": 1.0495019,
      "balance_loss_mlp": 1.02175319,
      "epoch": 0.5282270185775266,
      "flos": 19426488197760.0,
      "grad_norm": 1.5434821597672859,
      "language_loss": 0.79221666,
      "learning_rate": 1.9143393302981507e-06,
      "loss": 0.81405795,
      "num_input_tokens_seen": 94941125,
      "step": 4393,
      "time_per_iteration": 2.4823575019836426
    },
    {
      "auxiliary_loss_clip": 0.0115038,
      "auxiliary_loss_mlp": 0.01030459,
      "balance_loss_clip": 1.05084443,
      "balance_loss_mlp": 1.02305579,
      "epoch": 0.5283472614681657,
      "flos": 16399613934720.0,
      "grad_norm": 1.751339009408559,
      "language_loss": 0.83231217,
      "learning_rate": 1.913561079983252e-06,
      "loss": 0.85412061,
      "num_input_tokens_seen": 94959950,
      "step": 4394,
      "time_per_iteration": 2.5289502143859863
    },
    {
      "auxiliary_loss_clip": 0.01145563,
      "auxiliary_loss_mlp": 0.0103548,
      "balance_loss_clip": 1.04647255,
      "balance_loss_mlp": 1.02740359,
      "epoch": 0.5284675043588047,
      "flos": 26760524163840.0,
      "grad_norm": 3.3557891984931927,
      "language_loss": 0.75133133,
      "learning_rate": 1.9127828427810693e-06,
      "loss": 0.77314174,
      "num_input_tokens_seen": 94980515,
      "step": 4395,
      "time_per_iteration": 2.5643362998962402
    },
    {
      "auxiliary_loss_clip": 0.01140194,
      "auxiliary_loss_mlp": 0.01030469,
      "balance_loss_clip": 1.04749572,
      "balance_loss_mlp": 1.02244639,
      "epoch": 0.5285877472494439,
      "flos": 19899898473600.0,
      "grad_norm": 3.0418485402451054,
      "language_loss": 0.80992818,
      "learning_rate": 1.9120046188096607e-06,
      "loss": 0.83163476,
      "num_input_tokens_seen": 94998560,
      "step": 4396,
      "time_per_iteration": 2.5681049823760986
    },
    {
      "auxiliary_loss_clip": 0.01144862,
      "auxiliary_loss_mlp": 0.0103006,
      "balance_loss_clip": 1.05195403,
      "balance_loss_mlp": 1.02271676,
      "epoch": 0.528707990140083,
      "flos": 20011257613440.0,
      "grad_norm": 2.1932694368111165,
      "language_loss": 0.74121636,
      "learning_rate": 1.9112264081870804e-06,
      "loss": 0.76296556,
      "num_input_tokens_seen": 95016950,
      "step": 4397,
      "time_per_iteration": 2.51755690574646
    },
    {
      "auxiliary_loss_clip": 0.01131207,
      "auxiliary_loss_mlp": 0.0102875,
      "balance_loss_clip": 1.05012655,
      "balance_loss_mlp": 1.02092063,
      "epoch": 0.528828233030722,
      "flos": 20667956014080.0,
      "grad_norm": 2.1024444516879237,
      "language_loss": 0.75577879,
      "learning_rate": 1.9104482110313843e-06,
      "loss": 0.77737838,
      "num_input_tokens_seen": 95036540,
      "step": 4398,
      "time_per_iteration": 2.6656744480133057
    },
    {
      "auxiliary_loss_clip": 0.01162435,
      "auxiliary_loss_mlp": 0.01027374,
      "balance_loss_clip": 1.05326867,
      "balance_loss_mlp": 1.02020359,
      "epoch": 0.5289484759213612,
      "flos": 25192448956800.0,
      "grad_norm": 1.8354255960446761,
      "language_loss": 0.74446607,
      "learning_rate": 1.909670027460623e-06,
      "loss": 0.7663641,
      "num_input_tokens_seen": 95053840,
      "step": 4399,
      "time_per_iteration": 3.256376266479492
    },
    {
      "auxiliary_loss_clip": 0.01159229,
      "auxiliary_loss_mlp": 0.0102698,
      "balance_loss_clip": 1.05163705,
      "balance_loss_mlp": 1.01967227,
      "epoch": 0.5290687188120002,
      "flos": 31139255715840.0,
      "grad_norm": 1.663156585027762,
      "language_loss": 0.71822214,
      "learning_rate": 1.908891857592847e-06,
      "loss": 0.74008423,
      "num_input_tokens_seen": 95074910,
      "step": 4400,
      "time_per_iteration": 2.5710830688476562
    },
    {
      "auxiliary_loss_clip": 0.01125132,
      "auxiliary_loss_mlp": 0.01025298,
      "balance_loss_clip": 1.04747391,
      "balance_loss_mlp": 1.01802003,
      "epoch": 0.5291889617026393,
      "flos": 20119851406080.0,
      "grad_norm": 2.2483149101764326,
      "language_loss": 0.9009974,
      "learning_rate": 1.9081137015461034e-06,
      "loss": 0.92250168,
      "num_input_tokens_seen": 95090985,
      "step": 4401,
      "time_per_iteration": 2.5242116451263428
    },
    {
      "auxiliary_loss_clip": 0.01112247,
      "auxiliary_loss_mlp": 0.01030732,
      "balance_loss_clip": 1.04861689,
      "balance_loss_mlp": 1.02292049,
      "epoch": 0.5293092045932785,
      "flos": 19643747610240.0,
      "grad_norm": 2.4552449448926343,
      "language_loss": 0.9046737,
      "learning_rate": 1.9073355594384383e-06,
      "loss": 0.92610347,
      "num_input_tokens_seen": 95109225,
      "step": 4402,
      "time_per_iteration": 2.5764408111572266
    },
    {
      "auxiliary_loss_clip": 0.01126094,
      "auxiliary_loss_mlp": 0.01034631,
      "balance_loss_clip": 1.04888248,
      "balance_loss_mlp": 1.02664399,
      "epoch": 0.5294294474839175,
      "flos": 24317736958080.0,
      "grad_norm": 2.285835116377706,
      "language_loss": 0.80538219,
      "learning_rate": 1.906557431387895e-06,
      "loss": 0.82698941,
      "num_input_tokens_seen": 95128215,
      "step": 4403,
      "time_per_iteration": 2.574779510498047
    },
    {
      "auxiliary_loss_clip": 0.01132037,
      "auxiliary_loss_mlp": 0.01030325,
      "balance_loss_clip": 1.0531019,
      "balance_loss_mlp": 1.02295518,
      "epoch": 0.5295496903745566,
      "flos": 18875941464960.0,
      "grad_norm": 2.0857706501144984,
      "language_loss": 0.78453469,
      "learning_rate": 1.905779317512516e-06,
      "loss": 0.80615824,
      "num_input_tokens_seen": 95145760,
      "step": 4404,
      "time_per_iteration": 3.3134806156158447
    },
    {
      "auxiliary_loss_clip": 0.01158816,
      "auxiliary_loss_mlp": 0.01025009,
      "balance_loss_clip": 1.0508244,
      "balance_loss_mlp": 1.01829457,
      "epoch": 0.5296699332651957,
      "flos": 20923101296640.0,
      "grad_norm": 2.6545430066243054,
      "language_loss": 0.80971444,
      "learning_rate": 1.9050012179303385e-06,
      "loss": 0.83155268,
      "num_input_tokens_seen": 95164270,
      "step": 4405,
      "time_per_iteration": 2.493962526321411
    },
    {
      "auxiliary_loss_clip": 0.01160663,
      "auxiliary_loss_mlp": 0.01028863,
      "balance_loss_clip": 1.0493691,
      "balance_loss_mlp": 1.02064085,
      "epoch": 0.5297901761558348,
      "flos": 22046745525120.0,
      "grad_norm": 2.7379806873767016,
      "language_loss": 0.69077814,
      "learning_rate": 1.904223132759401e-06,
      "loss": 0.71267343,
      "num_input_tokens_seen": 95182870,
      "step": 4406,
      "time_per_iteration": 2.4954264163970947
    },
    {
      "auxiliary_loss_clip": 0.01161765,
      "auxiliary_loss_mlp": 0.0102382,
      "balance_loss_clip": 1.05055285,
      "balance_loss_mlp": 1.01678061,
      "epoch": 0.5299104190464738,
      "flos": 21798495653760.0,
      "grad_norm": 3.7744787774356308,
      "language_loss": 0.68673629,
      "learning_rate": 1.9034450621177383e-06,
      "loss": 0.70859212,
      "num_input_tokens_seen": 95201190,
      "step": 4407,
      "time_per_iteration": 4.310419797897339
    },
    {
      "auxiliary_loss_clip": 0.01161079,
      "auxiliary_loss_mlp": 0.01030384,
      "balance_loss_clip": 1.05250859,
      "balance_loss_mlp": 1.02205729,
      "epoch": 0.530030661937113,
      "flos": 14720790119040.0,
      "grad_norm": 2.023697770998391,
      "language_loss": 0.7009033,
      "learning_rate": 1.9026670061233824e-06,
      "loss": 0.7228179,
      "num_input_tokens_seen": 95218625,
      "step": 4408,
      "time_per_iteration": 2.4622395038604736
    },
    {
      "auxiliary_loss_clip": 0.01141148,
      "auxiliary_loss_mlp": 0.01028645,
      "balance_loss_clip": 1.05028999,
      "balance_loss_mlp": 1.02127159,
      "epoch": 0.5301509048277521,
      "flos": 21251504367360.0,
      "grad_norm": 1.826996368459706,
      "language_loss": 0.80667526,
      "learning_rate": 1.901888964894365e-06,
      "loss": 0.82837313,
      "num_input_tokens_seen": 95237665,
      "step": 4409,
      "time_per_iteration": 2.5433743000030518
    },
    {
      "auxiliary_loss_clip": 0.01174771,
      "auxiliary_loss_mlp": 0.01025956,
      "balance_loss_clip": 1.05117679,
      "balance_loss_mlp": 1.01829672,
      "epoch": 0.5302711477183911,
      "flos": 25957058791680.0,
      "grad_norm": 1.7406538165760426,
      "language_loss": 0.68002915,
      "learning_rate": 1.9011109385487134e-06,
      "loss": 0.70203638,
      "num_input_tokens_seen": 95258915,
      "step": 4410,
      "time_per_iteration": 2.493377208709717
    },
    {
      "auxiliary_loss_clip": 0.01176092,
      "auxiliary_loss_mlp": 0.0102995,
      "balance_loss_clip": 1.05364108,
      "balance_loss_mlp": 1.02223718,
      "epoch": 0.5303913906090303,
      "flos": 22273126992000.0,
      "grad_norm": 3.0419574743283335,
      "language_loss": 0.66576028,
      "learning_rate": 1.900332927204454e-06,
      "loss": 0.68782073,
      "num_input_tokens_seen": 95277365,
      "step": 4411,
      "time_per_iteration": 2.462980031967163
    },
    {
      "auxiliary_loss_clip": 0.01150751,
      "auxiliary_loss_mlp": 0.01029543,
      "balance_loss_clip": 1.04956889,
      "balance_loss_mlp": 1.02147245,
      "epoch": 0.5305116334996693,
      "flos": 24936010784640.0,
      "grad_norm": 1.7779364195046625,
      "language_loss": 0.7679646,
      "learning_rate": 1.8995549309796097e-06,
      "loss": 0.7897675,
      "num_input_tokens_seen": 95296670,
      "step": 4412,
      "time_per_iteration": 2.559704303741455
    },
    {
      "auxiliary_loss_clip": 0.0116517,
      "auxiliary_loss_mlp": 0.01028355,
      "balance_loss_clip": 1.05279732,
      "balance_loss_mlp": 1.02130973,
      "epoch": 0.5306318763903084,
      "flos": 20189338266240.0,
      "grad_norm": 3.9993124443153274,
      "language_loss": 0.76926541,
      "learning_rate": 1.8987769499922028e-06,
      "loss": 0.79120064,
      "num_input_tokens_seen": 95315640,
      "step": 4413,
      "time_per_iteration": 2.4937398433685303
    },
    {
      "auxiliary_loss_clip": 0.01159266,
      "auxiliary_loss_mlp": 0.00761142,
      "balance_loss_clip": 1.05210114,
      "balance_loss_mlp": 1.00022876,
      "epoch": 0.5307521192809476,
      "flos": 20266366982400.0,
      "grad_norm": 2.148915909986806,
      "language_loss": 0.70579553,
      "learning_rate": 1.897998984360252e-06,
      "loss": 0.72499955,
      "num_input_tokens_seen": 95334610,
      "step": 4414,
      "time_per_iteration": 2.499934196472168
    },
    {
      "auxiliary_loss_clip": 0.01144433,
      "auxiliary_loss_mlp": 0.01025323,
      "balance_loss_clip": 1.05083477,
      "balance_loss_mlp": 1.01859677,
      "epoch": 0.5308723621715866,
      "flos": 28844276976000.0,
      "grad_norm": 1.4505956442454657,
      "language_loss": 0.78499252,
      "learning_rate": 1.897221034201775e-06,
      "loss": 0.80669004,
      "num_input_tokens_seen": 95358350,
      "step": 4415,
      "time_per_iteration": 2.6479642391204834
    },
    {
      "auxiliary_loss_clip": 0.01131627,
      "auxiliary_loss_mlp": 0.01026016,
      "balance_loss_clip": 1.04696047,
      "balance_loss_mlp": 1.01912534,
      "epoch": 0.5309926050622257,
      "flos": 27457766040960.0,
      "grad_norm": 1.6868037061074566,
      "language_loss": 0.66852009,
      "learning_rate": 1.8964430996347842e-06,
      "loss": 0.6900965,
      "num_input_tokens_seen": 95379900,
      "step": 4416,
      "time_per_iteration": 2.6474878787994385
    },
    {
      "auxiliary_loss_clip": 0.01143994,
      "auxiliary_loss_mlp": 0.01036256,
      "balance_loss_clip": 1.04678059,
      "balance_loss_mlp": 1.02831697,
      "epoch": 0.5311128479528648,
      "flos": 20514545026560.0,
      "grad_norm": 1.614169259339339,
      "language_loss": 0.82227421,
      "learning_rate": 1.8956651807772931e-06,
      "loss": 0.84407675,
      "num_input_tokens_seen": 95397935,
      "step": 4417,
      "time_per_iteration": 2.5351674556732178
    },
    {
      "auxiliary_loss_clip": 0.0115948,
      "auxiliary_loss_mlp": 0.01030953,
      "balance_loss_clip": 1.05384541,
      "balance_loss_mlp": 1.02360392,
      "epoch": 0.5312330908435039,
      "flos": 21397660807680.0,
      "grad_norm": 1.7481634807499669,
      "language_loss": 0.83986986,
      "learning_rate": 1.8948872777473115e-06,
      "loss": 0.86177427,
      "num_input_tokens_seen": 95415890,
      "step": 4418,
      "time_per_iteration": 2.499054431915283
    },
    {
      "auxiliary_loss_clip": 0.01146993,
      "auxiliary_loss_mlp": 0.01038079,
      "balance_loss_clip": 1.04968715,
      "balance_loss_mlp": 1.03028321,
      "epoch": 0.531353333734143,
      "flos": 24717350741760.0,
      "grad_norm": 1.6885319249249462,
      "language_loss": 0.63422692,
      "learning_rate": 1.8941093906628458e-06,
      "loss": 0.65607762,
      "num_input_tokens_seen": 95433675,
      "step": 4419,
      "time_per_iteration": 2.5424652099609375
    },
    {
      "auxiliary_loss_clip": 0.01140721,
      "auxiliary_loss_mlp": 0.01028946,
      "balance_loss_clip": 1.04687119,
      "balance_loss_mlp": 1.02145946,
      "epoch": 0.531473576624782,
      "flos": 30480689808000.0,
      "grad_norm": 1.9045932746181244,
      "language_loss": 0.70768142,
      "learning_rate": 1.893331519641902e-06,
      "loss": 0.7293781,
      "num_input_tokens_seen": 95455820,
      "step": 4420,
      "time_per_iteration": 2.6080210208892822
    },
    {
      "auxiliary_loss_clip": 0.01119281,
      "auxiliary_loss_mlp": 0.01029903,
      "balance_loss_clip": 1.04257619,
      "balance_loss_mlp": 1.02188349,
      "epoch": 0.5315938195154212,
      "flos": 23002975440000.0,
      "grad_norm": 3.6919968066614866,
      "language_loss": 0.73719394,
      "learning_rate": 1.8925536648024815e-06,
      "loss": 0.75868583,
      "num_input_tokens_seen": 95473240,
      "step": 4421,
      "time_per_iteration": 2.567322015762329
    },
    {
      "auxiliary_loss_clip": 0.01174615,
      "auxiliary_loss_mlp": 0.01027695,
      "balance_loss_clip": 1.05221045,
      "balance_loss_mlp": 1.02046537,
      "epoch": 0.5317140624060602,
      "flos": 22748584343040.0,
      "grad_norm": 1.9861729381962434,
      "language_loss": 0.75827587,
      "learning_rate": 1.8917758262625849e-06,
      "loss": 0.78029895,
      "num_input_tokens_seen": 95493480,
      "step": 4422,
      "time_per_iteration": 2.5106558799743652
    },
    {
      "auxiliary_loss_clip": 0.01142491,
      "auxiliary_loss_mlp": 0.01029121,
      "balance_loss_clip": 1.049824,
      "balance_loss_mlp": 1.02188456,
      "epoch": 0.5318343052966993,
      "flos": 22821087945600.0,
      "grad_norm": 1.6406390786960448,
      "language_loss": 0.80905747,
      "learning_rate": 1.8909980041402089e-06,
      "loss": 0.83077359,
      "num_input_tokens_seen": 95512075,
      "step": 4423,
      "time_per_iteration": 2.5378849506378174
    },
    {
      "auxiliary_loss_clip": 0.01154046,
      "auxiliary_loss_mlp": 0.01029898,
      "balance_loss_clip": 1.04880512,
      "balance_loss_mlp": 1.02189898,
      "epoch": 0.5319545481873384,
      "flos": 13626089274240.0,
      "grad_norm": 2.3868523916218294,
      "language_loss": 0.65501714,
      "learning_rate": 1.8902201985533494e-06,
      "loss": 0.67685658,
      "num_input_tokens_seen": 95529340,
      "step": 4424,
      "time_per_iteration": 2.479931116104126
    },
    {
      "auxiliary_loss_clip": 0.01146561,
      "auxiliary_loss_mlp": 0.01031817,
      "balance_loss_clip": 1.05030584,
      "balance_loss_mlp": 1.02447963,
      "epoch": 0.5320747910779775,
      "flos": 22162522037760.0,
      "grad_norm": 1.6649935899486359,
      "language_loss": 0.74840987,
      "learning_rate": 1.8894424096199983e-06,
      "loss": 0.7701937,
      "num_input_tokens_seen": 95548545,
      "step": 4425,
      "time_per_iteration": 3.25303316116333
    },
    {
      "auxiliary_loss_clip": 0.01160153,
      "auxiliary_loss_mlp": 0.01036194,
      "balance_loss_clip": 1.05159712,
      "balance_loss_mlp": 1.02851999,
      "epoch": 0.5321950339686166,
      "flos": 18588081870720.0,
      "grad_norm": 1.9135010158834054,
      "language_loss": 0.85844207,
      "learning_rate": 1.8886646374581463e-06,
      "loss": 0.88040555,
      "num_input_tokens_seen": 95567770,
      "step": 4426,
      "time_per_iteration": 2.4940221309661865
    },
    {
      "auxiliary_loss_clip": 0.01156601,
      "auxiliary_loss_mlp": 0.01032727,
      "balance_loss_clip": 1.0490334,
      "balance_loss_mlp": 1.02496672,
      "epoch": 0.5323152768592557,
      "flos": 22856818999680.0,
      "grad_norm": 2.293364493754872,
      "language_loss": 0.71158993,
      "learning_rate": 1.8878868821857795e-06,
      "loss": 0.73348325,
      "num_input_tokens_seen": 95587420,
      "step": 4427,
      "time_per_iteration": 2.4957005977630615
    },
    {
      "auxiliary_loss_clip": 0.01112921,
      "auxiliary_loss_mlp": 0.01027017,
      "balance_loss_clip": 1.04328394,
      "balance_loss_mlp": 1.01894701,
      "epoch": 0.5324355197498948,
      "flos": 33948690998400.0,
      "grad_norm": 2.3541649837753225,
      "language_loss": 0.75268912,
      "learning_rate": 1.8871091439208838e-06,
      "loss": 0.7740885,
      "num_input_tokens_seen": 95609030,
      "step": 4428,
      "time_per_iteration": 2.712125062942505
    },
    {
      "auxiliary_loss_clip": 0.011154,
      "auxiliary_loss_mlp": 0.01033328,
      "balance_loss_clip": 1.04590797,
      "balance_loss_mlp": 1.02564502,
      "epoch": 0.5325557626405338,
      "flos": 23256720092160.0,
      "grad_norm": 2.368165545659527,
      "language_loss": 0.77642465,
      "learning_rate": 1.8863314227814414e-06,
      "loss": 0.79791188,
      "num_input_tokens_seen": 95627340,
      "step": 4429,
      "time_per_iteration": 2.6075961589813232
    },
    {
      "auxiliary_loss_clip": 0.01165231,
      "auxiliary_loss_mlp": 0.01025583,
      "balance_loss_clip": 1.05224514,
      "balance_loss_mlp": 1.01789403,
      "epoch": 0.532676005531173,
      "flos": 26718687797760.0,
      "grad_norm": 2.2912272871090744,
      "language_loss": 0.48693445,
      "learning_rate": 1.8855537188854313e-06,
      "loss": 0.50884259,
      "num_input_tokens_seen": 95646315,
      "step": 4430,
      "time_per_iteration": 3.3498008251190186
    },
    {
      "auxiliary_loss_clip": 0.01160151,
      "auxiliary_loss_mlp": 0.01027839,
      "balance_loss_clip": 1.04789162,
      "balance_loss_mlp": 1.02033472,
      "epoch": 0.5327962484218121,
      "flos": 17894610921600.0,
      "grad_norm": 2.020206974813297,
      "language_loss": 0.78396046,
      "learning_rate": 1.8847760323508315e-06,
      "loss": 0.80584037,
      "num_input_tokens_seen": 95665220,
      "step": 4431,
      "time_per_iteration": 2.4697282314300537
    },
    {
      "auxiliary_loss_clip": 0.01141404,
      "auxiliary_loss_mlp": 0.01032261,
      "balance_loss_clip": 1.0496223,
      "balance_loss_mlp": 1.02447081,
      "epoch": 0.5329164913124511,
      "flos": 17925385898880.0,
      "grad_norm": 1.8023067303552525,
      "language_loss": 0.75665951,
      "learning_rate": 1.883998363295616e-06,
      "loss": 0.77839613,
      "num_input_tokens_seen": 95682700,
      "step": 4432,
      "time_per_iteration": 3.348810911178589
    },
    {
      "auxiliary_loss_clip": 0.01045385,
      "auxiliary_loss_mlp": 0.01002677,
      "balance_loss_clip": 1.01550305,
      "balance_loss_mlp": 1.00160432,
      "epoch": 0.5330367342030903,
      "flos": 57254178781440.0,
      "grad_norm": 0.9125730505056111,
      "language_loss": 0.62652493,
      "learning_rate": 1.8832207118377565e-06,
      "loss": 0.64700556,
      "num_input_tokens_seen": 95738070,
      "step": 4433,
      "time_per_iteration": 3.8186235427856445
    },
    {
      "auxiliary_loss_clip": 0.01172922,
      "auxiliary_loss_mlp": 0.01024899,
      "balance_loss_clip": 1.05188727,
      "balance_loss_mlp": 1.01766598,
      "epoch": 0.5331569770937293,
      "flos": 17420518287360.0,
      "grad_norm": 1.9228915631523582,
      "language_loss": 0.69224364,
      "learning_rate": 1.882443078095222e-06,
      "loss": 0.71422184,
      "num_input_tokens_seen": 95756950,
      "step": 4434,
      "time_per_iteration": 2.4902453422546387
    },
    {
      "auxiliary_loss_clip": 0.01029279,
      "auxiliary_loss_mlp": 0.01002644,
      "balance_loss_clip": 1.01856995,
      "balance_loss_mlp": 1.00144005,
      "epoch": 0.5332772199843684,
      "flos": 56750783627520.0,
      "grad_norm": 0.8701158485646526,
      "language_loss": 0.66783118,
      "learning_rate": 1.8816654621859794e-06,
      "loss": 0.68815041,
      "num_input_tokens_seen": 95816615,
      "step": 4435,
      "time_per_iteration": 3.097675085067749
    },
    {
      "auxiliary_loss_clip": 0.01169594,
      "auxiliary_loss_mlp": 0.0102353,
      "balance_loss_clip": 1.05089664,
      "balance_loss_mlp": 1.01584125,
      "epoch": 0.5333974628750076,
      "flos": 18697753071360.0,
      "grad_norm": 2.3394478657223168,
      "language_loss": 0.72325099,
      "learning_rate": 1.8808878642279915e-06,
      "loss": 0.74518222,
      "num_input_tokens_seen": 95832020,
      "step": 4436,
      "time_per_iteration": 2.447885036468506
    },
    {
      "auxiliary_loss_clip": 0.01130881,
      "auxiliary_loss_mlp": 0.01030151,
      "balance_loss_clip": 1.04192603,
      "balance_loss_mlp": 1.02234852,
      "epoch": 0.5335177057656466,
      "flos": 23805507058560.0,
      "grad_norm": 2.2045138127337864,
      "language_loss": 0.65519035,
      "learning_rate": 1.8801102843392209e-06,
      "loss": 0.67680073,
      "num_input_tokens_seen": 95851425,
      "step": 4437,
      "time_per_iteration": 2.59413743019104
    },
    {
      "auxiliary_loss_clip": 0.01131073,
      "auxiliary_loss_mlp": 0.01029531,
      "balance_loss_clip": 1.04727364,
      "balance_loss_mlp": 1.02203023,
      "epoch": 0.5336379486562857,
      "flos": 25078683605760.0,
      "grad_norm": 1.516046661611078,
      "language_loss": 0.85024703,
      "learning_rate": 1.8793327226376238e-06,
      "loss": 0.87185305,
      "num_input_tokens_seen": 95870745,
      "step": 4438,
      "time_per_iteration": 2.6353368759155273
    },
    {
      "auxiliary_loss_clip": 0.01150797,
      "auxiliary_loss_mlp": 0.01027784,
      "balance_loss_clip": 1.04963958,
      "balance_loss_mlp": 1.01991642,
      "epoch": 0.5337581915469248,
      "flos": 21396691140480.0,
      "grad_norm": 1.7027554336218107,
      "language_loss": 0.80287766,
      "learning_rate": 1.8785551792411569e-06,
      "loss": 0.82466352,
      "num_input_tokens_seen": 95889755,
      "step": 4439,
      "time_per_iteration": 2.544616937637329
    },
    {
      "auxiliary_loss_clip": 0.01145728,
      "auxiliary_loss_mlp": 0.01029855,
      "balance_loss_clip": 1.04948401,
      "balance_loss_mlp": 1.02278304,
      "epoch": 0.5338784344375639,
      "flos": 14865905064960.0,
      "grad_norm": 2.34673440018098,
      "language_loss": 0.82273519,
      "learning_rate": 1.8777776542677733e-06,
      "loss": 0.844491,
      "num_input_tokens_seen": 95907805,
      "step": 4440,
      "time_per_iteration": 2.5227468013763428
    },
    {
      "auxiliary_loss_clip": 0.01127763,
      "auxiliary_loss_mlp": 0.01031006,
      "balance_loss_clip": 1.04321957,
      "balance_loss_mlp": 1.02344775,
      "epoch": 0.5339986773282029,
      "flos": 20813501923200.0,
      "grad_norm": 1.8023244957121562,
      "language_loss": 0.72490841,
      "learning_rate": 1.8770001478354216e-06,
      "loss": 0.74649608,
      "num_input_tokens_seen": 95927480,
      "step": 4441,
      "time_per_iteration": 2.57855224609375
    },
    {
      "auxiliary_loss_clip": 0.01153997,
      "auxiliary_loss_mlp": 0.01032089,
      "balance_loss_clip": 1.04863811,
      "balance_loss_mlp": 1.0236187,
      "epoch": 0.5341189202188421,
      "flos": 17969089772160.0,
      "grad_norm": 2.135999440791739,
      "language_loss": 0.84045374,
      "learning_rate": 1.8762226600620504e-06,
      "loss": 0.86231458,
      "num_input_tokens_seen": 95946095,
      "step": 4442,
      "time_per_iteration": 2.489802598953247
    },
    {
      "auxiliary_loss_clip": 0.0115057,
      "auxiliary_loss_mlp": 0.01031583,
      "balance_loss_clip": 1.04802692,
      "balance_loss_mlp": 1.02331889,
      "epoch": 0.5342391631094812,
      "flos": 11031866328960.0,
      "grad_norm": 2.4171818755116226,
      "language_loss": 0.58958864,
      "learning_rate": 1.8754451910656031e-06,
      "loss": 0.61141014,
      "num_input_tokens_seen": 95959995,
      "step": 4443,
      "time_per_iteration": 2.46627140045166
    },
    {
      "auxiliary_loss_clip": 0.01124783,
      "auxiliary_loss_mlp": 0.01030408,
      "balance_loss_clip": 1.04706109,
      "balance_loss_mlp": 1.02270126,
      "epoch": 0.5343594060001202,
      "flos": 15339135772800.0,
      "grad_norm": 1.839153863723415,
      "language_loss": 0.82988954,
      "learning_rate": 1.8746677409640212e-06,
      "loss": 0.85144144,
      "num_input_tokens_seen": 95977095,
      "step": 4444,
      "time_per_iteration": 2.5980372428894043
    },
    {
      "auxiliary_loss_clip": 0.01163224,
      "auxiliary_loss_mlp": 0.01026952,
      "balance_loss_clip": 1.05195904,
      "balance_loss_mlp": 1.01920891,
      "epoch": 0.5344796488907594,
      "flos": 26900898514560.0,
      "grad_norm": 1.9269222256468725,
      "language_loss": 0.84881926,
      "learning_rate": 1.8738903098752432e-06,
      "loss": 0.8707211,
      "num_input_tokens_seen": 95996225,
      "step": 4445,
      "time_per_iteration": 2.539308547973633
    },
    {
      "auxiliary_loss_clip": 0.01146262,
      "auxiliary_loss_mlp": 0.01026366,
      "balance_loss_clip": 1.04908776,
      "balance_loss_mlp": 1.01895094,
      "epoch": 0.5345998917813984,
      "flos": 25411216740480.0,
      "grad_norm": 2.062658350395964,
      "language_loss": 0.73178345,
      "learning_rate": 1.8731128979172052e-06,
      "loss": 0.7535097,
      "num_input_tokens_seen": 96015425,
      "step": 4446,
      "time_per_iteration": 2.5703680515289307
    },
    {
      "auxiliary_loss_clip": 0.01142583,
      "auxiliary_loss_mlp": 0.0103012,
      "balance_loss_clip": 1.04870081,
      "balance_loss_mlp": 1.02289891,
      "epoch": 0.5347201346720375,
      "flos": 32853379622400.0,
      "grad_norm": 2.2962804503528385,
      "language_loss": 0.66764522,
      "learning_rate": 1.8723355052078394e-06,
      "loss": 0.6893723,
      "num_input_tokens_seen": 96035460,
      "step": 4447,
      "time_per_iteration": 2.6124227046966553
    },
    {
      "auxiliary_loss_clip": 0.01156624,
      "auxiliary_loss_mlp": 0.01034654,
      "balance_loss_clip": 1.04926443,
      "balance_loss_mlp": 1.02611268,
      "epoch": 0.5348403775626767,
      "flos": 17967940536960.0,
      "grad_norm": 2.337581893688643,
      "language_loss": 0.77254909,
      "learning_rate": 1.8715581318650765e-06,
      "loss": 0.79446185,
      "num_input_tokens_seen": 96054515,
      "step": 4448,
      "time_per_iteration": 2.500372886657715
    },
    {
      "auxiliary_loss_clip": 0.01142239,
      "auxiliary_loss_mlp": 0.01028428,
      "balance_loss_clip": 1.04916811,
      "balance_loss_mlp": 1.02041745,
      "epoch": 0.5349606204533157,
      "flos": 17603339535360.0,
      "grad_norm": 2.270237626143738,
      "language_loss": 0.82302612,
      "learning_rate": 1.8707807780068422e-06,
      "loss": 0.84473276,
      "num_input_tokens_seen": 96072330,
      "step": 4449,
      "time_per_iteration": 2.5262610912323
    },
    {
      "auxiliary_loss_clip": 0.01143574,
      "auxiliary_loss_mlp": 0.01026196,
      "balance_loss_clip": 1.04831672,
      "balance_loss_mlp": 1.0186621,
      "epoch": 0.5350808633439548,
      "flos": 29167831710720.0,
      "grad_norm": 2.5900882185817684,
      "language_loss": 0.66462624,
      "learning_rate": 1.8700034437510611e-06,
      "loss": 0.68632394,
      "num_input_tokens_seen": 96092425,
      "step": 4450,
      "time_per_iteration": 2.624983549118042
    },
    {
      "auxiliary_loss_clip": 0.01121282,
      "auxiliary_loss_mlp": 0.01027506,
      "balance_loss_clip": 1.04568756,
      "balance_loss_mlp": 1.02017725,
      "epoch": 0.5352011062345938,
      "flos": 19499997381120.0,
      "grad_norm": 2.175401096904989,
      "language_loss": 0.81153232,
      "learning_rate": 1.8692261292156549e-06,
      "loss": 0.83302015,
      "num_input_tokens_seen": 96111660,
      "step": 4451,
      "time_per_iteration": 3.3375275135040283
    },
    {
      "auxiliary_loss_clip": 0.01173132,
      "auxiliary_loss_mlp": 0.01029126,
      "balance_loss_clip": 1.05340767,
      "balance_loss_mlp": 1.0215981,
      "epoch": 0.535321349125233,
      "flos": 23477642691840.0,
      "grad_norm": 1.8408051671204206,
      "language_loss": 0.81070155,
      "learning_rate": 1.8684488345185401e-06,
      "loss": 0.83272409,
      "num_input_tokens_seen": 96131835,
      "step": 4452,
      "time_per_iteration": 2.5121216773986816
    },
    {
      "auxiliary_loss_clip": 0.01177003,
      "auxiliary_loss_mlp": 0.010317,
      "balance_loss_clip": 1.05446005,
      "balance_loss_mlp": 1.02372456,
      "epoch": 0.535441592015872,
      "flos": 20478059786880.0,
      "grad_norm": 10.988948838778407,
      "language_loss": 0.77946466,
      "learning_rate": 1.8676715597776332e-06,
      "loss": 0.8015517,
      "num_input_tokens_seen": 96150180,
      "step": 4453,
      "time_per_iteration": 2.46423602104187
    },
    {
      "auxiliary_loss_clip": 0.01106696,
      "auxiliary_loss_mlp": 0.01023304,
      "balance_loss_clip": 1.04242778,
      "balance_loss_mlp": 1.01610935,
      "epoch": 0.5355618349065111,
      "flos": 19573147428480.0,
      "grad_norm": 1.7500527988517882,
      "language_loss": 0.75889862,
      "learning_rate": 1.8668943051108455e-06,
      "loss": 0.78019857,
      "num_input_tokens_seen": 96167485,
      "step": 4454,
      "time_per_iteration": 2.5958218574523926
    },
    {
      "auxiliary_loss_clip": 0.01143839,
      "auxiliary_loss_mlp": 0.01026388,
      "balance_loss_clip": 1.04769111,
      "balance_loss_mlp": 1.01826096,
      "epoch": 0.5356820777971503,
      "flos": 24024633978240.0,
      "grad_norm": 1.765273801880341,
      "language_loss": 0.76601541,
      "learning_rate": 1.8661170706360856e-06,
      "loss": 0.7877177,
      "num_input_tokens_seen": 96186650,
      "step": 4455,
      "time_per_iteration": 2.558471918106079
    },
    {
      "auxiliary_loss_clip": 0.01161158,
      "auxiliary_loss_mlp": 0.01026036,
      "balance_loss_clip": 1.05371547,
      "balance_loss_mlp": 1.01892531,
      "epoch": 0.5358023206877893,
      "flos": 20884676722560.0,
      "grad_norm": 1.5631177381213135,
      "language_loss": 0.81655216,
      "learning_rate": 1.8653398564712594e-06,
      "loss": 0.83842409,
      "num_input_tokens_seen": 96205595,
      "step": 4456,
      "time_per_iteration": 3.293365001678467
    },
    {
      "auxiliary_loss_clip": 0.01157913,
      "auxiliary_loss_mlp": 0.01028821,
      "balance_loss_clip": 1.05089378,
      "balance_loss_mlp": 1.02162075,
      "epoch": 0.5359225635784284,
      "flos": 22418996123520.0,
      "grad_norm": 1.8996444173804905,
      "language_loss": 0.81881469,
      "learning_rate": 1.8645626627342704e-06,
      "loss": 0.84068203,
      "num_input_tokens_seen": 96226360,
      "step": 4457,
      "time_per_iteration": 2.5387778282165527
    },
    {
      "auxiliary_loss_clip": 0.01162694,
      "auxiliary_loss_mlp": 0.01030302,
      "balance_loss_clip": 1.05122495,
      "balance_loss_mlp": 1.02299714,
      "epoch": 0.5360428064690675,
      "flos": 24097784025600.0,
      "grad_norm": 2.0141314830521475,
      "language_loss": 0.80981112,
      "learning_rate": 1.8637854895430172e-06,
      "loss": 0.83174098,
      "num_input_tokens_seen": 96245625,
      "step": 4458,
      "time_per_iteration": 2.510599374771118
    },
    {
      "auxiliary_loss_clip": 0.0112025,
      "auxiliary_loss_mlp": 0.01027348,
      "balance_loss_clip": 1.04441714,
      "balance_loss_mlp": 1.01928949,
      "epoch": 0.5361630493597066,
      "flos": 21434505183360.0,
      "grad_norm": 2.1173316283715775,
      "language_loss": 0.69382256,
      "learning_rate": 1.8630083370153978e-06,
      "loss": 0.71529853,
      "num_input_tokens_seen": 96265265,
      "step": 4459,
      "time_per_iteration": 3.3858582973480225
    },
    {
      "auxiliary_loss_clip": 0.01016352,
      "auxiliary_loss_mlp": 0.01004181,
      "balance_loss_clip": 1.01470375,
      "balance_loss_mlp": 1.00307262,
      "epoch": 0.5362832922503457,
      "flos": 68888696520960.0,
      "grad_norm": 0.7451312143488024,
      "language_loss": 0.55405617,
      "learning_rate": 1.8622312052693041e-06,
      "loss": 0.57426155,
      "num_input_tokens_seen": 96326445,
      "step": 4460,
      "time_per_iteration": 3.2886176109313965
    },
    {
      "auxiliary_loss_clip": 0.01151324,
      "auxiliary_loss_mlp": 0.01033527,
      "balance_loss_clip": 1.04578388,
      "balance_loss_mlp": 1.02609479,
      "epoch": 0.5364035351409848,
      "flos": 9793702563840.0,
      "grad_norm": 2.105664831456696,
      "language_loss": 0.72145271,
      "learning_rate": 1.8614540944226267e-06,
      "loss": 0.74330115,
      "num_input_tokens_seen": 96343115,
      "step": 4461,
      "time_per_iteration": 2.5095982551574707
    },
    {
      "auxiliary_loss_clip": 0.01141333,
      "auxiliary_loss_mlp": 0.01028935,
      "balance_loss_clip": 1.05006933,
      "balance_loss_mlp": 1.02214575,
      "epoch": 0.5365237780316239,
      "flos": 23290080848640.0,
      "grad_norm": 1.704474858223676,
      "language_loss": 0.67844009,
      "learning_rate": 1.8606770045932537e-06,
      "loss": 0.70014274,
      "num_input_tokens_seen": 96362230,
      "step": 4462,
      "time_per_iteration": 2.5441324710845947
    },
    {
      "auxiliary_loss_clip": 0.01121509,
      "auxiliary_loss_mlp": 0.01035583,
      "balance_loss_clip": 1.04023874,
      "balance_loss_mlp": 1.02767348,
      "epoch": 0.5366440209222629,
      "flos": 26578133879040.0,
      "grad_norm": 1.828827218882342,
      "language_loss": 0.81735229,
      "learning_rate": 1.859899935899068e-06,
      "loss": 0.83892322,
      "num_input_tokens_seen": 96382085,
      "step": 4463,
      "time_per_iteration": 2.614748954772949
    },
    {
      "auxiliary_loss_clip": 0.01143401,
      "auxiliary_loss_mlp": 0.01030956,
      "balance_loss_clip": 1.05098569,
      "balance_loss_mlp": 1.02317762,
      "epoch": 0.5367642638129021,
      "flos": 19608052469760.0,
      "grad_norm": 1.6608918783865303,
      "language_loss": 0.78986573,
      "learning_rate": 1.8591228884579506e-06,
      "loss": 0.81160927,
      "num_input_tokens_seen": 96400580,
      "step": 4464,
      "time_per_iteration": 2.521594762802124
    },
    {
      "auxiliary_loss_clip": 0.01136067,
      "auxiliary_loss_mlp": 0.01026697,
      "balance_loss_clip": 1.04796863,
      "balance_loss_mlp": 1.01956284,
      "epoch": 0.5368845067035412,
      "flos": 23915214172800.0,
      "grad_norm": 2.2011985068275712,
      "language_loss": 0.8212477,
      "learning_rate": 1.8583458623877795e-06,
      "loss": 0.84287536,
      "num_input_tokens_seen": 96419680,
      "step": 4465,
      "time_per_iteration": 2.5981311798095703
    },
    {
      "auxiliary_loss_clip": 0.0116146,
      "auxiliary_loss_mlp": 0.01021485,
      "balance_loss_clip": 1.05128348,
      "balance_loss_mlp": 1.01420176,
      "epoch": 0.5370047495941802,
      "flos": 16873131951360.0,
      "grad_norm": 1.817968359570094,
      "language_loss": 0.74478555,
      "learning_rate": 1.8575688578064281e-06,
      "loss": 0.76661503,
      "num_input_tokens_seen": 96437805,
      "step": 4466,
      "time_per_iteration": 2.465724468231201
    },
    {
      "auxiliary_loss_clip": 0.01162055,
      "auxiliary_loss_mlp": 0.01024858,
      "balance_loss_clip": 1.05199075,
      "balance_loss_mlp": 1.01702619,
      "epoch": 0.5371249924848194,
      "flos": 20740926493440.0,
      "grad_norm": 1.709303147357437,
      "language_loss": 0.76505345,
      "learning_rate": 1.8567918748317674e-06,
      "loss": 0.78692257,
      "num_input_tokens_seen": 96457155,
      "step": 4467,
      "time_per_iteration": 2.525341749191284
    },
    {
      "auxiliary_loss_clip": 0.0113031,
      "auxiliary_loss_mlp": 0.01032194,
      "balance_loss_clip": 1.04422677,
      "balance_loss_mlp": 1.02466643,
      "epoch": 0.5372452353754584,
      "flos": 17968120104960.0,
      "grad_norm": 1.9889036520477918,
      "language_loss": 0.83060801,
      "learning_rate": 1.8560149135816659e-06,
      "loss": 0.85223299,
      "num_input_tokens_seen": 96473990,
      "step": 4468,
      "time_per_iteration": 2.5402350425720215
    },
    {
      "auxiliary_loss_clip": 0.01152784,
      "auxiliary_loss_mlp": 0.01026837,
      "balance_loss_clip": 1.04588079,
      "balance_loss_mlp": 1.01952994,
      "epoch": 0.5373654782660975,
      "flos": 15377021642880.0,
      "grad_norm": 2.3633388977720795,
      "language_loss": 0.84287512,
      "learning_rate": 1.8552379741739873e-06,
      "loss": 0.86467135,
      "num_input_tokens_seen": 96491335,
      "step": 4469,
      "time_per_iteration": 2.500734806060791
    },
    {
      "auxiliary_loss_clip": 0.0103571,
      "auxiliary_loss_mlp": 0.00751224,
      "balance_loss_clip": 1.0147953,
      "balance_loss_mlp": 1.00020075,
      "epoch": 0.5374857211567367,
      "flos": 69000091574400.0,
      "grad_norm": 0.9125018518267768,
      "language_loss": 0.55738187,
      "learning_rate": 1.8544610567265935e-06,
      "loss": 0.57525122,
      "num_input_tokens_seen": 96545275,
      "step": 4470,
      "time_per_iteration": 3.1079840660095215
    },
    {
      "auxiliary_loss_clip": 0.01147531,
      "auxiliary_loss_mlp": 0.00760696,
      "balance_loss_clip": 1.05141044,
      "balance_loss_mlp": 1.00021029,
      "epoch": 0.5376059640473757,
      "flos": 15085355207040.0,
      "grad_norm": 1.9935672688120891,
      "language_loss": 0.83279264,
      "learning_rate": 1.853684161357341e-06,
      "loss": 0.85187489,
      "num_input_tokens_seen": 96562935,
      "step": 4471,
      "time_per_iteration": 2.532824993133545
    },
    {
      "auxiliary_loss_clip": 0.01155476,
      "auxiliary_loss_mlp": 0.00761138,
      "balance_loss_clip": 1.05097783,
      "balance_loss_mlp": 1.00020647,
      "epoch": 0.5377262069380148,
      "flos": 19792597570560.0,
      "grad_norm": 2.0970012599413446,
      "language_loss": 0.76886868,
      "learning_rate": 1.852907288184085e-06,
      "loss": 0.78803486,
      "num_input_tokens_seen": 96581820,
      "step": 4472,
      "time_per_iteration": 2.529510498046875
    },
    {
      "auxiliary_loss_clip": 0.0111844,
      "auxiliary_loss_mlp": 0.01032032,
      "balance_loss_clip": 1.04508138,
      "balance_loss_mlp": 1.02409244,
      "epoch": 0.5378464498286539,
      "flos": 30003077640960.0,
      "grad_norm": 2.0251511400939806,
      "language_loss": 0.69904172,
      "learning_rate": 1.8521304373246762e-06,
      "loss": 0.72054648,
      "num_input_tokens_seen": 96602865,
      "step": 4473,
      "time_per_iteration": 2.6825625896453857
    },
    {
      "auxiliary_loss_clip": 0.01160559,
      "auxiliary_loss_mlp": 0.01037969,
      "balance_loss_clip": 1.0489161,
      "balance_loss_mlp": 1.02932012,
      "epoch": 0.537966692719293,
      "flos": 21251217058560.0,
      "grad_norm": 2.4484471137880277,
      "language_loss": 0.88913715,
      "learning_rate": 1.8513536088969626e-06,
      "loss": 0.91112244,
      "num_input_tokens_seen": 96620530,
      "step": 4474,
      "time_per_iteration": 2.5299735069274902
    },
    {
      "auxiliary_loss_clip": 0.01161013,
      "auxiliary_loss_mlp": 0.01035562,
      "balance_loss_clip": 1.05250239,
      "balance_loss_mlp": 1.02782845,
      "epoch": 0.538086935609932,
      "flos": 21543170803200.0,
      "grad_norm": 1.6210206238210152,
      "language_loss": 0.80452764,
      "learning_rate": 1.8505768030187884e-06,
      "loss": 0.82649338,
      "num_input_tokens_seen": 96640660,
      "step": 4475,
      "time_per_iteration": 2.515212297439575
    },
    {
      "auxiliary_loss_clip": 0.01142302,
      "auxiliary_loss_mlp": 0.01029956,
      "balance_loss_clip": 1.05038404,
      "balance_loss_mlp": 1.02278531,
      "epoch": 0.5382071785005712,
      "flos": 22747219626240.0,
      "grad_norm": 1.468934461139235,
      "language_loss": 0.79741752,
      "learning_rate": 1.849800019807995e-06,
      "loss": 0.81914008,
      "num_input_tokens_seen": 96661885,
      "step": 4476,
      "time_per_iteration": 3.354647159576416
    },
    {
      "auxiliary_loss_clip": 0.01130145,
      "auxiliary_loss_mlp": 0.01033572,
      "balance_loss_clip": 1.04804325,
      "balance_loss_mlp": 1.02605641,
      "epoch": 0.5383274213912103,
      "flos": 24934574240640.0,
      "grad_norm": 1.8552170874302316,
      "language_loss": 0.71007848,
      "learning_rate": 1.8490232593824186e-06,
      "loss": 0.73171568,
      "num_input_tokens_seen": 96678340,
      "step": 4477,
      "time_per_iteration": 2.5758397579193115
    },
    {
      "auxiliary_loss_clip": 0.01144866,
      "auxiliary_loss_mlp": 0.01026737,
      "balance_loss_clip": 1.05040896,
      "balance_loss_mlp": 1.01971841,
      "epoch": 0.5384476642818493,
      "flos": 22310186849280.0,
      "grad_norm": 1.6504780033505906,
      "language_loss": 0.84555912,
      "learning_rate": 1.8482465218598935e-06,
      "loss": 0.86727524,
      "num_input_tokens_seen": 96698285,
      "step": 4478,
      "time_per_iteration": 2.575528144836426
    },
    {
      "auxiliary_loss_clip": 0.01129872,
      "auxiliary_loss_mlp": 0.01025152,
      "balance_loss_clip": 1.04638457,
      "balance_loss_mlp": 1.01725447,
      "epoch": 0.5385679071724885,
      "flos": 22711021695360.0,
      "grad_norm": 1.8055930768958812,
      "language_loss": 0.83314675,
      "learning_rate": 1.8474698073582508e-06,
      "loss": 0.85469699,
      "num_input_tokens_seen": 96719655,
      "step": 4479,
      "time_per_iteration": 2.601571559906006
    },
    {
      "auxiliary_loss_clip": 0.01135968,
      "auxiliary_loss_mlp": 0.01029195,
      "balance_loss_clip": 1.04698145,
      "balance_loss_mlp": 1.02200055,
      "epoch": 0.5386881500631275,
      "flos": 15953746412160.0,
      "grad_norm": 2.141489712044353,
      "language_loss": 0.87691998,
      "learning_rate": 1.8466931159953166e-06,
      "loss": 0.89857161,
      "num_input_tokens_seen": 96736290,
      "step": 4480,
      "time_per_iteration": 2.5692572593688965
    },
    {
      "auxiliary_loss_clip": 0.0114809,
      "auxiliary_loss_mlp": 0.0102632,
      "balance_loss_clip": 1.0509088,
      "balance_loss_mlp": 1.01818979,
      "epoch": 0.5388083929537666,
      "flos": 24060041809920.0,
      "grad_norm": 1.8220826673824824,
      "language_loss": 0.83873266,
      "learning_rate": 1.8459164478889158e-06,
      "loss": 0.86047673,
      "num_input_tokens_seen": 96757685,
      "step": 4481,
      "time_per_iteration": 2.5797581672668457
    },
    {
      "auxiliary_loss_clip": 0.01124031,
      "auxiliary_loss_mlp": 0.01027651,
      "balance_loss_clip": 1.04424179,
      "balance_loss_mlp": 1.0200603,
      "epoch": 0.5389286358444056,
      "flos": 22236893147520.0,
      "grad_norm": 1.7677118968228311,
      "language_loss": 0.75871253,
      "learning_rate": 1.8451398031568663e-06,
      "loss": 0.78022933,
      "num_input_tokens_seen": 96777310,
      "step": 4482,
      "time_per_iteration": 3.380509376525879
    },
    {
      "auxiliary_loss_clip": 0.01130063,
      "auxiliary_loss_mlp": 0.01031211,
      "balance_loss_clip": 1.04836655,
      "balance_loss_mlp": 1.02382874,
      "epoch": 0.5390488787350448,
      "flos": 24281718595200.0,
      "grad_norm": 2.5652906914081544,
      "language_loss": 0.74674416,
      "learning_rate": 1.844363181916986e-06,
      "loss": 0.76835692,
      "num_input_tokens_seen": 96798035,
      "step": 4483,
      "time_per_iteration": 2.6464931964874268
    },
    {
      "auxiliary_loss_clip": 0.01155997,
      "auxiliary_loss_mlp": 0.01026132,
      "balance_loss_clip": 1.05019522,
      "balance_loss_mlp": 1.01778698,
      "epoch": 0.5391691216256839,
      "flos": 16581393688320.0,
      "grad_norm": 1.6962975320754579,
      "language_loss": 0.83378923,
      "learning_rate": 1.8435865842870868e-06,
      "loss": 0.85561049,
      "num_input_tokens_seen": 96815975,
      "step": 4484,
      "time_per_iteration": 3.2903785705566406
    },
    {
      "auxiliary_loss_clip": 0.01136349,
      "auxiliary_loss_mlp": 0.00761666,
      "balance_loss_clip": 1.04597604,
      "balance_loss_mlp": 1.00019944,
      "epoch": 0.5392893645163229,
      "flos": 23330049707520.0,
      "grad_norm": 1.85705123718222,
      "language_loss": 0.71888125,
      "learning_rate": 1.8428100103849787e-06,
      "loss": 0.73786139,
      "num_input_tokens_seen": 96835770,
      "step": 4485,
      "time_per_iteration": 3.3344228267669678
    },
    {
      "auxiliary_loss_clip": 0.01144678,
      "auxiliary_loss_mlp": 0.01028024,
      "balance_loss_clip": 1.05186725,
      "balance_loss_mlp": 1.02066851,
      "epoch": 0.5394096074069621,
      "flos": 15669801400320.0,
      "grad_norm": 2.135677202505748,
      "language_loss": 0.72932041,
      "learning_rate": 1.842033460328467e-06,
      "loss": 0.75104737,
      "num_input_tokens_seen": 96854490,
      "step": 4486,
      "time_per_iteration": 2.510364294052124
    },
    {
      "auxiliary_loss_clip": 0.01147514,
      "auxiliary_loss_mlp": 0.00760921,
      "balance_loss_clip": 1.04751956,
      "balance_loss_mlp": 1.00020921,
      "epoch": 0.5395298502976011,
      "flos": 22893447893760.0,
      "grad_norm": 1.803426852019116,
      "language_loss": 0.75032246,
      "learning_rate": 1.8412569342353541e-06,
      "loss": 0.7694068,
      "num_input_tokens_seen": 96874645,
      "step": 4487,
      "time_per_iteration": 2.584289073944092
    },
    {
      "auxiliary_loss_clip": 0.01154079,
      "auxiliary_loss_mlp": 0.01037151,
      "balance_loss_clip": 1.05457401,
      "balance_loss_mlp": 1.02893174,
      "epoch": 0.5396500931882402,
      "flos": 23842135952640.0,
      "grad_norm": 1.9759234765766658,
      "language_loss": 0.84889162,
      "learning_rate": 1.840480432223438e-06,
      "loss": 0.87080383,
      "num_input_tokens_seen": 96893650,
      "step": 4488,
      "time_per_iteration": 2.5503032207489014
    },
    {
      "auxiliary_loss_clip": 0.01149372,
      "auxiliary_loss_mlp": 0.01028147,
      "balance_loss_clip": 1.04876661,
      "balance_loss_mlp": 1.0212599,
      "epoch": 0.5397703360788794,
      "flos": 26322988596480.0,
      "grad_norm": 3.709495855450797,
      "language_loss": 0.77414691,
      "learning_rate": 1.8397039544105131e-06,
      "loss": 0.7959221,
      "num_input_tokens_seen": 96912735,
      "step": 4489,
      "time_per_iteration": 2.590378761291504
    },
    {
      "auxiliary_loss_clip": 0.01137775,
      "auxiliary_loss_mlp": 0.01025817,
      "balance_loss_clip": 1.04409647,
      "balance_loss_mlp": 1.01784766,
      "epoch": 0.5398905789695184,
      "flos": 21214588164480.0,
      "grad_norm": 1.982253255251768,
      "language_loss": 0.69592452,
      "learning_rate": 1.8389275009143711e-06,
      "loss": 0.71756041,
      "num_input_tokens_seen": 96932475,
      "step": 4490,
      "time_per_iteration": 2.6660993099212646
    },
    {
      "auxiliary_loss_clip": 0.01173792,
      "auxiliary_loss_mlp": 0.01026089,
      "balance_loss_clip": 1.05329561,
      "balance_loss_mlp": 1.01897526,
      "epoch": 0.5400108218601575,
      "flos": 25080335631360.0,
      "grad_norm": 1.7354977439049641,
      "language_loss": 0.73378897,
      "learning_rate": 1.8381510718527988e-06,
      "loss": 0.75578773,
      "num_input_tokens_seen": 96952085,
      "step": 4491,
      "time_per_iteration": 2.524327516555786
    },
    {
      "auxiliary_loss_clip": 0.01145586,
      "auxiliary_loss_mlp": 0.01030434,
      "balance_loss_clip": 1.04594111,
      "balance_loss_mlp": 1.02248633,
      "epoch": 0.5401310647507966,
      "flos": 26357498588160.0,
      "grad_norm": 1.927023761829669,
      "language_loss": 0.63551515,
      "learning_rate": 1.8373746673435812e-06,
      "loss": 0.65727532,
      "num_input_tokens_seen": 96973110,
      "step": 4492,
      "time_per_iteration": 2.5774412155151367
    },
    {
      "auxiliary_loss_clip": 0.01174717,
      "auxiliary_loss_mlp": 0.01029205,
      "balance_loss_clip": 1.05405939,
      "balance_loss_mlp": 1.02153063,
      "epoch": 0.5402513076414357,
      "flos": 27855332749440.0,
      "grad_norm": 1.6486967081605235,
      "language_loss": 0.78771687,
      "learning_rate": 1.8365982875044964e-06,
      "loss": 0.80975604,
      "num_input_tokens_seen": 96993420,
      "step": 4493,
      "time_per_iteration": 2.523552656173706
    },
    {
      "auxiliary_loss_clip": 0.01164604,
      "auxiliary_loss_mlp": 0.00761842,
      "balance_loss_clip": 1.05187476,
      "balance_loss_mlp": 1.00020528,
      "epoch": 0.5403715505320748,
      "flos": 22893771116160.0,
      "grad_norm": 2.275372689746067,
      "language_loss": 0.76386875,
      "learning_rate": 1.8358219324533217e-06,
      "loss": 0.78313321,
      "num_input_tokens_seen": 97013685,
      "step": 4494,
      "time_per_iteration": 2.524921417236328
    },
    {
      "auxiliary_loss_clip": 0.01142394,
      "auxiliary_loss_mlp": 0.01025541,
      "balance_loss_clip": 1.04801798,
      "balance_loss_mlp": 1.01871896,
      "epoch": 0.5404917934227139,
      "flos": 30224143895040.0,
      "grad_norm": 1.745199544479273,
      "language_loss": 0.70424396,
      "learning_rate": 1.8350456023078292e-06,
      "loss": 0.7259233,
      "num_input_tokens_seen": 97036060,
      "step": 4495,
      "time_per_iteration": 2.6190342903137207
    },
    {
      "auxiliary_loss_clip": 0.01176764,
      "auxiliary_loss_mlp": 0.01035194,
      "balance_loss_clip": 1.05210423,
      "balance_loss_mlp": 1.02686083,
      "epoch": 0.540612036313353,
      "flos": 19938502615680.0,
      "grad_norm": 2.0953170076416643,
      "language_loss": 0.78009176,
      "learning_rate": 1.8342692971857874e-06,
      "loss": 0.80221134,
      "num_input_tokens_seen": 97055260,
      "step": 4496,
      "time_per_iteration": 2.477479934692383
    },
    {
      "auxiliary_loss_clip": 0.01143397,
      "auxiliary_loss_mlp": 0.01024081,
      "balance_loss_clip": 1.04975176,
      "balance_loss_mlp": 1.0165056,
      "epoch": 0.540732279203992,
      "flos": 24279599692800.0,
      "grad_norm": 2.3060717890601996,
      "language_loss": 0.71488214,
      "learning_rate": 1.833493017204962e-06,
      "loss": 0.73655701,
      "num_input_tokens_seen": 97075365,
      "step": 4497,
      "time_per_iteration": 2.556281566619873
    },
    {
      "auxiliary_loss_clip": 0.01172283,
      "auxiliary_loss_mlp": 0.01027442,
      "balance_loss_clip": 1.05061162,
      "balance_loss_mlp": 1.01999462,
      "epoch": 0.5408525220946312,
      "flos": 20193216935040.0,
      "grad_norm": 1.8808488507904944,
      "language_loss": 0.7801308,
      "learning_rate": 1.8327167624831134e-06,
      "loss": 0.80212808,
      "num_input_tokens_seen": 97093095,
      "step": 4498,
      "time_per_iteration": 2.4964852333068848
    },
    {
      "auxiliary_loss_clip": 0.01172707,
      "auxiliary_loss_mlp": 0.01028641,
      "balance_loss_clip": 1.05267334,
      "balance_loss_mlp": 1.0211935,
      "epoch": 0.5409727649852702,
      "flos": 24134448833280.0,
      "grad_norm": 1.7847695363344214,
      "language_loss": 0.70628053,
      "learning_rate": 1.831940533137999e-06,
      "loss": 0.72829401,
      "num_input_tokens_seen": 97112000,
      "step": 4499,
      "time_per_iteration": 2.477630615234375
    },
    {
      "auxiliary_loss_clip": 0.01158061,
      "auxiliary_loss_mlp": 0.0102923,
      "balance_loss_clip": 1.05260396,
      "balance_loss_mlp": 1.02188981,
      "epoch": 0.5410930078759093,
      "flos": 23912700220800.0,
      "grad_norm": 1.7170739771960104,
      "language_loss": 0.72135407,
      "learning_rate": 1.8311643292873718e-06,
      "loss": 0.74322701,
      "num_input_tokens_seen": 97130820,
      "step": 4500,
      "time_per_iteration": 2.5614125728607178
    },
    {
      "auxiliary_loss_clip": 0.01157025,
      "auxiliary_loss_mlp": 0.01030275,
      "balance_loss_clip": 1.05044675,
      "balance_loss_mlp": 1.02302086,
      "epoch": 0.5412132507665485,
      "flos": 21105132445440.0,
      "grad_norm": 1.8136881216515928,
      "language_loss": 0.87830108,
      "learning_rate": 1.8303881510489818e-06,
      "loss": 0.90017408,
      "num_input_tokens_seen": 97149210,
      "step": 4501,
      "time_per_iteration": 2.495850086212158
    },
    {
      "auxiliary_loss_clip": 0.01148965,
      "auxiliary_loss_mlp": 0.0103366,
      "balance_loss_clip": 1.05043435,
      "balance_loss_mlp": 1.02622414,
      "epoch": 0.5413334936571875,
      "flos": 30227340205440.0,
      "grad_norm": 1.9684557899779136,
      "language_loss": 0.69585156,
      "learning_rate": 1.829611998540574e-06,
      "loss": 0.71767783,
      "num_input_tokens_seen": 97170415,
      "step": 4502,
      "time_per_iteration": 3.368846893310547
    },
    {
      "auxiliary_loss_clip": 0.01157482,
      "auxiliary_loss_mlp": 0.00761328,
      "balance_loss_clip": 1.04896832,
      "balance_loss_mlp": 1.00022316,
      "epoch": 0.5414537365478266,
      "flos": 24279635606400.0,
      "grad_norm": 1.6846173387796528,
      "language_loss": 0.8000071,
      "learning_rate": 1.8288358718798914e-06,
      "loss": 0.81919521,
      "num_input_tokens_seen": 97189605,
      "step": 4503,
      "time_per_iteration": 2.5297632217407227
    },
    {
      "auxiliary_loss_clip": 0.01155633,
      "auxiliary_loss_mlp": 0.0076106,
      "balance_loss_clip": 1.05196702,
      "balance_loss_mlp": 1.00023913,
      "epoch": 0.5415739794384657,
      "flos": 16654543735680.0,
      "grad_norm": 2.2871008551117504,
      "language_loss": 0.7251265,
      "learning_rate": 1.8280597711846703e-06,
      "loss": 0.74429345,
      "num_input_tokens_seen": 97207845,
      "step": 4504,
      "time_per_iteration": 2.4851527214050293
    },
    {
      "auxiliary_loss_clip": 0.01154776,
      "auxiliary_loss_mlp": 0.01034451,
      "balance_loss_clip": 1.05053353,
      "balance_loss_mlp": 1.02679777,
      "epoch": 0.5416942223291048,
      "flos": 23185724860800.0,
      "grad_norm": 1.8814552803251026,
      "language_loss": 0.83340228,
      "learning_rate": 1.8272836965726455e-06,
      "loss": 0.85529459,
      "num_input_tokens_seen": 97226780,
      "step": 4505,
      "time_per_iteration": 2.5215506553649902
    },
    {
      "auxiliary_loss_clip": 0.01100691,
      "auxiliary_loss_mlp": 0.01029298,
      "balance_loss_clip": 1.04120219,
      "balance_loss_mlp": 1.02141821,
      "epoch": 0.5418144652197439,
      "flos": 20303247271680.0,
      "grad_norm": 1.7301348795031142,
      "language_loss": 0.78312564,
      "learning_rate": 1.8265076481615461e-06,
      "loss": 0.80442548,
      "num_input_tokens_seen": 97246695,
      "step": 4506,
      "time_per_iteration": 2.6491940021514893
    },
    {
      "auxiliary_loss_clip": 0.01146621,
      "auxiliary_loss_mlp": 0.01034919,
      "balance_loss_clip": 1.05332005,
      "balance_loss_mlp": 1.02703333,
      "epoch": 0.541934708110383,
      "flos": 12458633431680.0,
      "grad_norm": 2.1755207207520444,
      "language_loss": 0.87128305,
      "learning_rate": 1.8257316260690987e-06,
      "loss": 0.89309841,
      "num_input_tokens_seen": 97264480,
      "step": 4507,
      "time_per_iteration": 2.508147954940796
    },
    {
      "auxiliary_loss_clip": 0.01156727,
      "auxiliary_loss_mlp": 0.0102686,
      "balance_loss_clip": 1.04938114,
      "balance_loss_mlp": 1.01974022,
      "epoch": 0.5420549510010221,
      "flos": 21253802837760.0,
      "grad_norm": 1.5970420017082163,
      "language_loss": 0.75961983,
      "learning_rate": 1.8249556304130254e-06,
      "loss": 0.78145564,
      "num_input_tokens_seen": 97285760,
      "step": 4508,
      "time_per_iteration": 3.299811840057373
    },
    {
      "auxiliary_loss_clip": 0.01133842,
      "auxiliary_loss_mlp": 0.01028509,
      "balance_loss_clip": 1.04428852,
      "balance_loss_mlp": 1.02056098,
      "epoch": 0.5421751938916611,
      "flos": 29490524519040.0,
      "grad_norm": 2.0632923057076344,
      "language_loss": 0.68370616,
      "learning_rate": 1.824179661311044e-06,
      "loss": 0.70532972,
      "num_input_tokens_seen": 97304510,
      "step": 4509,
      "time_per_iteration": 2.5958242416381836
    },
    {
      "auxiliary_loss_clip": 0.01115107,
      "auxiliary_loss_mlp": 0.01029668,
      "balance_loss_clip": 1.04215336,
      "balance_loss_mlp": 1.02261996,
      "epoch": 0.5422954367823003,
      "flos": 18734238311040.0,
      "grad_norm": 1.803659263850139,
      "language_loss": 0.79886162,
      "learning_rate": 1.823403718880868e-06,
      "loss": 0.8203094,
      "num_input_tokens_seen": 97323270,
      "step": 4510,
      "time_per_iteration": 3.4088265895843506
    },
    {
      "auxiliary_loss_clip": 0.01143507,
      "auxiliary_loss_mlp": 0.01031541,
      "balance_loss_clip": 1.04439902,
      "balance_loss_mlp": 1.02410853,
      "epoch": 0.5424156796729394,
      "flos": 39969006940800.0,
      "grad_norm": 1.781754194865738,
      "language_loss": 0.66485459,
      "learning_rate": 1.822627803240207e-06,
      "loss": 0.6866051,
      "num_input_tokens_seen": 97345600,
      "step": 4511,
      "time_per_iteration": 3.419186592102051
    },
    {
      "auxiliary_loss_clip": 0.01135781,
      "auxiliary_loss_mlp": 0.01027568,
      "balance_loss_clip": 1.04751039,
      "balance_loss_mlp": 1.02036488,
      "epoch": 0.5425359225635784,
      "flos": 11546538353280.0,
      "grad_norm": 1.9982019981274868,
      "language_loss": 0.85235786,
      "learning_rate": 1.8218519145067675e-06,
      "loss": 0.87399137,
      "num_input_tokens_seen": 97361220,
      "step": 4512,
      "time_per_iteration": 2.554074287414551
    },
    {
      "auxiliary_loss_clip": 0.01124083,
      "auxiliary_loss_mlp": 0.01033454,
      "balance_loss_clip": 1.04473352,
      "balance_loss_mlp": 1.02609575,
      "epoch": 0.5426561654542175,
      "flos": 20229702174720.0,
      "grad_norm": 1.772711627749825,
      "language_loss": 0.89455473,
      "learning_rate": 1.8210760527982508e-06,
      "loss": 0.91613013,
      "num_input_tokens_seen": 97381505,
      "step": 4513,
      "time_per_iteration": 2.5712273120880127
    },
    {
      "auxiliary_loss_clip": 0.01148283,
      "auxiliary_loss_mlp": 0.00760623,
      "balance_loss_clip": 1.05099165,
      "balance_loss_mlp": 1.000211,
      "epoch": 0.5427764083448566,
      "flos": 21871681614720.0,
      "grad_norm": 1.8169514762452057,
      "language_loss": 0.74991077,
      "learning_rate": 1.8203002182323552e-06,
      "loss": 0.76899981,
      "num_input_tokens_seen": 97399060,
      "step": 4514,
      "time_per_iteration": 2.5578527450561523
    },
    {
      "auxiliary_loss_clip": 0.01144746,
      "auxiliary_loss_mlp": 0.01031498,
      "balance_loss_clip": 1.04771018,
      "balance_loss_mlp": 1.02454519,
      "epoch": 0.5428966512354957,
      "flos": 19640946349440.0,
      "grad_norm": 1.8161369424481932,
      "language_loss": 0.75657761,
      "learning_rate": 1.819524410926773e-06,
      "loss": 0.77834004,
      "num_input_tokens_seen": 97416740,
      "step": 4515,
      "time_per_iteration": 2.5369153022766113
    },
    {
      "auxiliary_loss_clip": 0.01103555,
      "auxiliary_loss_mlp": 0.01030787,
      "balance_loss_clip": 1.04614091,
      "balance_loss_mlp": 1.02326441,
      "epoch": 0.5430168941261347,
      "flos": 22382187661440.0,
      "grad_norm": 1.420428915875673,
      "language_loss": 0.77123809,
      "learning_rate": 1.8187486309991944e-06,
      "loss": 0.7925815,
      "num_input_tokens_seen": 97437620,
      "step": 4516,
      "time_per_iteration": 2.627680778503418
    },
    {
      "auxiliary_loss_clip": 0.01162488,
      "auxiliary_loss_mlp": 0.01030583,
      "balance_loss_clip": 1.05113792,
      "balance_loss_mlp": 1.02363896,
      "epoch": 0.5431371370167739,
      "flos": 18764187275520.0,
      "grad_norm": 1.8368545068615874,
      "language_loss": 0.77168787,
      "learning_rate": 1.817972878567304e-06,
      "loss": 0.79361856,
      "num_input_tokens_seen": 97456275,
      "step": 4517,
      "time_per_iteration": 2.5082075595855713
    },
    {
      "auxiliary_loss_clip": 0.01149986,
      "auxiliary_loss_mlp": 0.01026037,
      "balance_loss_clip": 1.04822576,
      "balance_loss_mlp": 1.01847291,
      "epoch": 0.543257379907413,
      "flos": 18806023641600.0,
      "grad_norm": 1.8084849392479445,
      "language_loss": 0.76467508,
      "learning_rate": 1.8171971537487834e-06,
      "loss": 0.78643531,
      "num_input_tokens_seen": 97474925,
      "step": 4518,
      "time_per_iteration": 2.5497963428497314
    },
    {
      "auxiliary_loss_clip": 0.01175975,
      "auxiliary_loss_mlp": 0.01033807,
      "balance_loss_clip": 1.05270016,
      "balance_loss_mlp": 1.0256803,
      "epoch": 0.543377622798052,
      "flos": 17493381025920.0,
      "grad_norm": 1.9440421124742004,
      "language_loss": 0.81079578,
      "learning_rate": 1.8164214566613093e-06,
      "loss": 0.83289361,
      "num_input_tokens_seen": 97493550,
      "step": 4519,
      "time_per_iteration": 2.4588522911071777
    },
    {
      "auxiliary_loss_clip": 0.01170882,
      "auxiliary_loss_mlp": 0.01029211,
      "balance_loss_clip": 1.05079317,
      "balance_loss_mlp": 1.02199888,
      "epoch": 0.5434978656886912,
      "flos": 18989311766400.0,
      "grad_norm": 3.7631665621295713,
      "language_loss": 0.65725094,
      "learning_rate": 1.8156457874225547e-06,
      "loss": 0.67925185,
      "num_input_tokens_seen": 97512010,
      "step": 4520,
      "time_per_iteration": 2.4552416801452637
    },
    {
      "auxiliary_loss_clip": 0.01139593,
      "auxiliary_loss_mlp": 0.01027861,
      "balance_loss_clip": 1.05067205,
      "balance_loss_mlp": 1.02007651,
      "epoch": 0.5436181085793302,
      "flos": 17274936464640.0,
      "grad_norm": 1.7592235726309076,
      "language_loss": 0.80290997,
      "learning_rate": 1.814870146150187e-06,
      "loss": 0.82458448,
      "num_input_tokens_seen": 97530120,
      "step": 4521,
      "time_per_iteration": 2.5324466228485107
    },
    {
      "auxiliary_loss_clip": 0.01150186,
      "auxiliary_loss_mlp": 0.01033101,
      "balance_loss_clip": 1.04792738,
      "balance_loss_mlp": 1.02452993,
      "epoch": 0.5437383514699693,
      "flos": 19098587917440.0,
      "grad_norm": 1.9239791888381743,
      "language_loss": 0.78686953,
      "learning_rate": 1.814094532961871e-06,
      "loss": 0.80870235,
      "num_input_tokens_seen": 97548695,
      "step": 4522,
      "time_per_iteration": 2.5349996089935303
    },
    {
      "auxiliary_loss_clip": 0.01118216,
      "auxiliary_loss_mlp": 0.01027373,
      "balance_loss_clip": 1.04298949,
      "balance_loss_mlp": 1.02029848,
      "epoch": 0.5438585943606085,
      "flos": 22602715211520.0,
      "grad_norm": 1.9188857387357585,
      "language_loss": 0.83647019,
      "learning_rate": 1.8133189479752666e-06,
      "loss": 0.85792607,
      "num_input_tokens_seen": 97567625,
      "step": 4523,
      "time_per_iteration": 2.6168689727783203
    },
    {
      "auxiliary_loss_clip": 0.01172812,
      "auxiliary_loss_mlp": 0.01026924,
      "balance_loss_clip": 1.05243993,
      "balance_loss_mlp": 1.01946712,
      "epoch": 0.5439788372512475,
      "flos": 21798495653760.0,
      "grad_norm": 1.8368094410215154,
      "language_loss": 0.81802088,
      "learning_rate": 1.8125433913080292e-06,
      "loss": 0.84001827,
      "num_input_tokens_seen": 97585325,
      "step": 4524,
      "time_per_iteration": 2.4799692630767822
    },
    {
      "auxiliary_loss_clip": 0.01048274,
      "auxiliary_loss_mlp": 0.01032826,
      "balance_loss_clip": 1.03594112,
      "balance_loss_mlp": 1.02575374,
      "epoch": 0.5440990801418866,
      "flos": 16399362539520.0,
      "grad_norm": 2.6994370033089234,
      "language_loss": 0.82791448,
      "learning_rate": 1.811767863077811e-06,
      "loss": 0.8487255,
      "num_input_tokens_seen": 97604275,
      "step": 4525,
      "time_per_iteration": 2.914135694503784
    },
    {
      "auxiliary_loss_clip": 0.01094441,
      "auxiliary_loss_mlp": 0.01026666,
      "balance_loss_clip": 1.04537809,
      "balance_loss_mlp": 1.01935887,
      "epoch": 0.5442193230325257,
      "flos": 21615638492160.0,
      "grad_norm": 1.656496293934783,
      "language_loss": 0.78256232,
      "learning_rate": 1.8109923634022577e-06,
      "loss": 0.8037734,
      "num_input_tokens_seen": 97624300,
      "step": 4526,
      "time_per_iteration": 2.8230338096618652
    },
    {
      "auxiliary_loss_clip": 0.01176438,
      "auxiliary_loss_mlp": 0.01028634,
      "balance_loss_clip": 1.05319309,
      "balance_loss_mlp": 1.02120709,
      "epoch": 0.5443395659231648,
      "flos": 15481198062720.0,
      "grad_norm": 2.125115692683203,
      "language_loss": 0.86672091,
      "learning_rate": 1.8102168923990128e-06,
      "loss": 0.88877165,
      "num_input_tokens_seen": 97637845,
      "step": 4527,
      "time_per_iteration": 2.4410343170166016
    },
    {
      "auxiliary_loss_clip": 0.01160525,
      "auxiliary_loss_mlp": 0.00761243,
      "balance_loss_clip": 1.05057931,
      "balance_loss_mlp": 1.00024617,
      "epoch": 0.5444598088138038,
      "flos": 18770436241920.0,
      "grad_norm": 1.7039613556941138,
      "language_loss": 0.79923749,
      "learning_rate": 1.809441450185714e-06,
      "loss": 0.81845516,
      "num_input_tokens_seen": 97656330,
      "step": 4528,
      "time_per_iteration": 3.453847885131836
    },
    {
      "auxiliary_loss_clip": 0.01146962,
      "auxiliary_loss_mlp": 0.01030389,
      "balance_loss_clip": 1.04405475,
      "balance_loss_mlp": 1.02277184,
      "epoch": 0.544580051704443,
      "flos": 21142335957120.0,
      "grad_norm": 1.999675578149202,
      "language_loss": 0.73321521,
      "learning_rate": 1.8086660368799958e-06,
      "loss": 0.75498873,
      "num_input_tokens_seen": 97674380,
      "step": 4529,
      "time_per_iteration": 2.5472123622894287
    },
    {
      "auxiliary_loss_clip": 0.01145623,
      "auxiliary_loss_mlp": 0.0102803,
      "balance_loss_clip": 1.04833508,
      "balance_loss_mlp": 1.02041864,
      "epoch": 0.5447002945950821,
      "flos": 32491508054400.0,
      "grad_norm": 3.1535506303916554,
      "language_loss": 0.77384734,
      "learning_rate": 1.807890652599488e-06,
      "loss": 0.79558384,
      "num_input_tokens_seen": 97698765,
      "step": 4530,
      "time_per_iteration": 2.632446527481079
    },
    {
      "auxiliary_loss_clip": 0.01169974,
      "auxiliary_loss_mlp": 0.01024509,
      "balance_loss_clip": 1.05073071,
      "balance_loss_mlp": 1.01796484,
      "epoch": 0.5448205374857211,
      "flos": 11798307757440.0,
      "grad_norm": 2.2201867164719298,
      "language_loss": 0.82376647,
      "learning_rate": 1.8071152974618156e-06,
      "loss": 0.84571135,
      "num_input_tokens_seen": 97716565,
      "step": 4531,
      "time_per_iteration": 2.444945812225342
    },
    {
      "auxiliary_loss_clip": 0.01131397,
      "auxiliary_loss_mlp": 0.00760998,
      "balance_loss_clip": 1.04531264,
      "balance_loss_mlp": 1.0001967,
      "epoch": 0.5449407803763603,
      "flos": 24133766474880.0,
      "grad_norm": 2.3115406773166693,
      "language_loss": 0.7847929,
      "learning_rate": 1.806339971584599e-06,
      "loss": 0.80371684,
      "num_input_tokens_seen": 97733225,
      "step": 4532,
      "time_per_iteration": 2.615262985229492
    },
    {
      "auxiliary_loss_clip": 0.01174283,
      "auxiliary_loss_mlp": 0.01027129,
      "balance_loss_clip": 1.05225706,
      "balance_loss_mlp": 1.02025354,
      "epoch": 0.5450610232669993,
      "flos": 23258551685760.0,
      "grad_norm": 2.0105623028128727,
      "language_loss": 0.85277081,
      "learning_rate": 1.8055646750854546e-06,
      "loss": 0.87478495,
      "num_input_tokens_seen": 97752735,
      "step": 4533,
      "time_per_iteration": 2.461588144302368
    },
    {
      "auxiliary_loss_clip": 0.01148912,
      "auxiliary_loss_mlp": 0.01031431,
      "balance_loss_clip": 1.04862261,
      "balance_loss_mlp": 1.02437115,
      "epoch": 0.5451812661576384,
      "flos": 17785083375360.0,
      "grad_norm": 3.9404851440783757,
      "language_loss": 0.815759,
      "learning_rate": 1.8047894080819945e-06,
      "loss": 0.83756238,
      "num_input_tokens_seen": 97769985,
      "step": 4534,
      "time_per_iteration": 3.4477219581604004
    },
    {
      "auxiliary_loss_clip": 0.01068408,
      "auxiliary_loss_mlp": 0.01001064,
      "balance_loss_clip": 1.0182842,
      "balance_loss_mlp": 0.99998468,
      "epoch": 0.5453015090482776,
      "flos": 71062586513280.0,
      "grad_norm": 0.7231144361059072,
      "language_loss": 0.63193268,
      "learning_rate": 1.8040141706918258e-06,
      "loss": 0.65262741,
      "num_input_tokens_seen": 97831225,
      "step": 4535,
      "time_per_iteration": 3.2205827236175537
    },
    {
      "auxiliary_loss_clip": 0.01146099,
      "auxiliary_loss_mlp": 0.01036797,
      "balance_loss_clip": 1.05124354,
      "balance_loss_mlp": 1.0294416,
      "epoch": 0.5454217519389166,
      "flos": 25552201622400.0,
      "grad_norm": 1.8177189080241651,
      "language_loss": 0.76761949,
      "learning_rate": 1.8032389630325525e-06,
      "loss": 0.78944844,
      "num_input_tokens_seen": 97849975,
      "step": 4536,
      "time_per_iteration": 3.3661441802978516
    },
    {
      "auxiliary_loss_clip": 0.0114362,
      "auxiliary_loss_mlp": 0.01024108,
      "balance_loss_clip": 1.04488063,
      "balance_loss_mlp": 1.01684821,
      "epoch": 0.5455419948295557,
      "flos": 23658345037440.0,
      "grad_norm": 1.6596121196440627,
      "language_loss": 0.75721228,
      "learning_rate": 1.8024637852217707e-06,
      "loss": 0.77888954,
      "num_input_tokens_seen": 97869700,
      "step": 4537,
      "time_per_iteration": 3.3125128746032715
    },
    {
      "auxiliary_loss_clip": 0.01145971,
      "auxiliary_loss_mlp": 0.01031654,
      "balance_loss_clip": 1.04975867,
      "balance_loss_mlp": 1.02489495,
      "epoch": 0.5456622377201948,
      "flos": 23403989854080.0,
      "grad_norm": 1.7108320441418228,
      "language_loss": 0.84732544,
      "learning_rate": 1.8016886373770766e-06,
      "loss": 0.8691017,
      "num_input_tokens_seen": 97888215,
      "step": 4538,
      "time_per_iteration": 2.5284061431884766
    },
    {
      "auxiliary_loss_clip": 0.01143021,
      "auxiliary_loss_mlp": 0.01030581,
      "balance_loss_clip": 1.04706335,
      "balance_loss_mlp": 1.02353549,
      "epoch": 0.5457824806108339,
      "flos": 23988040997760.0,
      "grad_norm": 1.644916313372015,
      "language_loss": 0.78873444,
      "learning_rate": 1.8009135196160579e-06,
      "loss": 0.81047046,
      "num_input_tokens_seen": 97907090,
      "step": 4539,
      "time_per_iteration": 2.560939311981201
    },
    {
      "auxiliary_loss_clip": 0.01125117,
      "auxiliary_loss_mlp": 0.01030086,
      "balance_loss_clip": 1.04413843,
      "balance_loss_mlp": 1.02297544,
      "epoch": 0.545902723501473,
      "flos": 22565870835840.0,
      "grad_norm": 1.6988995798717297,
      "language_loss": 0.84048647,
      "learning_rate": 1.8001384320563e-06,
      "loss": 0.86203849,
      "num_input_tokens_seen": 97927345,
      "step": 4540,
      "time_per_iteration": 2.5556275844573975
    },
    {
      "auxiliary_loss_clip": 0.01067289,
      "auxiliary_loss_mlp": 0.01002623,
      "balance_loss_clip": 1.01714301,
      "balance_loss_mlp": 1.00154436,
      "epoch": 0.5460229663921121,
      "flos": 55198399685760.0,
      "grad_norm": 0.7714194369362715,
      "language_loss": 0.57780403,
      "learning_rate": 1.7993633748153833e-06,
      "loss": 0.59850311,
      "num_input_tokens_seen": 97981950,
      "step": 4541,
      "time_per_iteration": 2.9763591289520264
    },
    {
      "auxiliary_loss_clip": 0.01162699,
      "auxiliary_loss_mlp": 0.01033448,
      "balance_loss_clip": 1.05059314,
      "balance_loss_mlp": 1.02566361,
      "epoch": 0.5461432092827512,
      "flos": 15413866018560.0,
      "grad_norm": 1.828538665541505,
      "language_loss": 0.72662723,
      "learning_rate": 1.7985883480108834e-06,
      "loss": 0.74858868,
      "num_input_tokens_seen": 97999585,
      "step": 4542,
      "time_per_iteration": 2.4843533039093018
    },
    {
      "auxiliary_loss_clip": 0.01153242,
      "auxiliary_loss_mlp": 0.01029447,
      "balance_loss_clip": 1.04845071,
      "balance_loss_mlp": 1.02203846,
      "epoch": 0.5462634521733902,
      "flos": 24024921287040.0,
      "grad_norm": 1.5135727600232125,
      "language_loss": 0.72103,
      "learning_rate": 1.797813351760371e-06,
      "loss": 0.74285686,
      "num_input_tokens_seen": 98021290,
      "step": 4543,
      "time_per_iteration": 2.5319628715515137
    },
    {
      "auxiliary_loss_clip": 0.01173669,
      "auxiliary_loss_mlp": 0.01030659,
      "balance_loss_clip": 1.05092025,
      "balance_loss_mlp": 1.0224576,
      "epoch": 0.5463836950640293,
      "flos": 22820944291200.0,
      "grad_norm": 1.6934831955837053,
      "language_loss": 0.78101063,
      "learning_rate": 1.7970383861814116e-06,
      "loss": 0.80305398,
      "num_input_tokens_seen": 98041060,
      "step": 4544,
      "time_per_iteration": 2.484126091003418
    },
    {
      "auxiliary_loss_clip": 0.0116058,
      "auxiliary_loss_mlp": 0.01029918,
      "balance_loss_clip": 1.05111182,
      "balance_loss_mlp": 1.0222683,
      "epoch": 0.5465039379546685,
      "flos": 20448290390400.0,
      "grad_norm": 1.8501570565190402,
      "language_loss": 0.73714715,
      "learning_rate": 1.7962634513915684e-06,
      "loss": 0.75905216,
      "num_input_tokens_seen": 98058410,
      "step": 4545,
      "time_per_iteration": 2.478583574295044
    },
    {
      "auxiliary_loss_clip": 0.01170668,
      "auxiliary_loss_mlp": 0.01026131,
      "balance_loss_clip": 1.05007172,
      "balance_loss_mlp": 1.01894236,
      "epoch": 0.5466241808453075,
      "flos": 17343310003200.0,
      "grad_norm": 1.6417209251406886,
      "language_loss": 0.7935577,
      "learning_rate": 1.7954885475083969e-06,
      "loss": 0.81552565,
      "num_input_tokens_seen": 98076080,
      "step": 4546,
      "time_per_iteration": 2.455613851547241
    },
    {
      "auxiliary_loss_clip": 0.01175657,
      "auxiliary_loss_mlp": 0.0102767,
      "balance_loss_clip": 1.05358899,
      "balance_loss_mlp": 1.01974249,
      "epoch": 0.5467444237359466,
      "flos": 21617039122560.0,
      "grad_norm": 1.9834666222739512,
      "language_loss": 0.72459865,
      "learning_rate": 1.7947136746494513e-06,
      "loss": 0.74663192,
      "num_input_tokens_seen": 98096995,
      "step": 4547,
      "time_per_iteration": 2.5557448863983154
    },
    {
      "auxiliary_loss_clip": 0.01156711,
      "auxiliary_loss_mlp": 0.01034309,
      "balance_loss_clip": 1.05012679,
      "balance_loss_mlp": 1.02667999,
      "epoch": 0.5468646666265857,
      "flos": 24170467196160.0,
      "grad_norm": 1.9321732620102325,
      "language_loss": 0.87753856,
      "learning_rate": 1.793938832932277e-06,
      "loss": 0.89944869,
      "num_input_tokens_seen": 98115105,
      "step": 4548,
      "time_per_iteration": 2.5109670162200928
    },
    {
      "auxiliary_loss_clip": 0.01173557,
      "auxiliary_loss_mlp": 0.01028551,
      "balance_loss_clip": 1.05136085,
      "balance_loss_mlp": 1.02087426,
      "epoch": 0.5469849095172248,
      "flos": 27527001505920.0,
      "grad_norm": 1.829343692529943,
      "language_loss": 0.70278496,
      "learning_rate": 1.7931640224744185e-06,
      "loss": 0.72480607,
      "num_input_tokens_seen": 98135655,
      "step": 4549,
      "time_per_iteration": 2.492534637451172
    },
    {
      "auxiliary_loss_clip": 0.01117806,
      "auxiliary_loss_mlp": 0.0102899,
      "balance_loss_clip": 1.04053283,
      "balance_loss_mlp": 1.02150655,
      "epoch": 0.5471051524078638,
      "flos": 27964680727680.0,
      "grad_norm": 2.260880458872386,
      "language_loss": 0.73672605,
      "learning_rate": 1.7923892433934127e-06,
      "loss": 0.75819409,
      "num_input_tokens_seen": 98156730,
      "step": 4550,
      "time_per_iteration": 2.633251428604126
    },
    {
      "auxiliary_loss_clip": 0.01145147,
      "auxiliary_loss_mlp": 0.0076131,
      "balance_loss_clip": 1.04831123,
      "balance_loss_mlp": 1.0002234,
      "epoch": 0.547225395298503,
      "flos": 18150510389760.0,
      "grad_norm": 1.6664445758579618,
      "language_loss": 0.78824514,
      "learning_rate": 1.7916144958067939e-06,
      "loss": 0.80730969,
      "num_input_tokens_seen": 98174590,
      "step": 4551,
      "time_per_iteration": 2.517235517501831
    },
    {
      "auxiliary_loss_clip": 0.01160045,
      "auxiliary_loss_mlp": 0.0102216,
      "balance_loss_clip": 1.04984426,
      "balance_loss_mlp": 1.01506388,
      "epoch": 0.5473456381891421,
      "flos": 21361498790400.0,
      "grad_norm": 1.750205297808644,
      "language_loss": 0.792377,
      "learning_rate": 1.7908397798320905e-06,
      "loss": 0.81419903,
      "num_input_tokens_seen": 98194325,
      "step": 4552,
      "time_per_iteration": 2.491636037826538
    },
    {
      "auxiliary_loss_clip": 0.0116203,
      "auxiliary_loss_mlp": 0.00761608,
      "balance_loss_clip": 1.05250609,
      "balance_loss_mlp": 1.0002079,
      "epoch": 0.5474658810797811,
      "flos": 19932145908480.0,
      "grad_norm": 1.6173497418781952,
      "language_loss": 0.74571586,
      "learning_rate": 1.7900650955868265e-06,
      "loss": 0.76495218,
      "num_input_tokens_seen": 98213970,
      "step": 4553,
      "time_per_iteration": 2.4878342151641846
    },
    {
      "auxiliary_loss_clip": 0.01163259,
      "auxiliary_loss_mlp": 0.00761305,
      "balance_loss_clip": 1.05455184,
      "balance_loss_mlp": 1.0002178,
      "epoch": 0.5475861239704203,
      "flos": 50476217264640.0,
      "grad_norm": 1.3778755152156756,
      "language_loss": 0.76573819,
      "learning_rate": 1.7892904431885202e-06,
      "loss": 0.78498387,
      "num_input_tokens_seen": 98241145,
      "step": 4554,
      "time_per_iteration": 2.76833176612854
    },
    {
      "auxiliary_loss_clip": 0.01117304,
      "auxiliary_loss_mlp": 0.01026813,
      "balance_loss_clip": 1.0427562,
      "balance_loss_mlp": 1.02029538,
      "epoch": 0.5477063668610593,
      "flos": 20705123612160.0,
      "grad_norm": 1.7712828200376882,
      "language_loss": 0.75249958,
      "learning_rate": 1.788515822754686e-06,
      "loss": 0.77394074,
      "num_input_tokens_seen": 98261565,
      "step": 4555,
      "time_per_iteration": 3.402116537094116
    },
    {
      "auxiliary_loss_clip": 0.0112919,
      "auxiliary_loss_mlp": 0.01028574,
      "balance_loss_clip": 1.04337561,
      "balance_loss_mlp": 1.02060485,
      "epoch": 0.5478266097516984,
      "flos": 19609740408960.0,
      "grad_norm": 2.107175163416886,
      "language_loss": 0.78708494,
      "learning_rate": 1.7877412344028335e-06,
      "loss": 0.80866253,
      "num_input_tokens_seen": 98281370,
      "step": 4556,
      "time_per_iteration": 2.570615530014038
    },
    {
      "auxiliary_loss_clip": 0.0116199,
      "auxiliary_loss_mlp": 0.0102634,
      "balance_loss_clip": 1.05012286,
      "balance_loss_mlp": 1.01919341,
      "epoch": 0.5479468526423376,
      "flos": 12896599962240.0,
      "grad_norm": 2.4578022591741076,
      "language_loss": 0.77394599,
      "learning_rate": 1.7869666782504668e-06,
      "loss": 0.7958293,
      "num_input_tokens_seen": 98297950,
      "step": 4557,
      "time_per_iteration": 2.461474657058716
    },
    {
      "auxiliary_loss_clip": 0.01133866,
      "auxiliary_loss_mlp": 0.01026149,
      "balance_loss_clip": 1.04528844,
      "balance_loss_mlp": 1.01891243,
      "epoch": 0.5480670955329766,
      "flos": 18588800142720.0,
      "grad_norm": 1.8186025090460671,
      "language_loss": 0.69103527,
      "learning_rate": 1.7861921544150867e-06,
      "loss": 0.7126354,
      "num_input_tokens_seen": 98316800,
      "step": 4558,
      "time_per_iteration": 2.5157859325408936
    },
    {
      "auxiliary_loss_clip": 0.01090958,
      "auxiliary_loss_mlp": 0.00761339,
      "balance_loss_clip": 1.04270601,
      "balance_loss_mlp": 1.00019062,
      "epoch": 0.5481873384236157,
      "flos": 15954608338560.0,
      "grad_norm": 1.7055631689701296,
      "language_loss": 0.76412821,
      "learning_rate": 1.7854176630141856e-06,
      "loss": 0.78265119,
      "num_input_tokens_seen": 98333935,
      "step": 4559,
      "time_per_iteration": 2.6016974449157715
    },
    {
      "auxiliary_loss_clip": 0.01176187,
      "auxiliary_loss_mlp": 0.01029932,
      "balance_loss_clip": 1.05330229,
      "balance_loss_mlp": 1.02183819,
      "epoch": 0.5483075813142548,
      "flos": 22783812606720.0,
      "grad_norm": 2.1427418609319004,
      "language_loss": 0.84529459,
      "learning_rate": 1.784643204165255e-06,
      "loss": 0.86735582,
      "num_input_tokens_seen": 98353255,
      "step": 4560,
      "time_per_iteration": 3.2781002521514893
    },
    {
      "auxiliary_loss_clip": 0.01152323,
      "auxiliary_loss_mlp": 0.01026916,
      "balance_loss_clip": 1.0506742,
      "balance_loss_mlp": 1.01950729,
      "epoch": 0.5484278242048939,
      "flos": 19317212046720.0,
      "grad_norm": 1.9663156552665761,
      "language_loss": 0.77177197,
      "learning_rate": 1.7838687779857783e-06,
      "loss": 0.79356432,
      "num_input_tokens_seen": 98371130,
      "step": 4561,
      "time_per_iteration": 2.4860503673553467
    },
    {
      "auxiliary_loss_clip": 0.0113656,
      "auxiliary_loss_mlp": 0.01028841,
      "balance_loss_clip": 1.04474592,
      "balance_loss_mlp": 1.02136338,
      "epoch": 0.5485480670955329,
      "flos": 22816024128000.0,
      "grad_norm": 1.7899858350517748,
      "language_loss": 0.64244628,
      "learning_rate": 1.7830943845932366e-06,
      "loss": 0.66410035,
      "num_input_tokens_seen": 98390455,
      "step": 4562,
      "time_per_iteration": 3.2955172061920166
    },
    {
      "auxiliary_loss_clip": 0.01147174,
      "auxiliary_loss_mlp": 0.01024357,
      "balance_loss_clip": 1.04909444,
      "balance_loss_mlp": 1.01722217,
      "epoch": 0.5486683099861721,
      "flos": 22671304231680.0,
      "grad_norm": 4.059680989245851,
      "language_loss": 0.75091404,
      "learning_rate": 1.7823200241051044e-06,
      "loss": 0.77262926,
      "num_input_tokens_seen": 98409370,
      "step": 4563,
      "time_per_iteration": 3.34077787399292
    },
    {
      "auxiliary_loss_clip": 0.01175585,
      "auxiliary_loss_mlp": 0.01028363,
      "balance_loss_clip": 1.05474854,
      "balance_loss_mlp": 1.02142513,
      "epoch": 0.5487885528768112,
      "flos": 23149383275520.0,
      "grad_norm": 1.8910477433949866,
      "language_loss": 0.80734062,
      "learning_rate": 1.7815456966388513e-06,
      "loss": 0.8293801,
      "num_input_tokens_seen": 98428465,
      "step": 4564,
      "time_per_iteration": 2.4698710441589355
    },
    {
      "auxiliary_loss_clip": 0.01130069,
      "auxiliary_loss_mlp": 0.01027893,
      "balance_loss_clip": 1.04536223,
      "balance_loss_mlp": 1.02038836,
      "epoch": 0.5489087957674502,
      "flos": 22053928245120.0,
      "grad_norm": 1.9555679514594209,
      "language_loss": 0.80994964,
      "learning_rate": 1.780771402311943e-06,
      "loss": 0.83152926,
      "num_input_tokens_seen": 98447300,
      "step": 4565,
      "time_per_iteration": 2.5577876567840576
    },
    {
      "auxiliary_loss_clip": 0.01147415,
      "auxiliary_loss_mlp": 0.01028587,
      "balance_loss_clip": 1.05255055,
      "balance_loss_mlp": 1.02125597,
      "epoch": 0.5490290386580894,
      "flos": 24315977191680.0,
      "grad_norm": 1.6599988998710056,
      "language_loss": 0.7865994,
      "learning_rate": 1.7799971412418374e-06,
      "loss": 0.8083595,
      "num_input_tokens_seen": 98468695,
      "step": 4566,
      "time_per_iteration": 2.534700393676758
    },
    {
      "auxiliary_loss_clip": 0.01128641,
      "auxiliary_loss_mlp": 0.01034545,
      "balance_loss_clip": 1.04632866,
      "balance_loss_mlp": 1.02618885,
      "epoch": 0.5491492815487284,
      "flos": 18294942977280.0,
      "grad_norm": 2.0046325819587287,
      "language_loss": 0.7382971,
      "learning_rate": 1.7792229135459918e-06,
      "loss": 0.75992894,
      "num_input_tokens_seen": 98485345,
      "step": 4567,
      "time_per_iteration": 2.522423505783081
    },
    {
      "auxiliary_loss_clip": 0.01022027,
      "auxiliary_loss_mlp": 0.01004236,
      "balance_loss_clip": 1.01978636,
      "balance_loss_mlp": 1.00310314,
      "epoch": 0.5492695244393675,
      "flos": 64550257050240.0,
      "grad_norm": 0.7343310433939847,
      "language_loss": 0.61704183,
      "learning_rate": 1.7784487193418538e-06,
      "loss": 0.63730448,
      "num_input_tokens_seen": 98543195,
      "step": 4568,
      "time_per_iteration": 3.0791804790496826
    },
    {
      "auxiliary_loss_clip": 0.01114263,
      "auxiliary_loss_mlp": 0.0102549,
      "balance_loss_clip": 1.04030704,
      "balance_loss_mlp": 1.01722026,
      "epoch": 0.5493897673300067,
      "flos": 17379579761280.0,
      "grad_norm": 1.7656708256883211,
      "language_loss": 0.61381847,
      "learning_rate": 1.7776745587468698e-06,
      "loss": 0.635216,
      "num_input_tokens_seen": 98560620,
      "step": 4569,
      "time_per_iteration": 2.523169755935669
    },
    {
      "auxiliary_loss_clip": 0.01170477,
      "auxiliary_loss_mlp": 0.01027694,
      "balance_loss_clip": 1.0485729,
      "balance_loss_mlp": 1.02011275,
      "epoch": 0.5495100102206457,
      "flos": 19901765980800.0,
      "grad_norm": 2.2555221531247156,
      "language_loss": 0.81925845,
      "learning_rate": 1.7769004318784776e-06,
      "loss": 0.84124017,
      "num_input_tokens_seen": 98578265,
      "step": 4570,
      "time_per_iteration": 2.4462785720825195
    },
    {
      "auxiliary_loss_clip": 0.0115996,
      "auxiliary_loss_mlp": 0.010267,
      "balance_loss_clip": 1.04938567,
      "balance_loss_mlp": 1.01895773,
      "epoch": 0.5496302531112848,
      "flos": 16727190992640.0,
      "grad_norm": 1.744062514430034,
      "language_loss": 0.80500263,
      "learning_rate": 1.776126338854113e-06,
      "loss": 0.82686925,
      "num_input_tokens_seen": 98596055,
      "step": 4571,
      "time_per_iteration": 2.4639976024627686
    },
    {
      "auxiliary_loss_clip": 0.01152624,
      "auxiliary_loss_mlp": 0.01025076,
      "balance_loss_clip": 1.04954004,
      "balance_loss_mlp": 1.01815319,
      "epoch": 0.5497504960019239,
      "flos": 24572343536640.0,
      "grad_norm": 1.5742426786391595,
      "language_loss": 0.84396642,
      "learning_rate": 1.7753522797912044e-06,
      "loss": 0.8657434,
      "num_input_tokens_seen": 98616140,
      "step": 4572,
      "time_per_iteration": 2.531416177749634
    },
    {
      "auxiliary_loss_clip": 0.01150438,
      "auxiliary_loss_mlp": 0.01028978,
      "balance_loss_clip": 1.04715669,
      "balance_loss_mlp": 1.02133048,
      "epoch": 0.549870738892563,
      "flos": 15450494912640.0,
      "grad_norm": 2.2784223848045064,
      "language_loss": 0.69566965,
      "learning_rate": 1.7745782548071765e-06,
      "loss": 0.71746379,
      "num_input_tokens_seen": 98633035,
      "step": 4573,
      "time_per_iteration": 2.509826898574829
    },
    {
      "auxiliary_loss_clip": 0.01123955,
      "auxiliary_loss_mlp": 0.01029468,
      "balance_loss_clip": 1.04771638,
      "balance_loss_mlp": 1.02252364,
      "epoch": 0.549990981783202,
      "flos": 21069114082560.0,
      "grad_norm": 1.6586114935805851,
      "language_loss": 0.74221957,
      "learning_rate": 1.7738042640194482e-06,
      "loss": 0.76375383,
      "num_input_tokens_seen": 98652700,
      "step": 4574,
      "time_per_iteration": 2.5439114570617676
    },
    {
      "auxiliary_loss_clip": 0.01170217,
      "auxiliary_loss_mlp": 0.0102612,
      "balance_loss_clip": 1.04890943,
      "balance_loss_mlp": 1.01859164,
      "epoch": 0.5501112246738411,
      "flos": 21395901041280.0,
      "grad_norm": 1.5356503032671067,
      "language_loss": 0.70789921,
      "learning_rate": 1.7730303075454335e-06,
      "loss": 0.72986257,
      "num_input_tokens_seen": 98671590,
      "step": 4575,
      "time_per_iteration": 2.4466545581817627
    },
    {
      "auxiliary_loss_clip": 0.01133062,
      "auxiliary_loss_mlp": 0.01037204,
      "balance_loss_clip": 1.04504657,
      "balance_loss_mlp": 1.02987289,
      "epoch": 0.5502314675644803,
      "flos": 17456931699840.0,
      "grad_norm": 2.562781383827615,
      "language_loss": 0.84948564,
      "learning_rate": 1.7722563855025402e-06,
      "loss": 0.87118828,
      "num_input_tokens_seen": 98689620,
      "step": 4576,
      "time_per_iteration": 2.5518813133239746
    },
    {
      "auxiliary_loss_clip": 0.01142781,
      "auxiliary_loss_mlp": 0.01026494,
      "balance_loss_clip": 1.04436898,
      "balance_loss_mlp": 1.01886439,
      "epoch": 0.5503517104551193,
      "flos": 24310410583680.0,
      "grad_norm": 2.1242126012427676,
      "language_loss": 0.70565581,
      "learning_rate": 1.7714824980081721e-06,
      "loss": 0.72734857,
      "num_input_tokens_seen": 98708915,
      "step": 4577,
      "time_per_iteration": 2.577956438064575
    },
    {
      "auxiliary_loss_clip": 0.01156753,
      "auxiliary_loss_mlp": 0.01025,
      "balance_loss_clip": 1.0519948,
      "balance_loss_mlp": 1.01735926,
      "epoch": 0.5504719533457584,
      "flos": 22419427086720.0,
      "grad_norm": 1.7776206477358734,
      "language_loss": 0.73763883,
      "learning_rate": 1.7707086451797276e-06,
      "loss": 0.7594564,
      "num_input_tokens_seen": 98729790,
      "step": 4578,
      "time_per_iteration": 2.5243818759918213
    },
    {
      "auxiliary_loss_clip": 0.01035108,
      "auxiliary_loss_mlp": 0.0100029,
      "balance_loss_clip": 1.01422596,
      "balance_loss_mlp": 0.99912137,
      "epoch": 0.5505921962363975,
      "flos": 67294155968640.0,
      "grad_norm": 0.6977149925371015,
      "language_loss": 0.52379906,
      "learning_rate": 1.7699348271345993e-06,
      "loss": 0.54415298,
      "num_input_tokens_seen": 98792415,
      "step": 4579,
      "time_per_iteration": 3.0923850536346436
    },
    {
      "auxiliary_loss_clip": 0.01030531,
      "auxiliary_loss_mlp": 0.01000115,
      "balance_loss_clip": 1.01635742,
      "balance_loss_mlp": 0.99888676,
      "epoch": 0.5507124391270366,
      "flos": 45685125578880.0,
      "grad_norm": 0.7126373152363171,
      "language_loss": 0.54478097,
      "learning_rate": 1.7691610439901753e-06,
      "loss": 0.56508744,
      "num_input_tokens_seen": 98855350,
      "step": 4580,
      "time_per_iteration": 3.262882947921753
    },
    {
      "auxiliary_loss_clip": 0.01160494,
      "auxiliary_loss_mlp": 0.01033492,
      "balance_loss_clip": 1.05000532,
      "balance_loss_mlp": 1.02638984,
      "epoch": 0.5508326820176757,
      "flos": 22273845264000.0,
      "grad_norm": 1.7906486332283178,
      "language_loss": 0.75606775,
      "learning_rate": 1.7683872958638367e-06,
      "loss": 0.77800763,
      "num_input_tokens_seen": 98874230,
      "step": 4581,
      "time_per_iteration": 3.25884747505188
    },
    {
      "auxiliary_loss_clip": 0.01140895,
      "auxiliary_loss_mlp": 0.01028802,
      "balance_loss_clip": 1.04528356,
      "balance_loss_mlp": 1.02155149,
      "epoch": 0.5509529249083148,
      "flos": 20012442762240.0,
      "grad_norm": 2.4949820609558615,
      "language_loss": 0.83936059,
      "learning_rate": 1.7676135828729614e-06,
      "loss": 0.86105764,
      "num_input_tokens_seen": 98893940,
      "step": 4582,
      "time_per_iteration": 2.5592641830444336
    },
    {
      "auxiliary_loss_clip": 0.01157778,
      "auxiliary_loss_mlp": 0.01027056,
      "balance_loss_clip": 1.05166364,
      "balance_loss_mlp": 1.01915836,
      "epoch": 0.5510731677989539,
      "flos": 21834801325440.0,
      "grad_norm": 4.714340527888407,
      "language_loss": 0.82974112,
      "learning_rate": 1.7668399051349205e-06,
      "loss": 0.8515895,
      "num_input_tokens_seen": 98913620,
      "step": 4583,
      "time_per_iteration": 2.5226328372955322
    },
    {
      "auxiliary_loss_clip": 0.01125208,
      "auxiliary_loss_mlp": 0.01027993,
      "balance_loss_clip": 1.04520416,
      "balance_loss_mlp": 1.02044725,
      "epoch": 0.5511934106895929,
      "flos": 21467901853440.0,
      "grad_norm": 1.8477823110538816,
      "language_loss": 0.83591437,
      "learning_rate": 1.766066262767081e-06,
      "loss": 0.85744643,
      "num_input_tokens_seen": 98931460,
      "step": 4584,
      "time_per_iteration": 2.574920177459717
    },
    {
      "auxiliary_loss_clip": 0.01140331,
      "auxiliary_loss_mlp": 0.01025159,
      "balance_loss_clip": 1.0485034,
      "balance_loss_mlp": 1.01798558,
      "epoch": 0.5513136535802321,
      "flos": 21068934514560.0,
      "grad_norm": 2.270652175854672,
      "language_loss": 0.76980036,
      "learning_rate": 1.765292655886803e-06,
      "loss": 0.79145527,
      "num_input_tokens_seen": 98950105,
      "step": 4585,
      "time_per_iteration": 2.5350422859191895
    },
    {
      "auxiliary_loss_clip": 0.01137724,
      "auxiliary_loss_mlp": 0.01029298,
      "balance_loss_clip": 1.04688191,
      "balance_loss_mlp": 1.02231586,
      "epoch": 0.5514338964708712,
      "flos": 27815004754560.0,
      "grad_norm": 1.7868467903114176,
      "language_loss": 0.70571232,
      "learning_rate": 1.764519084611443e-06,
      "loss": 0.7273826,
      "num_input_tokens_seen": 98970560,
      "step": 4586,
      "time_per_iteration": 3.395937919616699
    },
    {
      "auxiliary_loss_clip": 0.01141787,
      "auxiliary_loss_mlp": 0.01027977,
      "balance_loss_clip": 1.04618323,
      "balance_loss_mlp": 1.02019858,
      "epoch": 0.5515541393615102,
      "flos": 21908525990400.0,
      "grad_norm": 1.919060028342727,
      "language_loss": 0.77785206,
      "learning_rate": 1.7637455490583505e-06,
      "loss": 0.7995497,
      "num_input_tokens_seen": 98989885,
      "step": 4587,
      "time_per_iteration": 2.5202691555023193
    },
    {
      "auxiliary_loss_clip": 0.01160296,
      "auxiliary_loss_mlp": 0.01027172,
      "balance_loss_clip": 1.05123818,
      "balance_loss_mlp": 1.02049959,
      "epoch": 0.5516743822521494,
      "flos": 20485422074880.0,
      "grad_norm": 1.9198999756093793,
      "language_loss": 0.77259994,
      "learning_rate": 1.7629720493448701e-06,
      "loss": 0.79447466,
      "num_input_tokens_seen": 99007180,
      "step": 4588,
      "time_per_iteration": 3.2994871139526367
    },
    {
      "auxiliary_loss_clip": 0.01149638,
      "auxiliary_loss_mlp": 0.0102815,
      "balance_loss_clip": 1.04642558,
      "balance_loss_mlp": 1.02020431,
      "epoch": 0.5517946251427884,
      "flos": 14940383915520.0,
      "grad_norm": 1.755884411483945,
      "language_loss": 0.84661424,
      "learning_rate": 1.7621985855883418e-06,
      "loss": 0.86839211,
      "num_input_tokens_seen": 99023880,
      "step": 4589,
      "time_per_iteration": 3.2289061546325684
    },
    {
      "auxiliary_loss_clip": 0.01137643,
      "auxiliary_loss_mlp": 0.01025895,
      "balance_loss_clip": 1.04694676,
      "balance_loss_mlp": 1.01895738,
      "epoch": 0.5519148680334275,
      "flos": 18404865573120.0,
      "grad_norm": 1.7308637367114907,
      "language_loss": 0.72175002,
      "learning_rate": 1.7614251579060983e-06,
      "loss": 0.74338543,
      "num_input_tokens_seen": 99042475,
      "step": 4590,
      "time_per_iteration": 2.514972686767578
    },
    {
      "auxiliary_loss_clip": 0.01130647,
      "auxiliary_loss_mlp": 0.01027326,
      "balance_loss_clip": 1.04660034,
      "balance_loss_mlp": 1.01999509,
      "epoch": 0.5520351109240667,
      "flos": 25113337251840.0,
      "grad_norm": 1.7653783526061013,
      "language_loss": 0.84728837,
      "learning_rate": 1.76065176641547e-06,
      "loss": 0.86886811,
      "num_input_tokens_seen": 99065185,
      "step": 4591,
      "time_per_iteration": 2.635100841522217
    },
    {
      "auxiliary_loss_clip": 0.01156072,
      "auxiliary_loss_mlp": 0.01025392,
      "balance_loss_clip": 1.04614949,
      "balance_loss_mlp": 1.01855254,
      "epoch": 0.5521553538147057,
      "flos": 21069545045760.0,
      "grad_norm": 1.784774042962693,
      "language_loss": 0.77672464,
      "learning_rate": 1.759878411233777e-06,
      "loss": 0.79853928,
      "num_input_tokens_seen": 99083645,
      "step": 4592,
      "time_per_iteration": 2.5189290046691895
    },
    {
      "auxiliary_loss_clip": 0.01158806,
      "auxiliary_loss_mlp": 0.01024598,
      "balance_loss_clip": 1.05024934,
      "balance_loss_mlp": 1.0173173,
      "epoch": 0.5522755967053448,
      "flos": 18879999701760.0,
      "grad_norm": 2.06873187758352,
      "language_loss": 0.76119184,
      "learning_rate": 1.7591050924783388e-06,
      "loss": 0.78302592,
      "num_input_tokens_seen": 99100835,
      "step": 4593,
      "time_per_iteration": 2.4737982749938965
    },
    {
      "auxiliary_loss_clip": 0.01021753,
      "auxiliary_loss_mlp": 0.01005109,
      "balance_loss_clip": 1.01356292,
      "balance_loss_mlp": 1.00386906,
      "epoch": 0.5523958395959839,
      "flos": 64675622494080.0,
      "grad_norm": 0.8365527394717246,
      "language_loss": 0.57946771,
      "learning_rate": 1.7583318102664661e-06,
      "loss": 0.59973633,
      "num_input_tokens_seen": 99168400,
      "step": 4594,
      "time_per_iteration": 3.2207822799682617
    },
    {
      "auxiliary_loss_clip": 0.01157113,
      "auxiliary_loss_mlp": 0.01025458,
      "balance_loss_clip": 1.04565787,
      "balance_loss_mlp": 1.01822519,
      "epoch": 0.552516082486623,
      "flos": 10889732211840.0,
      "grad_norm": 1.845825808185984,
      "language_loss": 0.79114807,
      "learning_rate": 1.757558564715466e-06,
      "loss": 0.81297374,
      "num_input_tokens_seen": 99186475,
      "step": 4595,
      "time_per_iteration": 2.4805736541748047
    },
    {
      "auxiliary_loss_clip": 0.01156231,
      "auxiliary_loss_mlp": 0.01037088,
      "balance_loss_clip": 1.04692173,
      "balance_loss_mlp": 1.0293808,
      "epoch": 0.552636325377262,
      "flos": 22199797376640.0,
      "grad_norm": 2.8190629638858096,
      "language_loss": 0.74365175,
      "learning_rate": 1.7567853559426386e-06,
      "loss": 0.76558495,
      "num_input_tokens_seen": 99203525,
      "step": 4596,
      "time_per_iteration": 2.492171287536621
    },
    {
      "auxiliary_loss_clip": 0.01159664,
      "auxiliary_loss_mlp": 0.01027908,
      "balance_loss_clip": 1.04950225,
      "balance_loss_mlp": 1.02094007,
      "epoch": 0.5527565682679012,
      "flos": 23988184652160.0,
      "grad_norm": 2.8282993587365755,
      "language_loss": 0.74925625,
      "learning_rate": 1.7560121840652797e-06,
      "loss": 0.77113199,
      "num_input_tokens_seen": 99222910,
      "step": 4597,
      "time_per_iteration": 2.52915620803833
    },
    {
      "auxiliary_loss_clip": 0.01122148,
      "auxiliary_loss_mlp": 0.01031495,
      "balance_loss_clip": 1.04566026,
      "balance_loss_mlp": 1.02411628,
      "epoch": 0.5528768111585403,
      "flos": 19719267955200.0,
      "grad_norm": 2.552780706430194,
      "language_loss": 0.68997312,
      "learning_rate": 1.7552390492006782e-06,
      "loss": 0.71150959,
      "num_input_tokens_seen": 99241230,
      "step": 4598,
      "time_per_iteration": 2.588440179824829
    },
    {
      "auxiliary_loss_clip": 0.01118237,
      "auxiliary_loss_mlp": 0.00761326,
      "balance_loss_clip": 1.04118967,
      "balance_loss_mlp": 1.00025082,
      "epoch": 0.5529970540491793,
      "flos": 26215975002240.0,
      "grad_norm": 1.9736144122948742,
      "language_loss": 0.65401798,
      "learning_rate": 1.7544659514661184e-06,
      "loss": 0.67281365,
      "num_input_tokens_seen": 99264320,
      "step": 4599,
      "time_per_iteration": 2.6811771392822266
    },
    {
      "auxiliary_loss_clip": 0.01138025,
      "auxiliary_loss_mlp": 0.01027386,
      "balance_loss_clip": 1.0448494,
      "balance_loss_mlp": 1.02048695,
      "epoch": 0.5531172969398185,
      "flos": 24425971614720.0,
      "grad_norm": 2.3820874255264832,
      "language_loss": 0.79514092,
      "learning_rate": 1.7536928909788786e-06,
      "loss": 0.81679505,
      "num_input_tokens_seen": 99283625,
      "step": 4600,
      "time_per_iteration": 2.554486036300659
    },
    {
      "auxiliary_loss_clip": 0.01025663,
      "auxiliary_loss_mlp": 0.01001878,
      "balance_loss_clip": 1.01687181,
      "balance_loss_mlp": 1.00070941,
      "epoch": 0.5532375398304575,
      "flos": 64907316195840.0,
      "grad_norm": 0.8806827377861195,
      "language_loss": 0.61988175,
      "learning_rate": 1.752919867856231e-06,
      "loss": 0.6401571,
      "num_input_tokens_seen": 99335270,
      "step": 4601,
      "time_per_iteration": 2.9947681427001953
    },
    {
      "auxiliary_loss_clip": 0.01135255,
      "auxiliary_loss_mlp": 0.01024353,
      "balance_loss_clip": 1.04454625,
      "balance_loss_mlp": 1.01702142,
      "epoch": 0.5533577827210966,
      "flos": 19683105937920.0,
      "grad_norm": 1.9156916834007982,
      "language_loss": 0.78558278,
      "learning_rate": 1.7521468822154436e-06,
      "loss": 0.80717885,
      "num_input_tokens_seen": 99354185,
      "step": 4602,
      "time_per_iteration": 2.5461394786834717
    },
    {
      "auxiliary_loss_clip": 0.01137619,
      "auxiliary_loss_mlp": 0.01027113,
      "balance_loss_clip": 1.04837298,
      "balance_loss_mlp": 1.02017486,
      "epoch": 0.5534780256117358,
      "flos": 32306496076800.0,
      "grad_norm": 1.7866157117209083,
      "language_loss": 0.75305021,
      "learning_rate": 1.751373934173777e-06,
      "loss": 0.77469754,
      "num_input_tokens_seen": 99376930,
      "step": 4603,
      "time_per_iteration": 2.657331943511963
    },
    {
      "auxiliary_loss_clip": 0.01169563,
      "auxiliary_loss_mlp": 0.0103213,
      "balance_loss_clip": 1.04825735,
      "balance_loss_mlp": 1.02441132,
      "epoch": 0.5535982685023748,
      "flos": 23222425582080.0,
      "grad_norm": 1.602646427351878,
      "language_loss": 0.72952628,
      "learning_rate": 1.750601023848487e-06,
      "loss": 0.75154316,
      "num_input_tokens_seen": 99397655,
      "step": 4604,
      "time_per_iteration": 2.5450479984283447
    },
    {
      "auxiliary_loss_clip": 0.01169733,
      "auxiliary_loss_mlp": 0.00760857,
      "balance_loss_clip": 1.05099392,
      "balance_loss_mlp": 1.00021529,
      "epoch": 0.5537185113930139,
      "flos": 24352534258560.0,
      "grad_norm": 1.7861673179586828,
      "language_loss": 0.73968458,
      "learning_rate": 1.749828151356823e-06,
      "loss": 0.75899047,
      "num_input_tokens_seen": 99417850,
      "step": 4605,
      "time_per_iteration": 2.486889123916626
    },
    {
      "auxiliary_loss_clip": 0.01144102,
      "auxiliary_loss_mlp": 0.01024012,
      "balance_loss_clip": 1.04735708,
      "balance_loss_mlp": 1.01693368,
      "epoch": 0.553838754283653,
      "flos": 23549068886400.0,
      "grad_norm": 2.1854740239586605,
      "language_loss": 0.75353616,
      "learning_rate": 1.7490553168160297e-06,
      "loss": 0.77521729,
      "num_input_tokens_seen": 99438920,
      "step": 4606,
      "time_per_iteration": 3.258605718612671
    },
    {
      "auxiliary_loss_clip": 0.01140971,
      "auxiliary_loss_mlp": 0.01026212,
      "balance_loss_clip": 1.04708433,
      "balance_loss_mlp": 1.01924133,
      "epoch": 0.5539589971742921,
      "flos": 17275044205440.0,
      "grad_norm": 1.906224535031859,
      "language_loss": 0.76243055,
      "learning_rate": 1.748282520343345e-06,
      "loss": 0.78410238,
      "num_input_tokens_seen": 99457950,
      "step": 4607,
      "time_per_iteration": 2.523167848587036
    },
    {
      "auxiliary_loss_clip": 0.01162751,
      "auxiliary_loss_mlp": 0.01029503,
      "balance_loss_clip": 1.04967904,
      "balance_loss_mlp": 1.02179015,
      "epoch": 0.5540792400649311,
      "flos": 27564169104000.0,
      "grad_norm": 1.9139045386992566,
      "language_loss": 0.78512293,
      "learning_rate": 1.7475097620560023e-06,
      "loss": 0.80704546,
      "num_input_tokens_seen": 99478015,
      "step": 4608,
      "time_per_iteration": 2.544057607650757
    },
    {
      "auxiliary_loss_clip": 0.01168974,
      "auxiliary_loss_mlp": 0.01024899,
      "balance_loss_clip": 1.04892743,
      "balance_loss_mlp": 1.01739728,
      "epoch": 0.5541994829555702,
      "flos": 23878657105920.0,
      "grad_norm": 1.766375373592864,
      "language_loss": 0.70981735,
      "learning_rate": 1.746737042071228e-06,
      "loss": 0.73175609,
      "num_input_tokens_seen": 99496520,
      "step": 4609,
      "time_per_iteration": 2.4824166297912598
    },
    {
      "auxiliary_loss_clip": 0.01139466,
      "auxiliary_loss_mlp": 0.01027399,
      "balance_loss_clip": 1.04861856,
      "balance_loss_mlp": 1.02054739,
      "epoch": 0.5543197258462094,
      "flos": 20115721342080.0,
      "grad_norm": 1.7818094951726855,
      "language_loss": 0.78971183,
      "learning_rate": 1.7459643605062424e-06,
      "loss": 0.81138045,
      "num_input_tokens_seen": 99513780,
      "step": 4610,
      "time_per_iteration": 2.528555154800415
    },
    {
      "auxiliary_loss_clip": 0.01110759,
      "auxiliary_loss_mlp": 0.0102397,
      "balance_loss_clip": 1.04478633,
      "balance_loss_mlp": 1.01626277,
      "epoch": 0.5544399687368484,
      "flos": 20916565021440.0,
      "grad_norm": 2.5639698721382285,
      "language_loss": 0.80443585,
      "learning_rate": 1.745191717478262e-06,
      "loss": 0.82578313,
      "num_input_tokens_seen": 99532360,
      "step": 4611,
      "time_per_iteration": 2.583462715148926
    },
    {
      "auxiliary_loss_clip": 0.011365,
      "auxiliary_loss_mlp": 0.01027582,
      "balance_loss_clip": 1.04670727,
      "balance_loss_mlp": 1.02034855,
      "epoch": 0.5545602116274875,
      "flos": 25518661297920.0,
      "grad_norm": 1.6110318002346469,
      "language_loss": 0.79540324,
      "learning_rate": 1.7444191131044948e-06,
      "loss": 0.81704402,
      "num_input_tokens_seen": 99552635,
      "step": 4612,
      "time_per_iteration": 3.362720012664795
    },
    {
      "auxiliary_loss_clip": 0.01142719,
      "auxiliary_loss_mlp": 0.01030521,
      "balance_loss_clip": 1.04887104,
      "balance_loss_mlp": 1.02287114,
      "epoch": 0.5546804545181266,
      "flos": 20995568985600.0,
      "grad_norm": 2.0717871138498727,
      "language_loss": 0.72880232,
      "learning_rate": 1.7436465475021456e-06,
      "loss": 0.75053477,
      "num_input_tokens_seen": 99572685,
      "step": 4613,
      "time_per_iteration": 2.533323287963867
    },
    {
      "auxiliary_loss_clip": 0.01121071,
      "auxiliary_loss_mlp": 0.01025022,
      "balance_loss_clip": 1.04528475,
      "balance_loss_mlp": 1.01785493,
      "epoch": 0.5548006974087657,
      "flos": 26833638297600.0,
      "grad_norm": 1.8864294852591454,
      "language_loss": 0.71657252,
      "learning_rate": 1.7428740207884111e-06,
      "loss": 0.73803353,
      "num_input_tokens_seen": 99593565,
      "step": 4614,
      "time_per_iteration": 3.3799476623535156
    },
    {
      "auxiliary_loss_clip": 0.01113399,
      "auxiliary_loss_mlp": 0.01028249,
      "balance_loss_clip": 1.04592609,
      "balance_loss_mlp": 1.02073312,
      "epoch": 0.5549209402994048,
      "flos": 33656414031360.0,
      "grad_norm": 1.8150752715832288,
      "language_loss": 0.60730755,
      "learning_rate": 1.7421015330804833e-06,
      "loss": 0.6287241,
      "num_input_tokens_seen": 99613485,
      "step": 4615,
      "time_per_iteration": 3.5030436515808105
    },
    {
      "auxiliary_loss_clip": 0.011691,
      "auxiliary_loss_mlp": 0.01023656,
      "balance_loss_clip": 1.0490458,
      "balance_loss_mlp": 1.01652694,
      "epoch": 0.5550411831900439,
      "flos": 23769524609280.0,
      "grad_norm": 1.9619490981396115,
      "language_loss": 0.72246337,
      "learning_rate": 1.7413290844955475e-06,
      "loss": 0.7443909,
      "num_input_tokens_seen": 99633515,
      "step": 4616,
      "time_per_iteration": 2.5132598876953125
    },
    {
      "auxiliary_loss_clip": 0.01147811,
      "auxiliary_loss_mlp": 0.01034777,
      "balance_loss_clip": 1.04841411,
      "balance_loss_mlp": 1.02743351,
      "epoch": 0.555161426080683,
      "flos": 21651189978240.0,
      "grad_norm": 2.487092648152689,
      "language_loss": 0.7791363,
      "learning_rate": 1.7405566751507843e-06,
      "loss": 0.80096221,
      "num_input_tokens_seen": 99651560,
      "step": 4617,
      "time_per_iteration": 2.5018117427825928
    },
    {
      "auxiliary_loss_clip": 0.01126315,
      "auxiliary_loss_mlp": 0.01021726,
      "balance_loss_clip": 1.04531455,
      "balance_loss_mlp": 1.01446605,
      "epoch": 0.555281668971322,
      "flos": 49563116605440.0,
      "grad_norm": 1.9796996779566516,
      "language_loss": 0.67380661,
      "learning_rate": 1.7397843051633668e-06,
      "loss": 0.69528699,
      "num_input_tokens_seen": 99674255,
      "step": 4618,
      "time_per_iteration": 2.8374552726745605
    },
    {
      "auxiliary_loss_clip": 0.01151864,
      "auxiliary_loss_mlp": 0.01029583,
      "balance_loss_clip": 1.04822016,
      "balance_loss_mlp": 1.02189994,
      "epoch": 0.5554019118619612,
      "flos": 20741608851840.0,
      "grad_norm": 1.591226271172385,
      "language_loss": 0.71194839,
      "learning_rate": 1.739011974650464e-06,
      "loss": 0.73376286,
      "num_input_tokens_seen": 99693585,
      "step": 4619,
      "time_per_iteration": 2.531893014907837
    },
    {
      "auxiliary_loss_clip": 0.01119378,
      "auxiliary_loss_mlp": 0.01024682,
      "balance_loss_clip": 1.04598963,
      "balance_loss_mlp": 1.01673651,
      "epoch": 0.5555221547526003,
      "flos": 25483217552640.0,
      "grad_norm": 2.336783754261743,
      "language_loss": 0.76673329,
      "learning_rate": 1.7382396837292365e-06,
      "loss": 0.78817391,
      "num_input_tokens_seen": 99714045,
      "step": 4620,
      "time_per_iteration": 2.672349691390991
    },
    {
      "auxiliary_loss_clip": 0.01171932,
      "auxiliary_loss_mlp": 0.0103461,
      "balance_loss_clip": 1.05094671,
      "balance_loss_mlp": 1.02641439,
      "epoch": 0.5556423976432393,
      "flos": 21762513204480.0,
      "grad_norm": 1.6931209063671189,
      "language_loss": 0.73431176,
      "learning_rate": 1.737467432516841e-06,
      "loss": 0.75637716,
      "num_input_tokens_seen": 99734145,
      "step": 4621,
      "time_per_iteration": 2.481193780899048
    },
    {
      "auxiliary_loss_clip": 0.01141303,
      "auxiliary_loss_mlp": 0.01027502,
      "balance_loss_clip": 1.04395831,
      "balance_loss_mlp": 1.02012885,
      "epoch": 0.5557626405338785,
      "flos": 24900171989760.0,
      "grad_norm": 2.4136016985198823,
      "language_loss": 0.74212861,
      "learning_rate": 1.7366952211304274e-06,
      "loss": 0.76381671,
      "num_input_tokens_seen": 99751990,
      "step": 4622,
      "time_per_iteration": 2.54645037651062
    },
    {
      "auxiliary_loss_clip": 0.01134149,
      "auxiliary_loss_mlp": 0.01035514,
      "balance_loss_clip": 1.04438722,
      "balance_loss_mlp": 1.02743745,
      "epoch": 0.5558828834245175,
      "flos": 18697501676160.0,
      "grad_norm": 2.0203340606531617,
      "language_loss": 0.83093345,
      "learning_rate": 1.735923049687139e-06,
      "loss": 0.85263014,
      "num_input_tokens_seen": 99768565,
      "step": 4623,
      "time_per_iteration": 2.492771863937378
    },
    {
      "auxiliary_loss_clip": 0.01137106,
      "auxiliary_loss_mlp": 0.01026561,
      "balance_loss_clip": 1.04442549,
      "balance_loss_mlp": 1.01976657,
      "epoch": 0.5560031263151566,
      "flos": 27272179445760.0,
      "grad_norm": 1.5482846079620536,
      "language_loss": 0.73926288,
      "learning_rate": 1.7351509183041144e-06,
      "loss": 0.76089954,
      "num_input_tokens_seen": 99788895,
      "step": 4624,
      "time_per_iteration": 2.728402614593506
    },
    {
      "auxiliary_loss_clip": 0.01174557,
      "auxiliary_loss_mlp": 0.01036143,
      "balance_loss_clip": 1.05316412,
      "balance_loss_mlp": 1.02815592,
      "epoch": 0.5561233692057957,
      "flos": 23403738458880.0,
      "grad_norm": 1.726591346670488,
      "language_loss": 0.71410543,
      "learning_rate": 1.7343788270984852e-06,
      "loss": 0.73621237,
      "num_input_tokens_seen": 99808035,
      "step": 4625,
      "time_per_iteration": 2.4910695552825928
    },
    {
      "auxiliary_loss_clip": 0.0114347,
      "auxiliary_loss_mlp": 0.01034033,
      "balance_loss_clip": 1.05069256,
      "balance_loss_mlp": 1.02619171,
      "epoch": 0.5562436120964348,
      "flos": 37670867804160.0,
      "grad_norm": 1.7955759704205314,
      "language_loss": 0.74680638,
      "learning_rate": 1.7336067761873764e-06,
      "loss": 0.76858133,
      "num_input_tokens_seen": 99830460,
      "step": 4626,
      "time_per_iteration": 2.667536735534668
    },
    {
      "auxiliary_loss_clip": 0.01160114,
      "auxiliary_loss_mlp": 0.01031012,
      "balance_loss_clip": 1.04764056,
      "balance_loss_mlp": 1.02331161,
      "epoch": 0.5563638549870739,
      "flos": 25155245445120.0,
      "grad_norm": 2.151416586060699,
      "language_loss": 0.76120073,
      "learning_rate": 1.7328347656879076e-06,
      "loss": 0.78311199,
      "num_input_tokens_seen": 99850320,
      "step": 4627,
      "time_per_iteration": 2.515690565109253
    },
    {
      "auxiliary_loss_clip": 0.01126618,
      "auxiliary_loss_mlp": 0.01027623,
      "balance_loss_clip": 1.04442513,
      "balance_loss_mlp": 1.02022576,
      "epoch": 0.556484097877713,
      "flos": 13581810783360.0,
      "grad_norm": 2.2245665836225625,
      "language_loss": 0.68404913,
      "learning_rate": 1.7320627957171927e-06,
      "loss": 0.70559156,
      "num_input_tokens_seen": 99864980,
      "step": 4628,
      "time_per_iteration": 2.5076446533203125
    },
    {
      "auxiliary_loss_clip": 0.01171391,
      "auxiliary_loss_mlp": 0.01025021,
      "balance_loss_clip": 1.05173492,
      "balance_loss_mlp": 1.01766586,
      "epoch": 0.5566043407683521,
      "flos": 24681368292480.0,
      "grad_norm": 1.7752222058617542,
      "language_loss": 0.81730032,
      "learning_rate": 1.7312908663923382e-06,
      "loss": 0.83926451,
      "num_input_tokens_seen": 99881155,
      "step": 4629,
      "time_per_iteration": 2.478086233139038
    },
    {
      "auxiliary_loss_clip": 0.01151248,
      "auxiliary_loss_mlp": 0.01024616,
      "balance_loss_clip": 1.04657269,
      "balance_loss_mlp": 1.0169208,
      "epoch": 0.5567245836589911,
      "flos": 20588161950720.0,
      "grad_norm": 1.952799282367499,
      "language_loss": 0.67403734,
      "learning_rate": 1.7305189778304463e-06,
      "loss": 0.69579601,
      "num_input_tokens_seen": 99899330,
      "step": 4630,
      "time_per_iteration": 2.4783334732055664
    },
    {
      "auxiliary_loss_clip": 0.01147153,
      "auxiliary_loss_mlp": 0.01026958,
      "balance_loss_clip": 1.0527972,
      "balance_loss_mlp": 1.01954913,
      "epoch": 0.5568448265496303,
      "flos": 20704189858560.0,
      "grad_norm": 1.9646916834745636,
      "language_loss": 0.79886448,
      "learning_rate": 1.729747130148611e-06,
      "loss": 0.82060564,
      "num_input_tokens_seen": 99918525,
      "step": 4631,
      "time_per_iteration": 2.507913112640381
    },
    {
      "auxiliary_loss_clip": 0.01138027,
      "auxiliary_loss_mlp": 0.01030449,
      "balance_loss_clip": 1.04885781,
      "balance_loss_mlp": 1.02249801,
      "epoch": 0.5569650694402694,
      "flos": 25302910256640.0,
      "grad_norm": 2.660200901257454,
      "language_loss": 0.77180135,
      "learning_rate": 1.7289753234639208e-06,
      "loss": 0.79348612,
      "num_input_tokens_seen": 99937500,
      "step": 4632,
      "time_per_iteration": 3.3080809116363525
    },
    {
      "auxiliary_loss_clip": 0.01163526,
      "auxiliary_loss_mlp": 0.01027466,
      "balance_loss_clip": 1.05087543,
      "balance_loss_mlp": 1.01998532,
      "epoch": 0.5570853123309084,
      "flos": 19712623939200.0,
      "grad_norm": 2.0870008309430608,
      "language_loss": 0.76528943,
      "learning_rate": 1.7282035578934592e-06,
      "loss": 0.78719938,
      "num_input_tokens_seen": 99955665,
      "step": 4633,
      "time_per_iteration": 2.475905418395996
    },
    {
      "auxiliary_loss_clip": 0.01137961,
      "auxiliary_loss_mlp": 0.01028912,
      "balance_loss_clip": 1.05147576,
      "balance_loss_mlp": 1.02136946,
      "epoch": 0.5572055552215476,
      "flos": 16108091153280.0,
      "grad_norm": 1.6126106501923507,
      "language_loss": 0.79059958,
      "learning_rate": 1.727431833554301e-06,
      "loss": 0.81226838,
      "num_input_tokens_seen": 99974140,
      "step": 4634,
      "time_per_iteration": 2.4992637634277344
    },
    {
      "auxiliary_loss_clip": 0.0110609,
      "auxiliary_loss_mlp": 0.01023592,
      "balance_loss_clip": 1.04290271,
      "balance_loss_mlp": 1.01633215,
      "epoch": 0.5573257981121866,
      "flos": 17128815937920.0,
      "grad_norm": 1.8051671649210035,
      "language_loss": 0.77462441,
      "learning_rate": 1.7266601505635175e-06,
      "loss": 0.79592121,
      "num_input_tokens_seen": 99991480,
      "step": 4635,
      "time_per_iteration": 2.5944924354553223
    },
    {
      "auxiliary_loss_clip": 0.0115741,
      "auxiliary_loss_mlp": 0.01029751,
      "balance_loss_clip": 1.05051625,
      "balance_loss_mlp": 1.02156091,
      "epoch": 0.5574460410028257,
      "flos": 18807029222400.0,
      "grad_norm": 1.86951939446058,
      "language_loss": 0.75730181,
      "learning_rate": 1.7258885090381717e-06,
      "loss": 0.77917349,
      "num_input_tokens_seen": 100009520,
      "step": 4636,
      "time_per_iteration": 2.5282914638519287
    },
    {
      "auxiliary_loss_clip": 0.01145339,
      "auxiliary_loss_mlp": 0.01029249,
      "balance_loss_clip": 1.04603732,
      "balance_loss_mlp": 1.02178311,
      "epoch": 0.5575662838934649,
      "flos": 29642678530560.0,
      "grad_norm": 1.7286884013719412,
      "language_loss": 0.7826668,
      "learning_rate": 1.7251169090953213e-06,
      "loss": 0.80441272,
      "num_input_tokens_seen": 100029995,
      "step": 4637,
      "time_per_iteration": 2.5945987701416016
    },
    {
      "auxiliary_loss_clip": 0.01159705,
      "auxiliary_loss_mlp": 0.01027374,
      "balance_loss_clip": 1.04970825,
      "balance_loss_mlp": 1.01973248,
      "epoch": 0.5576865267841039,
      "flos": 22054466949120.0,
      "grad_norm": 2.4298655800550066,
      "language_loss": 0.76497126,
      "learning_rate": 1.7243453508520168e-06,
      "loss": 0.78684205,
      "num_input_tokens_seen": 100046980,
      "step": 4638,
      "time_per_iteration": 3.2878684997558594
    },
    {
      "auxiliary_loss_clip": 0.01141295,
      "auxiliary_loss_mlp": 0.0102628,
      "balance_loss_clip": 1.04538846,
      "balance_loss_mlp": 1.01904118,
      "epoch": 0.557806769674743,
      "flos": 17196040241280.0,
      "grad_norm": 2.0102914020847393,
      "language_loss": 0.84473485,
      "learning_rate": 1.7235738344253038e-06,
      "loss": 0.86641061,
      "num_input_tokens_seen": 100060610,
      "step": 4639,
      "time_per_iteration": 2.490902900695801
    },
    {
      "auxiliary_loss_clip": 0.01153572,
      "auxiliary_loss_mlp": 0.01033916,
      "balance_loss_clip": 1.0488323,
      "balance_loss_mlp": 1.02585721,
      "epoch": 0.557927012565382,
      "flos": 24712717887360.0,
      "grad_norm": 1.8036252211584771,
      "language_loss": 0.82696807,
      "learning_rate": 1.72280235993222e-06,
      "loss": 0.84884298,
      "num_input_tokens_seen": 100078915,
      "step": 4640,
      "time_per_iteration": 3.307020902633667
    },
    {
      "auxiliary_loss_clip": 0.01154933,
      "auxiliary_loss_mlp": 0.0076236,
      "balance_loss_clip": 1.0498383,
      "balance_loss_mlp": 1.00022531,
      "epoch": 0.5580472554560212,
      "flos": 16983090460800.0,
      "grad_norm": 2.012351276166846,
      "language_loss": 0.69446814,
      "learning_rate": 1.722030927489798e-06,
      "loss": 0.71364105,
      "num_input_tokens_seen": 100096195,
      "step": 4641,
      "time_per_iteration": 3.2306628227233887
    },
    {
      "auxiliary_loss_clip": 0.01133076,
      "auxiliary_loss_mlp": 0.01027385,
      "balance_loss_clip": 1.04918349,
      "balance_loss_mlp": 1.0192194,
      "epoch": 0.5581674983466602,
      "flos": 23509100027520.0,
      "grad_norm": 1.7136689339715356,
      "language_loss": 0.74278772,
      "learning_rate": 1.7212595372150634e-06,
      "loss": 0.76439238,
      "num_input_tokens_seen": 100116175,
      "step": 4642,
      "time_per_iteration": 2.5891661643981934
    },
    {
      "auxiliary_loss_clip": 0.01171721,
      "auxiliary_loss_mlp": 0.01028491,
      "balance_loss_clip": 1.05138588,
      "balance_loss_mlp": 1.02092671,
      "epoch": 0.5582877412372993,
      "flos": 13480291969920.0,
      "grad_norm": 2.5235358787036697,
      "language_loss": 0.72913927,
      "learning_rate": 1.720488189225035e-06,
      "loss": 0.75114143,
      "num_input_tokens_seen": 100133875,
      "step": 4643,
      "time_per_iteration": 2.439209461212158
    },
    {
      "auxiliary_loss_clip": 0.01161729,
      "auxiliary_loss_mlp": 0.01027017,
      "balance_loss_clip": 1.05036283,
      "balance_loss_mlp": 1.01920843,
      "epoch": 0.5584079841279385,
      "flos": 21903605827200.0,
      "grad_norm": 3.011949043129492,
      "language_loss": 0.79173553,
      "learning_rate": 1.7197168836367265e-06,
      "loss": 0.81362301,
      "num_input_tokens_seen": 100150685,
      "step": 4644,
      "time_per_iteration": 2.476140022277832
    },
    {
      "auxiliary_loss_clip": 0.01156935,
      "auxiliary_loss_mlp": 0.00761311,
      "balance_loss_clip": 1.04950452,
      "balance_loss_mlp": 1.00020027,
      "epoch": 0.5585282270185775,
      "flos": 18843550375680.0,
      "grad_norm": 1.890025443949501,
      "language_loss": 0.81859171,
      "learning_rate": 1.7189456205671433e-06,
      "loss": 0.83777416,
      "num_input_tokens_seen": 100169530,
      "step": 4645,
      "time_per_iteration": 2.4948794841766357
    },
    {
      "auxiliary_loss_clip": 0.01162683,
      "auxiliary_loss_mlp": 0.01027867,
      "balance_loss_clip": 1.05040622,
      "balance_loss_mlp": 1.02051759,
      "epoch": 0.5586484699092166,
      "flos": 21868449390720.0,
      "grad_norm": 1.9528033457532326,
      "language_loss": 0.82643896,
      "learning_rate": 1.7181744001332866e-06,
      "loss": 0.84834445,
      "num_input_tokens_seen": 100188140,
      "step": 4646,
      "time_per_iteration": 2.480687141418457
    },
    {
      "auxiliary_loss_clip": 0.01170579,
      "auxiliary_loss_mlp": 0.01030355,
      "balance_loss_clip": 1.0527662,
      "balance_loss_mlp": 1.02317238,
      "epoch": 0.5587687127998557,
      "flos": 22893232412160.0,
      "grad_norm": 2.100016970917802,
      "language_loss": 0.63202649,
      "learning_rate": 1.7174032224521493e-06,
      "loss": 0.65403581,
      "num_input_tokens_seen": 100206850,
      "step": 4647,
      "time_per_iteration": 2.4768612384796143
    },
    {
      "auxiliary_loss_clip": 0.01157009,
      "auxiliary_loss_mlp": 0.01025549,
      "balance_loss_clip": 1.04927123,
      "balance_loss_mlp": 1.01818824,
      "epoch": 0.5588889556904948,
      "flos": 20303067703680.0,
      "grad_norm": 1.6372271747489948,
      "language_loss": 0.69833201,
      "learning_rate": 1.7166320876407184e-06,
      "loss": 0.72015762,
      "num_input_tokens_seen": 100226270,
      "step": 4648,
      "time_per_iteration": 2.4930996894836426
    },
    {
      "auxiliary_loss_clip": 0.01173969,
      "auxiliary_loss_mlp": 0.00761918,
      "balance_loss_clip": 1.05153513,
      "balance_loss_mlp": 1.00022447,
      "epoch": 0.5590091985811338,
      "flos": 16472153450880.0,
      "grad_norm": 1.9496353978249683,
      "language_loss": 0.67954135,
      "learning_rate": 1.7158609958159742e-06,
      "loss": 0.69890022,
      "num_input_tokens_seen": 100243675,
      "step": 4649,
      "time_per_iteration": 2.442575693130493
    },
    {
      "auxiliary_loss_clip": 0.01106812,
      "auxiliary_loss_mlp": 0.0103034,
      "balance_loss_clip": 1.04320431,
      "balance_loss_mlp": 1.02283001,
      "epoch": 0.559129441471773,
      "flos": 14532186781440.0,
      "grad_norm": 37.40346569455095,
      "language_loss": 0.78086346,
      "learning_rate": 1.7150899470948911e-06,
      "loss": 0.80223501,
      "num_input_tokens_seen": 100258940,
      "step": 4650,
      "time_per_iteration": 2.5857198238372803
    },
    {
      "auxiliary_loss_clip": 0.01043183,
      "auxiliary_loss_mlp": 0.01003124,
      "balance_loss_clip": 1.01499832,
      "balance_loss_mlp": 1.00198579,
      "epoch": 0.5592496843624121,
      "flos": 60521009852160.0,
      "grad_norm": 0.8264820647344485,
      "language_loss": 0.56631988,
      "learning_rate": 1.7143189415944365e-06,
      "loss": 0.58678299,
      "num_input_tokens_seen": 100323400,
      "step": 4651,
      "time_per_iteration": 3.1711440086364746
    },
    {
      "auxiliary_loss_clip": 0.01159748,
      "auxiliary_loss_mlp": 0.01029028,
      "balance_loss_clip": 1.05241942,
      "balance_loss_mlp": 1.0213449,
      "epoch": 0.5593699272530511,
      "flos": 20886256920960.0,
      "grad_norm": 1.580686038357926,
      "language_loss": 0.76416612,
      "learning_rate": 1.7135479794315714e-06,
      "loss": 0.78605384,
      "num_input_tokens_seen": 100340355,
      "step": 4652,
      "time_per_iteration": 2.4965269565582275
    },
    {
      "auxiliary_loss_clip": 0.01128198,
      "auxiliary_loss_mlp": 0.01028664,
      "balance_loss_clip": 1.04696703,
      "balance_loss_mlp": 1.02116561,
      "epoch": 0.5594901701436903,
      "flos": 12896743616640.0,
      "grad_norm": 1.879331303285908,
      "language_loss": 0.79193711,
      "learning_rate": 1.7127770607232502e-06,
      "loss": 0.81350571,
      "num_input_tokens_seen": 100358900,
      "step": 4653,
      "time_per_iteration": 2.5659799575805664
    },
    {
      "auxiliary_loss_clip": 0.01135956,
      "auxiliary_loss_mlp": 0.01028448,
      "balance_loss_clip": 1.04701543,
      "balance_loss_mlp": 1.02122068,
      "epoch": 0.5596104130343293,
      "flos": 23112107936640.0,
      "grad_norm": 1.7552983604592889,
      "language_loss": 0.79951197,
      "learning_rate": 1.7120061855864204e-06,
      "loss": 0.82115602,
      "num_input_tokens_seen": 100378910,
      "step": 4654,
      "time_per_iteration": 2.577087163925171
    },
    {
      "auxiliary_loss_clip": 0.01159114,
      "auxiliary_loss_mlp": 0.0102742,
      "balance_loss_clip": 1.05184495,
      "balance_loss_mlp": 1.01936781,
      "epoch": 0.5597306559249684,
      "flos": 25957812977280.0,
      "grad_norm": 17.660739840368322,
      "language_loss": 0.70764959,
      "learning_rate": 1.7112353541380233e-06,
      "loss": 0.72951496,
      "num_input_tokens_seen": 100398770,
      "step": 4655,
      "time_per_iteration": 2.552063465118408
    },
    {
      "auxiliary_loss_clip": 0.01146349,
      "auxiliary_loss_mlp": 0.01037349,
      "balance_loss_clip": 1.05060923,
      "balance_loss_mlp": 1.02933216,
      "epoch": 0.5598508988156076,
      "flos": 22492289825280.0,
      "grad_norm": 1.404532588466725,
      "language_loss": 0.71928096,
      "learning_rate": 1.7104645664949931e-06,
      "loss": 0.74111801,
      "num_input_tokens_seen": 100421240,
      "step": 4656,
      "time_per_iteration": 2.577639102935791
    },
    {
      "auxiliary_loss_clip": 0.0114464,
      "auxiliary_loss_mlp": 0.01027891,
      "balance_loss_clip": 1.04690409,
      "balance_loss_mlp": 1.02058315,
      "epoch": 0.5599711417062466,
      "flos": 23112538899840.0,
      "grad_norm": 2.752699181863947,
      "language_loss": 0.71507049,
      "learning_rate": 1.7096938227742584e-06,
      "loss": 0.73679578,
      "num_input_tokens_seen": 100442370,
      "step": 4657,
      "time_per_iteration": 2.543750762939453
    },
    {
      "auxiliary_loss_clip": 0.01170865,
      "auxiliary_loss_mlp": 0.01027258,
      "balance_loss_clip": 1.05113864,
      "balance_loss_mlp": 1.01953018,
      "epoch": 0.5600913845968857,
      "flos": 22339345714560.0,
      "grad_norm": 2.193619622718537,
      "language_loss": 0.83700728,
      "learning_rate": 1.70892312309274e-06,
      "loss": 0.85898852,
      "num_input_tokens_seen": 100460260,
      "step": 4658,
      "time_per_iteration": 3.306978464126587
    },
    {
      "auxiliary_loss_clip": 0.01140039,
      "auxiliary_loss_mlp": 0.01025892,
      "balance_loss_clip": 1.04160523,
      "balance_loss_mlp": 1.01808417,
      "epoch": 0.5602116274875248,
      "flos": 17633791290240.0,
      "grad_norm": 2.142144520186431,
      "language_loss": 0.67997527,
      "learning_rate": 1.7081524675673523e-06,
      "loss": 0.70163465,
      "num_input_tokens_seen": 100475750,
      "step": 4659,
      "time_per_iteration": 2.504667282104492
    },
    {
      "auxiliary_loss_clip": 0.0104361,
      "auxiliary_loss_mlp": 0.01003097,
      "balance_loss_clip": 1.01433587,
      "balance_loss_mlp": 1.00202394,
      "epoch": 0.5603318703781639,
      "flos": 70115945529600.0,
      "grad_norm": 0.7695833117575557,
      "language_loss": 0.5958541,
      "learning_rate": 1.7073818563150026e-06,
      "loss": 0.61632115,
      "num_input_tokens_seen": 100537830,
      "step": 4660,
      "time_per_iteration": 3.2082507610321045
    },
    {
      "auxiliary_loss_clip": 0.0115285,
      "auxiliary_loss_mlp": 0.01031944,
      "balance_loss_clip": 1.04777515,
      "balance_loss_mlp": 1.023844,
      "epoch": 0.560452113268803,
      "flos": 18545850455040.0,
      "grad_norm": 2.059104674068944,
      "language_loss": 0.86711752,
      "learning_rate": 1.7066112894525935e-06,
      "loss": 0.88896549,
      "num_input_tokens_seen": 100555910,
      "step": 4661,
      "time_per_iteration": 2.4934444427490234
    },
    {
      "auxiliary_loss_clip": 0.01134284,
      "auxiliary_loss_mlp": 0.01033932,
      "balance_loss_clip": 1.04505026,
      "balance_loss_mlp": 1.02596331,
      "epoch": 0.5605723561594421,
      "flos": 25264665250560.0,
      "grad_norm": 1.5892462704764865,
      "language_loss": 0.7273941,
      "learning_rate": 1.7058407670970177e-06,
      "loss": 0.74907625,
      "num_input_tokens_seen": 100577385,
      "step": 4662,
      "time_per_iteration": 2.5945751667022705
    },
    {
      "auxiliary_loss_clip": 0.01159903,
      "auxiliary_loss_mlp": 0.01032872,
      "balance_loss_clip": 1.04984999,
      "balance_loss_mlp": 1.02537966,
      "epoch": 0.5606925990500812,
      "flos": 20594949621120.0,
      "grad_norm": 2.0886543267470707,
      "language_loss": 0.61353195,
      "learning_rate": 1.7050702893651643e-06,
      "loss": 0.63545966,
      "num_input_tokens_seen": 100596965,
      "step": 4663,
      "time_per_iteration": 2.4823267459869385
    },
    {
      "auxiliary_loss_clip": 0.01157207,
      "auxiliary_loss_mlp": 0.01027993,
      "balance_loss_clip": 1.0493108,
      "balance_loss_mlp": 1.02001786,
      "epoch": 0.5608128419407202,
      "flos": 35006044677120.0,
      "grad_norm": 2.1151450885914937,
      "language_loss": 0.75454581,
      "learning_rate": 1.7042998563739134e-06,
      "loss": 0.77639782,
      "num_input_tokens_seen": 100615315,
      "step": 4664,
      "time_per_iteration": 3.4269039630889893
    },
    {
      "auxiliary_loss_clip": 0.01152846,
      "auxiliary_loss_mlp": 0.01034973,
      "balance_loss_clip": 1.04965675,
      "balance_loss_mlp": 1.02622271,
      "epoch": 0.5609330848313594,
      "flos": 24639819235200.0,
      "grad_norm": 2.2004014635911253,
      "language_loss": 0.71598631,
      "learning_rate": 1.703529468240139e-06,
      "loss": 0.73786449,
      "num_input_tokens_seen": 100634185,
      "step": 4665,
      "time_per_iteration": 2.58793044090271
    },
    {
      "auxiliary_loss_clip": 0.01138424,
      "auxiliary_loss_mlp": 0.0102813,
      "balance_loss_clip": 1.04778934,
      "balance_loss_mlp": 1.02105474,
      "epoch": 0.5610533277219985,
      "flos": 18762894385920.0,
      "grad_norm": 2.1878507623987016,
      "language_loss": 0.73893267,
      "learning_rate": 1.7027591250807088e-06,
      "loss": 0.76059824,
      "num_input_tokens_seen": 100651360,
      "step": 4666,
      "time_per_iteration": 3.3178677558898926
    },
    {
      "auxiliary_loss_clip": 0.01173192,
      "auxiliary_loss_mlp": 0.0102744,
      "balance_loss_clip": 1.05208659,
      "balance_loss_mlp": 1.01908898,
      "epoch": 0.5611735706126375,
      "flos": 15012384727680.0,
      "grad_norm": 6.32096757296884,
      "language_loss": 0.85064149,
      "learning_rate": 1.7019888270124825e-06,
      "loss": 0.87264776,
      "num_input_tokens_seen": 100668525,
      "step": 4667,
      "time_per_iteration": 3.1835391521453857
    },
    {
      "auxiliary_loss_clip": 0.01162043,
      "auxiliary_loss_mlp": 0.01037877,
      "balance_loss_clip": 1.05240941,
      "balance_loss_mlp": 1.02993798,
      "epoch": 0.5612938135032767,
      "flos": 16468167041280.0,
      "grad_norm": 1.9448245754696594,
      "language_loss": 0.82294428,
      "learning_rate": 1.7012185741523147e-06,
      "loss": 0.84494352,
      "num_input_tokens_seen": 100684850,
      "step": 4668,
      "time_per_iteration": 2.4556496143341064
    },
    {
      "auxiliary_loss_clip": 0.01174698,
      "auxiliary_loss_mlp": 0.01027596,
      "balance_loss_clip": 1.05449259,
      "balance_loss_mlp": 1.01943636,
      "epoch": 0.5614140563939157,
      "flos": 25666433850240.0,
      "grad_norm": 1.925699945872736,
      "language_loss": 0.62518185,
      "learning_rate": 1.7004483666170514e-06,
      "loss": 0.64720476,
      "num_input_tokens_seen": 100705345,
      "step": 4669,
      "time_per_iteration": 2.5214765071868896
    },
    {
      "auxiliary_loss_clip": 0.01156078,
      "auxiliary_loss_mlp": 0.01025466,
      "balance_loss_clip": 1.04987502,
      "balance_loss_mlp": 1.01837254,
      "epoch": 0.5615342992845548,
      "flos": 24717566223360.0,
      "grad_norm": 2.369796405635374,
      "language_loss": 0.80598509,
      "learning_rate": 1.699678204523533e-06,
      "loss": 0.82780051,
      "num_input_tokens_seen": 100725210,
      "step": 4670,
      "time_per_iteration": 2.517056465148926
    },
    {
      "auxiliary_loss_clip": 0.01150588,
      "auxiliary_loss_mlp": 0.01033076,
      "balance_loss_clip": 1.05242991,
      "balance_loss_mlp": 1.02567315,
      "epoch": 0.5616545421751938,
      "flos": 22015934634240.0,
      "grad_norm": 2.6157707464290207,
      "language_loss": 0.68464273,
      "learning_rate": 1.6989080879885918e-06,
      "loss": 0.70647943,
      "num_input_tokens_seen": 100743070,
      "step": 4671,
      "time_per_iteration": 2.5485873222351074
    },
    {
      "auxiliary_loss_clip": 0.0103165,
      "auxiliary_loss_mlp": 0.01000972,
      "balance_loss_clip": 1.0126853,
      "balance_loss_mlp": 0.9997803,
      "epoch": 0.561774785065833,
      "flos": 53760358690560.0,
      "grad_norm": 0.9042835411606975,
      "language_loss": 0.61103189,
      "learning_rate": 1.6981380171290544e-06,
      "loss": 0.63135809,
      "num_input_tokens_seen": 100804095,
      "step": 4672,
      "time_per_iteration": 3.12965989112854
    },
    {
      "auxiliary_loss_clip": 0.01133956,
      "auxiliary_loss_mlp": 0.01032413,
      "balance_loss_clip": 1.04347491,
      "balance_loss_mlp": 1.02424669,
      "epoch": 0.5618950279564721,
      "flos": 19750007018880.0,
      "grad_norm": 1.8049872731158227,
      "language_loss": 0.74028748,
      "learning_rate": 1.6973679920617396e-06,
      "loss": 0.76195115,
      "num_input_tokens_seen": 100821630,
      "step": 4673,
      "time_per_iteration": 2.524620771408081
    },
    {
      "auxiliary_loss_clip": 0.01147641,
      "auxiliary_loss_mlp": 0.01025638,
      "balance_loss_clip": 1.05253696,
      "balance_loss_mlp": 1.01787126,
      "epoch": 0.5620152708471111,
      "flos": 16800592435200.0,
      "grad_norm": 4.597075982699051,
      "language_loss": 0.84780884,
      "learning_rate": 1.6965980129034603e-06,
      "loss": 0.86954165,
      "num_input_tokens_seen": 100839015,
      "step": 4674,
      "time_per_iteration": 2.5264580249786377
    },
    {
      "auxiliary_loss_clip": 0.01144914,
      "auxiliary_loss_mlp": 0.01026047,
      "balance_loss_clip": 1.05017066,
      "balance_loss_mlp": 1.01811361,
      "epoch": 0.5621355137377503,
      "flos": 26797799502720.0,
      "grad_norm": 2.6218738040127363,
      "language_loss": 0.76538324,
      "learning_rate": 1.6958280797710209e-06,
      "loss": 0.7870928,
      "num_input_tokens_seen": 100860940,
      "step": 4675,
      "time_per_iteration": 2.570274591445923
    },
    {
      "auxiliary_loss_clip": 0.01044453,
      "auxiliary_loss_mlp": 0.01002494,
      "balance_loss_clip": 1.0152514,
      "balance_loss_mlp": 1.00131404,
      "epoch": 0.5622557566283893,
      "flos": 61207046686080.0,
      "grad_norm": 0.7097308628567911,
      "language_loss": 0.54808426,
      "learning_rate": 1.6950581927812198e-06,
      "loss": 0.56855369,
      "num_input_tokens_seen": 100920510,
      "step": 4676,
      "time_per_iteration": 3.0174875259399414
    },
    {
      "auxiliary_loss_clip": 0.01155221,
      "auxiliary_loss_mlp": 0.01026443,
      "balance_loss_clip": 1.04787564,
      "balance_loss_mlp": 1.01925445,
      "epoch": 0.5623759995190284,
      "flos": 26468534505600.0,
      "grad_norm": 2.8683365770226614,
      "language_loss": 0.791179,
      "learning_rate": 1.6942883520508486e-06,
      "loss": 0.81299567,
      "num_input_tokens_seen": 100939245,
      "step": 4677,
      "time_per_iteration": 2.5699169635772705
    },
    {
      "auxiliary_loss_clip": 0.01155513,
      "auxiliary_loss_mlp": 0.01028539,
      "balance_loss_clip": 1.04934645,
      "balance_loss_mlp": 1.02168739,
      "epoch": 0.5624962424096676,
      "flos": 19390900798080.0,
      "grad_norm": 1.9714058886543886,
      "language_loss": 0.77566636,
      "learning_rate": 1.693518557696691e-06,
      "loss": 0.79750687,
      "num_input_tokens_seen": 100958385,
      "step": 4678,
      "time_per_iteration": 2.483856201171875
    },
    {
      "auxiliary_loss_clip": 0.01151801,
      "auxiliary_loss_mlp": 0.01023861,
      "balance_loss_clip": 1.04555488,
      "balance_loss_mlp": 1.01696146,
      "epoch": 0.5626164853003066,
      "flos": 20667345482880.0,
      "grad_norm": 2.364373708544208,
      "language_loss": 0.8901974,
      "learning_rate": 1.6927488098355252e-06,
      "loss": 0.91195399,
      "num_input_tokens_seen": 100976015,
      "step": 4679,
      "time_per_iteration": 2.474860668182373
    },
    {
      "auxiliary_loss_clip": 0.01024063,
      "auxiliary_loss_mlp": 0.01003328,
      "balance_loss_clip": 1.01409245,
      "balance_loss_mlp": 1.00210595,
      "epoch": 0.5627367281909457,
      "flos": 62766071665920.0,
      "grad_norm": 0.9166413937123754,
      "language_loss": 0.63206744,
      "learning_rate": 1.6919791085841201e-06,
      "loss": 0.65234125,
      "num_input_tokens_seen": 101033425,
      "step": 4680,
      "time_per_iteration": 3.1317784786224365
    },
    {
      "auxiliary_loss_clip": 0.01150049,
      "auxiliary_loss_mlp": 0.01027897,
      "balance_loss_clip": 1.04510307,
      "balance_loss_mlp": 1.02008843,
      "epoch": 0.5628569710815848,
      "flos": 12787144243200.0,
      "grad_norm": 2.0970192500471576,
      "language_loss": 0.78833318,
      "learning_rate": 1.6912094540592396e-06,
      "loss": 0.8101126,
      "num_input_tokens_seen": 101048945,
      "step": 4681,
      "time_per_iteration": 2.444520950317383
    },
    {
      "auxiliary_loss_clip": 0.01157512,
      "auxiliary_loss_mlp": 0.01029542,
      "balance_loss_clip": 1.05010509,
      "balance_loss_mlp": 1.02222276,
      "epoch": 0.5629772139722239,
      "flos": 13762082165760.0,
      "grad_norm": 2.50336754948972,
      "language_loss": 0.80706966,
      "learning_rate": 1.6904398463776393e-06,
      "loss": 0.82894015,
      "num_input_tokens_seen": 101062745,
      "step": 4682,
      "time_per_iteration": 2.4443960189819336
    },
    {
      "auxiliary_loss_clip": 0.01157223,
      "auxiliary_loss_mlp": 0.0102169,
      "balance_loss_clip": 1.04754293,
      "balance_loss_mlp": 1.01454926,
      "epoch": 0.5630974568628629,
      "flos": 21467830026240.0,
      "grad_norm": 1.7976092527850882,
      "language_loss": 0.72807276,
      "learning_rate": 1.6896702856560683e-06,
      "loss": 0.7498619,
      "num_input_tokens_seen": 101081840,
      "step": 4683,
      "time_per_iteration": 2.4879748821258545
    },
    {
      "auxiliary_loss_clip": 0.01125995,
      "auxiliary_loss_mlp": 0.01027316,
      "balance_loss_clip": 1.04219079,
      "balance_loss_mlp": 1.01965714,
      "epoch": 0.5632176997535021,
      "flos": 14245907385600.0,
      "grad_norm": 3.232454365918642,
      "language_loss": 0.69312018,
      "learning_rate": 1.6889007720112677e-06,
      "loss": 0.71465331,
      "num_input_tokens_seen": 101099585,
      "step": 4684,
      "time_per_iteration": 3.264641046524048
    },
    {
      "auxiliary_loss_clip": 0.01159789,
      "auxiliary_loss_mlp": 0.01023884,
      "balance_loss_clip": 1.04926109,
      "balance_loss_mlp": 1.01674628,
      "epoch": 0.5633379426441412,
      "flos": 20812244947200.0,
      "grad_norm": 1.6374168387131542,
      "language_loss": 0.7725004,
      "learning_rate": 1.6881313055599734e-06,
      "loss": 0.79433709,
      "num_input_tokens_seen": 101119515,
      "step": 4685,
      "time_per_iteration": 2.4857230186462402
    },
    {
      "auxiliary_loss_clip": 0.01128424,
      "auxiliary_loss_mlp": 0.01030252,
      "balance_loss_clip": 1.04269671,
      "balance_loss_mlp": 1.02269399,
      "epoch": 0.5634581855347802,
      "flos": 22600883617920.0,
      "grad_norm": 6.0398457423833465,
      "language_loss": 0.82155532,
      "learning_rate": 1.6873618864189117e-06,
      "loss": 0.84314209,
      "num_input_tokens_seen": 101135285,
      "step": 4686,
      "time_per_iteration": 2.50394606590271
    },
    {
      "auxiliary_loss_clip": 0.01157614,
      "auxiliary_loss_mlp": 0.01031941,
      "balance_loss_clip": 1.04932213,
      "balance_loss_mlp": 1.02451384,
      "epoch": 0.5635784284254194,
      "flos": 21506972872320.0,
      "grad_norm": 2.123408979510677,
      "language_loss": 0.78083766,
      "learning_rate": 1.686592514704803e-06,
      "loss": 0.80273318,
      "num_input_tokens_seen": 101152680,
      "step": 4687,
      "time_per_iteration": 2.488701343536377
    },
    {
      "auxiliary_loss_clip": 0.0114174,
      "auxiliary_loss_mlp": 0.0102681,
      "balance_loss_clip": 1.04765558,
      "balance_loss_mlp": 1.01969874,
      "epoch": 0.5636986713160584,
      "flos": 19827466698240.0,
      "grad_norm": 2.17357243903105,
      "language_loss": 0.70959914,
      "learning_rate": 1.685823190534361e-06,
      "loss": 0.73128468,
      "num_input_tokens_seen": 101170920,
      "step": 4688,
      "time_per_iteration": 2.5023059844970703
    },
    {
      "auxiliary_loss_clip": 0.01174388,
      "auxiliary_loss_mlp": 0.01033331,
      "balance_loss_clip": 1.05077767,
      "balance_loss_mlp": 1.02483082,
      "epoch": 0.5638189142066975,
      "flos": 19792453916160.0,
      "grad_norm": 1.745904522486701,
      "language_loss": 0.83401626,
      "learning_rate": 1.6850539140242907e-06,
      "loss": 0.85609341,
      "num_input_tokens_seen": 101190180,
      "step": 4689,
      "time_per_iteration": 2.4580090045928955
    },
    {
      "auxiliary_loss_clip": 0.01159014,
      "auxiliary_loss_mlp": 0.01024694,
      "balance_loss_clip": 1.04966068,
      "balance_loss_mlp": 1.01746416,
      "epoch": 0.5639391570973367,
      "flos": 22893771116160.0,
      "grad_norm": 1.7265505873506168,
      "language_loss": 0.81714904,
      "learning_rate": 1.684284685291292e-06,
      "loss": 0.83898616,
      "num_input_tokens_seen": 101211825,
      "step": 4690,
      "time_per_iteration": 3.2913572788238525
    },
    {
      "auxiliary_loss_clip": 0.01171071,
      "auxiliary_loss_mlp": 0.01029563,
      "balance_loss_clip": 1.05064762,
      "balance_loss_mlp": 1.02165985,
      "epoch": 0.5640593999879757,
      "flos": 23727077712000.0,
      "grad_norm": 2.2072397370150703,
      "language_loss": 0.81001627,
      "learning_rate": 1.683515504452055e-06,
      "loss": 0.83202261,
      "num_input_tokens_seen": 101229200,
      "step": 4691,
      "time_per_iteration": 2.479348659515381
    },
    {
      "auxiliary_loss_clip": 0.01120492,
      "auxiliary_loss_mlp": 0.0102963,
      "balance_loss_clip": 1.04355288,
      "balance_loss_mlp": 1.02139294,
      "epoch": 0.5641796428786148,
      "flos": 22710123855360.0,
      "grad_norm": 1.6090901078187176,
      "language_loss": 0.66300619,
      "learning_rate": 1.6827463716232648e-06,
      "loss": 0.68450743,
      "num_input_tokens_seen": 101249860,
      "step": 4692,
      "time_per_iteration": 3.3811419010162354
    },
    {
      "auxiliary_loss_clip": 0.01157977,
      "auxiliary_loss_mlp": 0.00761356,
      "balance_loss_clip": 1.04914451,
      "balance_loss_mlp": 1.00019836,
      "epoch": 0.5642998857692539,
      "flos": 19791987039360.0,
      "grad_norm": 1.7674524489475714,
      "language_loss": 0.75636649,
      "learning_rate": 1.6819772869215972e-06,
      "loss": 0.77555978,
      "num_input_tokens_seen": 101268940,
      "step": 4693,
      "time_per_iteration": 3.241947650909424
    },
    {
      "auxiliary_loss_clip": 0.01148239,
      "auxiliary_loss_mlp": 0.01026747,
      "balance_loss_clip": 1.04876781,
      "balance_loss_mlp": 1.02001429,
      "epoch": 0.564420128659893,
      "flos": 23185904428800.0,
      "grad_norm": 2.3893878759584766,
      "language_loss": 0.8230468,
      "learning_rate": 1.6812082504637228e-06,
      "loss": 0.84479666,
      "num_input_tokens_seen": 101290260,
      "step": 4694,
      "time_per_iteration": 2.5318498611450195
    },
    {
      "auxiliary_loss_clip": 0.01158178,
      "auxiliary_loss_mlp": 0.01029446,
      "balance_loss_clip": 1.05071712,
      "balance_loss_mlp": 1.022192,
      "epoch": 0.564540371550532,
      "flos": 23258264376960.0,
      "grad_norm": 1.4424786694870926,
      "language_loss": 0.74329853,
      "learning_rate": 1.6804392623663025e-06,
      "loss": 0.76517475,
      "num_input_tokens_seen": 101311465,
      "step": 4695,
      "time_per_iteration": 2.566643238067627
    },
    {
      "auxiliary_loss_clip": 0.01154172,
      "auxiliary_loss_mlp": 0.01027242,
      "balance_loss_clip": 1.04903102,
      "balance_loss_mlp": 1.02013755,
      "epoch": 0.5646606144411712,
      "flos": 25010058672000.0,
      "grad_norm": 1.7244124622414447,
      "language_loss": 0.78325689,
      "learning_rate": 1.6796703227459935e-06,
      "loss": 0.805071,
      "num_input_tokens_seen": 101329420,
      "step": 4696,
      "time_per_iteration": 2.52933669090271
    },
    {
      "auxiliary_loss_clip": 0.01106543,
      "auxiliary_loss_mlp": 0.01023202,
      "balance_loss_clip": 1.04043758,
      "balance_loss_mlp": 1.01587319,
      "epoch": 0.5647808573318103,
      "flos": 36539645806080.0,
      "grad_norm": 1.8135515673988654,
      "language_loss": 0.76151657,
      "learning_rate": 1.6789014317194407e-06,
      "loss": 0.78281403,
      "num_input_tokens_seen": 101350900,
      "step": 4697,
      "time_per_iteration": 2.7067673206329346
    },
    {
      "auxiliary_loss_clip": 0.01148271,
      "auxiliary_loss_mlp": 0.01026599,
      "balance_loss_clip": 1.04999113,
      "balance_loss_mlp": 1.01883268,
      "epoch": 0.5649011002224493,
      "flos": 22528451842560.0,
      "grad_norm": 2.066946898103636,
      "language_loss": 0.73208487,
      "learning_rate": 1.6781325894032853e-06,
      "loss": 0.75383353,
      "num_input_tokens_seen": 101369860,
      "step": 4698,
      "time_per_iteration": 2.5440471172332764
    },
    {
      "auxiliary_loss_clip": 0.01141142,
      "auxiliary_loss_mlp": 0.01028384,
      "balance_loss_clip": 1.04975247,
      "balance_loss_mlp": 1.02054,
      "epoch": 0.5650213431130885,
      "flos": 18515147304960.0,
      "grad_norm": 2.025519306826211,
      "language_loss": 0.91689956,
      "learning_rate": 1.6773637959141608e-06,
      "loss": 0.93859482,
      "num_input_tokens_seen": 101386835,
      "step": 4699,
      "time_per_iteration": 2.4736459255218506
    },
    {
      "auxiliary_loss_clip": 0.01134847,
      "auxiliary_loss_mlp": 0.01028046,
      "balance_loss_clip": 1.04553103,
      "balance_loss_mlp": 1.02065825,
      "epoch": 0.5651415860037275,
      "flos": 17526310819200.0,
      "grad_norm": 2.249312101779733,
      "language_loss": 0.66484773,
      "learning_rate": 1.6765950513686915e-06,
      "loss": 0.68647665,
      "num_input_tokens_seen": 101404945,
      "step": 4700,
      "time_per_iteration": 2.498988389968872
    },
    {
      "auxiliary_loss_clip": 0.0111461,
      "auxiliary_loss_mlp": 0.01029783,
      "balance_loss_clip": 1.0423075,
      "balance_loss_mlp": 1.0219276,
      "epoch": 0.5652618288943666,
      "flos": 25520026014720.0,
      "grad_norm": 1.655723079455022,
      "language_loss": 0.76202077,
      "learning_rate": 1.675826355883496e-06,
      "loss": 0.78346467,
      "num_input_tokens_seen": 101424160,
      "step": 4701,
      "time_per_iteration": 2.6138060092926025
    },
    {
      "auxiliary_loss_clip": 0.01141891,
      "auxiliary_loss_mlp": 0.01027356,
      "balance_loss_clip": 1.05114532,
      "balance_loss_mlp": 1.01929498,
      "epoch": 0.5653820717850057,
      "flos": 19683105937920.0,
      "grad_norm": 1.954112751204864,
      "language_loss": 0.79088557,
      "learning_rate": 1.6750577095751848e-06,
      "loss": 0.81257814,
      "num_input_tokens_seen": 101443270,
      "step": 4702,
      "time_per_iteration": 2.5127274990081787
    },
    {
      "auxiliary_loss_clip": 0.01170763,
      "auxiliary_loss_mlp": 0.01025843,
      "balance_loss_clip": 1.05073249,
      "balance_loss_mlp": 1.01848221,
      "epoch": 0.5655023146756448,
      "flos": 26979722910720.0,
      "grad_norm": 1.620227068573036,
      "language_loss": 0.72638381,
      "learning_rate": 1.6742891125603605e-06,
      "loss": 0.74834985,
      "num_input_tokens_seen": 101464175,
      "step": 4703,
      "time_per_iteration": 2.4916887283325195
    },
    {
      "auxiliary_loss_clip": 0.01158318,
      "auxiliary_loss_mlp": 0.01034976,
      "balance_loss_clip": 1.05114758,
      "balance_loss_mlp": 1.02666068,
      "epoch": 0.5656225575662839,
      "flos": 27669351104640.0,
      "grad_norm": 4.593682490544995,
      "language_loss": 0.71616763,
      "learning_rate": 1.6735205649556185e-06,
      "loss": 0.73810059,
      "num_input_tokens_seen": 101484045,
      "step": 4704,
      "time_per_iteration": 2.533649206161499
    },
    {
      "auxiliary_loss_clip": 0.01133596,
      "auxiliary_loss_mlp": 0.01032616,
      "balance_loss_clip": 1.0465107,
      "balance_loss_mlp": 1.02480805,
      "epoch": 0.5657428004569229,
      "flos": 24349732997760.0,
      "grad_norm": 1.644535686397196,
      "language_loss": 0.84582996,
      "learning_rate": 1.6727520668775476e-06,
      "loss": 0.86749202,
      "num_input_tokens_seen": 101504330,
      "step": 4705,
      "time_per_iteration": 2.571937084197998
    },
    {
      "auxiliary_loss_clip": 0.01172297,
      "auxiliary_loss_mlp": 0.01024949,
      "balance_loss_clip": 1.04983711,
      "balance_loss_mlp": 1.01759994,
      "epoch": 0.5658630433475621,
      "flos": 21944041562880.0,
      "grad_norm": 1.546911801957483,
      "language_loss": 0.75227195,
      "learning_rate": 1.6719836184427275e-06,
      "loss": 0.77424443,
      "num_input_tokens_seen": 101524635,
      "step": 4706,
      "time_per_iteration": 2.4802730083465576
    },
    {
      "auxiliary_loss_clip": 0.01141365,
      "auxiliary_loss_mlp": 0.01025931,
      "balance_loss_clip": 1.04625916,
      "balance_loss_mlp": 1.018296,
      "epoch": 0.5659832862382012,
      "flos": 30409012218240.0,
      "grad_norm": 1.630792929390597,
      "language_loss": 0.64228326,
      "learning_rate": 1.671215219767733e-06,
      "loss": 0.66395622,
      "num_input_tokens_seen": 101544095,
      "step": 4707,
      "time_per_iteration": 2.573749303817749
    },
    {
      "auxiliary_loss_clip": 0.01114366,
      "auxiliary_loss_mlp": 0.01027437,
      "balance_loss_clip": 1.0432303,
      "balance_loss_mlp": 1.02032065,
      "epoch": 0.5661035291288402,
      "flos": 13188194570880.0,
      "grad_norm": 1.9992210324585,
      "language_loss": 0.76020181,
      "learning_rate": 1.670446870969127e-06,
      "loss": 0.78161991,
      "num_input_tokens_seen": 101561760,
      "step": 4708,
      "time_per_iteration": 2.563624858856201
    },
    {
      "auxiliary_loss_clip": 0.01144444,
      "auxiliary_loss_mlp": 0.01031919,
      "balance_loss_clip": 1.04908228,
      "balance_loss_mlp": 1.0242722,
      "epoch": 0.5662237720194794,
      "flos": 16143032108160.0,
      "grad_norm": 2.1230560313472338,
      "language_loss": 0.79611075,
      "learning_rate": 1.6696785721634685e-06,
      "loss": 0.81787443,
      "num_input_tokens_seen": 101576245,
      "step": 4709,
      "time_per_iteration": 3.206770896911621
    },
    {
      "auxiliary_loss_clip": 0.01159654,
      "auxiliary_loss_mlp": 0.01027073,
      "balance_loss_clip": 1.0497359,
      "balance_loss_mlp": 1.01961601,
      "epoch": 0.5663440149101184,
      "flos": 17676848718720.0,
      "grad_norm": 1.811617341418638,
      "language_loss": 0.73536932,
      "learning_rate": 1.6689103234673086e-06,
      "loss": 0.7572366,
      "num_input_tokens_seen": 101594565,
      "step": 4710,
      "time_per_iteration": 2.4857959747314453
    },
    {
      "auxiliary_loss_clip": 0.01143083,
      "auxiliary_loss_mlp": 0.01026517,
      "balance_loss_clip": 1.05028582,
      "balance_loss_mlp": 1.01885736,
      "epoch": 0.5664642578007575,
      "flos": 23368330627200.0,
      "grad_norm": 1.7944437781282772,
      "language_loss": 0.76717889,
      "learning_rate": 1.668142124997189e-06,
      "loss": 0.78887486,
      "num_input_tokens_seen": 101614225,
      "step": 4711,
      "time_per_iteration": 2.558715343475342
    },
    {
      "auxiliary_loss_clip": 0.01034967,
      "auxiliary_loss_mlp": 0.0100403,
      "balance_loss_clip": 1.01378369,
      "balance_loss_mlp": 1.00290382,
      "epoch": 0.5665845006913967,
      "flos": 65516470945920.0,
      "grad_norm": 0.7239260722621347,
      "language_loss": 0.59804755,
      "learning_rate": 1.6673739768696453e-06,
      "loss": 0.61843753,
      "num_input_tokens_seen": 101680795,
      "step": 4712,
      "time_per_iteration": 3.140430450439453
    },
    {
      "auxiliary_loss_clip": 0.0114787,
      "auxiliary_loss_mlp": 0.01036949,
      "balance_loss_clip": 1.04599798,
      "balance_loss_mlp": 1.02926016,
      "epoch": 0.5667047435820357,
      "flos": 26140885620480.0,
      "grad_norm": 1.8939920823505374,
      "language_loss": 0.77390194,
      "learning_rate": 1.6666058792012052e-06,
      "loss": 0.79575014,
      "num_input_tokens_seen": 101701680,
      "step": 4713,
      "time_per_iteration": 2.5649683475494385
    },
    {
      "auxiliary_loss_clip": 0.01053424,
      "auxiliary_loss_mlp": 0.0100361,
      "balance_loss_clip": 1.01324797,
      "balance_loss_mlp": 1.00252569,
      "epoch": 0.5668249864726748,
      "flos": 71866949725440.0,
      "grad_norm": 0.875434634574743,
      "language_loss": 0.68785757,
      "learning_rate": 1.6658378321083878e-06,
      "loss": 0.70842791,
      "num_input_tokens_seen": 101766010,
      "step": 4714,
      "time_per_iteration": 3.1468007564544678
    },
    {
      "auxiliary_loss_clip": 0.0110213,
      "auxiliary_loss_mlp": 0.01035322,
      "balance_loss_clip": 1.04425716,
      "balance_loss_mlp": 1.02826464,
      "epoch": 0.5669452293633139,
      "flos": 22195667312640.0,
      "grad_norm": 1.713422510168844,
      "language_loss": 0.8225348,
      "learning_rate": 1.6650698357077055e-06,
      "loss": 0.84390938,
      "num_input_tokens_seen": 101783055,
      "step": 4715,
      "time_per_iteration": 3.3826870918273926
    },
    {
      "auxiliary_loss_clip": 0.01143608,
      "auxiliary_loss_mlp": 0.01028195,
      "balance_loss_clip": 1.04383314,
      "balance_loss_mlp": 1.02036858,
      "epoch": 0.567065472253953,
      "flos": 18223193560320.0,
      "grad_norm": 2.949168219420587,
      "language_loss": 0.80940127,
      "learning_rate": 1.6643018901156632e-06,
      "loss": 0.8311193,
      "num_input_tokens_seen": 101802150,
      "step": 4716,
      "time_per_iteration": 2.4970686435699463
    },
    {
      "auxiliary_loss_clip": 0.01144954,
      "auxiliary_loss_mlp": 0.01026346,
      "balance_loss_clip": 1.04538059,
      "balance_loss_mlp": 1.01909804,
      "epoch": 0.567185715144592,
      "flos": 20371548983040.0,
      "grad_norm": 2.2987399059017104,
      "language_loss": 0.79304308,
      "learning_rate": 1.6635339954487566e-06,
      "loss": 0.81475604,
      "num_input_tokens_seen": 101818025,
      "step": 4717,
      "time_per_iteration": 2.509399175643921
    },
    {
      "auxiliary_loss_clip": 0.01146553,
      "auxiliary_loss_mlp": 0.01027388,
      "balance_loss_clip": 1.04700041,
      "balance_loss_mlp": 1.01988387,
      "epoch": 0.5673059580352312,
      "flos": 23221348174080.0,
      "grad_norm": 1.815391248719157,
      "language_loss": 0.81782675,
      "learning_rate": 1.6627661518234765e-06,
      "loss": 0.83956623,
      "num_input_tokens_seen": 101837280,
      "step": 4718,
      "time_per_iteration": 3.334368944168091
    },
    {
      "auxiliary_loss_clip": 0.01118393,
      "auxiliary_loss_mlp": 0.01025842,
      "balance_loss_clip": 1.0461247,
      "balance_loss_mlp": 1.01849902,
      "epoch": 0.5674262009258703,
      "flos": 21719599430400.0,
      "grad_norm": 1.594474268232112,
      "language_loss": 0.85534954,
      "learning_rate": 1.661998359356302e-06,
      "loss": 0.87679183,
      "num_input_tokens_seen": 101856310,
      "step": 4719,
      "time_per_iteration": 3.3555374145507812
    },
    {
      "auxiliary_loss_clip": 0.01062342,
      "auxiliary_loss_mlp": 0.01002761,
      "balance_loss_clip": 1.01360273,
      "balance_loss_mlp": 1.00169981,
      "epoch": 0.5675464438165093,
      "flos": 67470369114240.0,
      "grad_norm": 0.7474482527793633,
      "language_loss": 0.55889952,
      "learning_rate": 1.6612306181637077e-06,
      "loss": 0.5795505,
      "num_input_tokens_seen": 101915635,
      "step": 4720,
      "time_per_iteration": 3.0358452796936035
    },
    {
      "auxiliary_loss_clip": 0.01128805,
      "auxiliary_loss_mlp": 0.01030949,
      "balance_loss_clip": 1.04492402,
      "balance_loss_mlp": 1.02355814,
      "epoch": 0.5676666867071485,
      "flos": 18879173688960.0,
      "grad_norm": 1.9787976402038354,
      "language_loss": 0.65674096,
      "learning_rate": 1.6604629283621598e-06,
      "loss": 0.67833853,
      "num_input_tokens_seen": 101933565,
      "step": 4721,
      "time_per_iteration": 2.5398221015930176
    },
    {
      "auxiliary_loss_clip": 0.01173405,
      "auxiliary_loss_mlp": 0.01030746,
      "balance_loss_clip": 1.04989457,
      "balance_loss_mlp": 1.02286077,
      "epoch": 0.5677869295977875,
      "flos": 33546778744320.0,
      "grad_norm": 1.8461646883160006,
      "language_loss": 0.74044049,
      "learning_rate": 1.6596952900681152e-06,
      "loss": 0.76248205,
      "num_input_tokens_seen": 101954325,
      "step": 4722,
      "time_per_iteration": 2.565617322921753
    },
    {
      "auxiliary_loss_clip": 0.01104007,
      "auxiliary_loss_mlp": 0.01030994,
      "balance_loss_clip": 1.04480565,
      "balance_loss_mlp": 1.02266121,
      "epoch": 0.5679071724884266,
      "flos": 28037256157440.0,
      "grad_norm": 2.135650447826687,
      "language_loss": 0.8220762,
      "learning_rate": 1.658927703398025e-06,
      "loss": 0.84342623,
      "num_input_tokens_seen": 101974390,
      "step": 4723,
      "time_per_iteration": 2.6109657287597656
    },
    {
      "auxiliary_loss_clip": 0.01116641,
      "auxiliary_loss_mlp": 0.01023953,
      "balance_loss_clip": 1.0407021,
      "balance_loss_mlp": 1.01636267,
      "epoch": 0.5680274153790658,
      "flos": 23550110380800.0,
      "grad_norm": 2.2225134011602443,
      "language_loss": 0.77526283,
      "learning_rate": 1.6581601684683309e-06,
      "loss": 0.79666883,
      "num_input_tokens_seen": 101994815,
      "step": 4724,
      "time_per_iteration": 2.619114637374878
    },
    {
      "auxiliary_loss_clip": 0.01160505,
      "auxiliary_loss_mlp": 0.01028972,
      "balance_loss_clip": 1.05162573,
      "balance_loss_mlp": 1.02097249,
      "epoch": 0.5681476582697048,
      "flos": 22455158140800.0,
      "grad_norm": 2.2800728144588525,
      "language_loss": 0.69007611,
      "learning_rate": 1.6573926853954674e-06,
      "loss": 0.71197087,
      "num_input_tokens_seen": 102012400,
      "step": 4725,
      "time_per_iteration": 2.49540638923645
    },
    {
      "auxiliary_loss_clip": 0.01138347,
      "auxiliary_loss_mlp": 0.01029286,
      "balance_loss_clip": 1.04508293,
      "balance_loss_mlp": 1.02208018,
      "epoch": 0.5682679011603439,
      "flos": 19536913584000.0,
      "grad_norm": 1.850848854559763,
      "language_loss": 0.83151168,
      "learning_rate": 1.6566252542958608e-06,
      "loss": 0.85318804,
      "num_input_tokens_seen": 102031900,
      "step": 4726,
      "time_per_iteration": 2.5470423698425293
    },
    {
      "auxiliary_loss_clip": 0.01125061,
      "auxiliary_loss_mlp": 0.01033273,
      "balance_loss_clip": 1.04676056,
      "balance_loss_mlp": 1.02551234,
      "epoch": 0.568388144050983,
      "flos": 28765488493440.0,
      "grad_norm": 2.063937981027075,
      "language_loss": 0.78575444,
      "learning_rate": 1.6558578752859305e-06,
      "loss": 0.8073377,
      "num_input_tokens_seen": 102050860,
      "step": 4727,
      "time_per_iteration": 2.613264799118042
    },
    {
      "auxiliary_loss_clip": 0.01127821,
      "auxiliary_loss_mlp": 0.01024756,
      "balance_loss_clip": 1.04459763,
      "balance_loss_mlp": 1.01716876,
      "epoch": 0.5685083869416221,
      "flos": 21209452519680.0,
      "grad_norm": 2.0132146713521912,
      "language_loss": 0.78609711,
      "learning_rate": 1.6550905484820865e-06,
      "loss": 0.80762285,
      "num_input_tokens_seen": 102069320,
      "step": 4728,
      "time_per_iteration": 2.5841875076293945
    },
    {
      "auxiliary_loss_clip": 0.011704,
      "auxiliary_loss_mlp": 0.01029724,
      "balance_loss_clip": 1.04892254,
      "balance_loss_mlp": 1.02149272,
      "epoch": 0.5686286298322611,
      "flos": 24827021942400.0,
      "grad_norm": 2.31013763200404,
      "language_loss": 0.78846002,
      "learning_rate": 1.6543232740007328e-06,
      "loss": 0.81046128,
      "num_input_tokens_seen": 102086435,
      "step": 4729,
      "time_per_iteration": 2.4803056716918945
    },
    {
      "auxiliary_loss_clip": 0.01159068,
      "auxiliary_loss_mlp": 0.0102581,
      "balance_loss_clip": 1.04888821,
      "balance_loss_mlp": 1.0180881,
      "epoch": 0.5687488727229003,
      "flos": 26615121909120.0,
      "grad_norm": 2.141097078508931,
      "language_loss": 0.66942346,
      "learning_rate": 1.653556051958263e-06,
      "loss": 0.69127226,
      "num_input_tokens_seen": 102106115,
      "step": 4730,
      "time_per_iteration": 2.5520994663238525
    },
    {
      "auxiliary_loss_clip": 0.01084782,
      "auxiliary_loss_mlp": 0.01026626,
      "balance_loss_clip": 1.04078555,
      "balance_loss_mlp": 1.01953578,
      "epoch": 0.5688691156135394,
      "flos": 20808725414400.0,
      "grad_norm": 1.8917559384178562,
      "language_loss": 0.73782313,
      "learning_rate": 1.6527888824710642e-06,
      "loss": 0.75893724,
      "num_input_tokens_seen": 102125715,
      "step": 4731,
      "time_per_iteration": 2.64235258102417
    },
    {
      "auxiliary_loss_clip": 0.01122456,
      "auxiliary_loss_mlp": 0.01028205,
      "balance_loss_clip": 1.04186225,
      "balance_loss_mlp": 1.0203073,
      "epoch": 0.5689893585041784,
      "flos": 25880963829120.0,
      "grad_norm": 1.9935340461320874,
      "language_loss": 0.76443493,
      "learning_rate": 1.6520217656555166e-06,
      "loss": 0.78594154,
      "num_input_tokens_seen": 102145005,
      "step": 4732,
      "time_per_iteration": 2.600533962249756
    },
    {
      "auxiliary_loss_clip": 0.01131997,
      "auxiliary_loss_mlp": 0.01029244,
      "balance_loss_clip": 1.04349184,
      "balance_loss_mlp": 1.02184987,
      "epoch": 0.5691096013948175,
      "flos": 23477463123840.0,
      "grad_norm": 1.5751842152210997,
      "language_loss": 0.70677924,
      "learning_rate": 1.65125470162799e-06,
      "loss": 0.72839165,
      "num_input_tokens_seen": 102165360,
      "step": 4733,
      "time_per_iteration": 2.5750174522399902
    },
    {
      "auxiliary_loss_clip": 0.01130985,
      "auxiliary_loss_mlp": 0.01029351,
      "balance_loss_clip": 1.04427886,
      "balance_loss_mlp": 1.02093482,
      "epoch": 0.5692298442854566,
      "flos": 18075600576000.0,
      "grad_norm": 2.2551627821073725,
      "language_loss": 0.70095229,
      "learning_rate": 1.6504876905048485e-06,
      "loss": 0.72255564,
      "num_input_tokens_seen": 102182320,
      "step": 4734,
      "time_per_iteration": 2.531982421875
    },
    {
      "auxiliary_loss_clip": 0.01171416,
      "auxiliary_loss_mlp": 0.01029621,
      "balance_loss_clip": 1.0508697,
      "balance_loss_mlp": 1.02267122,
      "epoch": 0.5693500871760957,
      "flos": 23039317025280.0,
      "grad_norm": 1.5176183120504734,
      "language_loss": 0.72159135,
      "learning_rate": 1.6497207324024464e-06,
      "loss": 0.74360168,
      "num_input_tokens_seen": 102201220,
      "step": 4735,
      "time_per_iteration": 2.465909481048584
    },
    {
      "auxiliary_loss_clip": 0.01148003,
      "auxiliary_loss_mlp": 0.01026684,
      "balance_loss_clip": 1.04666221,
      "balance_loss_mlp": 1.01883125,
      "epoch": 0.5694703300667348,
      "flos": 18989670902400.0,
      "grad_norm": 1.8092032623703358,
      "language_loss": 0.82208848,
      "learning_rate": 1.6489538274371305e-06,
      "loss": 0.84383529,
      "num_input_tokens_seen": 102219825,
      "step": 4736,
      "time_per_iteration": 3.329573392868042
    },
    {
      "auxiliary_loss_clip": 0.0115354,
      "auxiliary_loss_mlp": 0.01025248,
      "balance_loss_clip": 1.05006933,
      "balance_loss_mlp": 1.0181905,
      "epoch": 0.5695905729573739,
      "flos": 21908705558400.0,
      "grad_norm": 1.9254229382147208,
      "language_loss": 0.83206773,
      "learning_rate": 1.6481869757252396e-06,
      "loss": 0.85385561,
      "num_input_tokens_seen": 102238160,
      "step": 4737,
      "time_per_iteration": 2.4985244274139404
    },
    {
      "auxiliary_loss_clip": 0.01158208,
      "auxiliary_loss_mlp": 0.01024939,
      "balance_loss_clip": 1.04970181,
      "balance_loss_mlp": 1.0174706,
      "epoch": 0.569710815848013,
      "flos": 28476659232000.0,
      "grad_norm": 1.6906625988672503,
      "language_loss": 0.71727645,
      "learning_rate": 1.647420177383105e-06,
      "loss": 0.73910797,
      "num_input_tokens_seen": 102261030,
      "step": 4738,
      "time_per_iteration": 2.573963165283203
    },
    {
      "auxiliary_loss_clip": 0.01156697,
      "auxiliary_loss_mlp": 0.01031471,
      "balance_loss_clip": 1.05189919,
      "balance_loss_mlp": 1.02422297,
      "epoch": 0.569831058738652,
      "flos": 28366162018560.0,
      "grad_norm": 2.448566560510273,
      "language_loss": 0.72625095,
      "learning_rate": 1.646653432527049e-06,
      "loss": 0.74813259,
      "num_input_tokens_seen": 102281670,
      "step": 4739,
      "time_per_iteration": 2.561300754547119
    },
    {
      "auxiliary_loss_clip": 0.01134758,
      "auxiliary_loss_mlp": 0.01026975,
      "balance_loss_clip": 1.04905844,
      "balance_loss_mlp": 1.01972091,
      "epoch": 0.5699513016292912,
      "flos": 25849973370240.0,
      "grad_norm": 1.4404881304372572,
      "language_loss": 0.74263299,
      "learning_rate": 1.645886741273387e-06,
      "loss": 0.76425028,
      "num_input_tokens_seen": 102303485,
      "step": 4740,
      "time_per_iteration": 2.6033706665039062
    },
    {
      "auxiliary_loss_clip": 0.01124602,
      "auxiliary_loss_mlp": 0.01027407,
      "balance_loss_clip": 1.04941273,
      "balance_loss_mlp": 1.01967013,
      "epoch": 0.5700715445199303,
      "flos": 18037858360320.0,
      "grad_norm": 2.0936335757412086,
      "language_loss": 0.74031961,
      "learning_rate": 1.645120103738424e-06,
      "loss": 0.76183963,
      "num_input_tokens_seen": 102320995,
      "step": 4741,
      "time_per_iteration": 3.3269333839416504
    },
    {
      "auxiliary_loss_clip": 0.01148442,
      "auxiliary_loss_mlp": 0.0076103,
      "balance_loss_clip": 1.04849577,
      "balance_loss_mlp": 1.00021005,
      "epoch": 0.5701917874105693,
      "flos": 11473352392320.0,
      "grad_norm": 2.0110076806483197,
      "language_loss": 0.83219767,
      "learning_rate": 1.6443535200384591e-06,
      "loss": 0.85129237,
      "num_input_tokens_seen": 102339170,
      "step": 4742,
      "time_per_iteration": 2.4744019508361816
    },
    {
      "auxiliary_loss_clip": 0.01173984,
      "auxiliary_loss_mlp": 0.01032777,
      "balance_loss_clip": 1.05339146,
      "balance_loss_mlp": 1.02592242,
      "epoch": 0.5703120303012085,
      "flos": 21761759018880.0,
      "grad_norm": 1.6039343600038112,
      "language_loss": 0.70433724,
      "learning_rate": 1.6435869902897827e-06,
      "loss": 0.72640479,
      "num_input_tokens_seen": 102357750,
      "step": 4743,
      "time_per_iteration": 2.487847089767456
    },
    {
      "auxiliary_loss_clip": 0.0103572,
      "auxiliary_loss_mlp": 0.01007134,
      "balance_loss_clip": 1.01628327,
      "balance_loss_mlp": 1.00598323,
      "epoch": 0.5704322731918475,
      "flos": 56746258513920.0,
      "grad_norm": 0.7918647137310475,
      "language_loss": 0.62044322,
      "learning_rate": 1.6428205146086764e-06,
      "loss": 0.64087176,
      "num_input_tokens_seen": 102419730,
      "step": 4744,
      "time_per_iteration": 3.2042245864868164
    },
    {
      "auxiliary_loss_clip": 0.0114968,
      "auxiliary_loss_mlp": 0.01029254,
      "balance_loss_clip": 1.0479182,
      "balance_loss_mlp": 1.02178836,
      "epoch": 0.5705525160824866,
      "flos": 20741141975040.0,
      "grad_norm": 1.5074041409012287,
      "language_loss": 0.70648509,
      "learning_rate": 1.6420540931114142e-06,
      "loss": 0.72827446,
      "num_input_tokens_seen": 102440320,
      "step": 4745,
      "time_per_iteration": 4.0821545124053955
    },
    {
      "auxiliary_loss_clip": 0.01148687,
      "auxiliary_loss_mlp": 0.01033886,
      "balance_loss_clip": 1.04960227,
      "balance_loss_mlp": 1.02589941,
      "epoch": 0.5706727589731257,
      "flos": 18771262254720.0,
      "grad_norm": 2.0127515635864546,
      "language_loss": 0.79113483,
      "learning_rate": 1.6412877259142616e-06,
      "loss": 0.81296062,
      "num_input_tokens_seen": 102460240,
      "step": 4746,
      "time_per_iteration": 2.5455076694488525
    },
    {
      "auxiliary_loss_clip": 0.01141756,
      "auxiliary_loss_mlp": 0.01029052,
      "balance_loss_clip": 1.04783261,
      "balance_loss_mlp": 1.0211184,
      "epoch": 0.5707930018637648,
      "flos": 27634733372160.0,
      "grad_norm": 2.911127026230653,
      "language_loss": 0.73569691,
      "learning_rate": 1.6405214131334757e-06,
      "loss": 0.75740504,
      "num_input_tokens_seen": 102478765,
      "step": 4747,
      "time_per_iteration": 2.588075637817383
    },
    {
      "auxiliary_loss_clip": 0.0111761,
      "auxiliary_loss_mlp": 0.01028036,
      "balance_loss_clip": 1.0471909,
      "balance_loss_mlp": 1.02029657,
      "epoch": 0.5709132447544039,
      "flos": 27597673514880.0,
      "grad_norm": 1.759458085911633,
      "language_loss": 0.79256952,
      "learning_rate": 1.6397551548853052e-06,
      "loss": 0.814026,
      "num_input_tokens_seen": 102496930,
      "step": 4748,
      "time_per_iteration": 2.6519582271575928
    },
    {
      "auxiliary_loss_clip": 0.01145668,
      "auxiliary_loss_mlp": 0.01024858,
      "balance_loss_clip": 1.05089509,
      "balance_loss_mlp": 1.01686549,
      "epoch": 0.571033487645043,
      "flos": 21686095019520.0,
      "grad_norm": 1.522793647113733,
      "language_loss": 0.71272659,
      "learning_rate": 1.6389889512859917e-06,
      "loss": 0.7344318,
      "num_input_tokens_seen": 102516590,
      "step": 4749,
      "time_per_iteration": 2.5416009426116943
    },
    {
      "auxiliary_loss_clip": 0.01043416,
      "auxiliary_loss_mlp": 0.01000553,
      "balance_loss_clip": 1.01502728,
      "balance_loss_mlp": 0.9994567,
      "epoch": 0.5711537305356821,
      "flos": 70181445980160.0,
      "grad_norm": 0.8118761218628681,
      "language_loss": 0.60390252,
      "learning_rate": 1.638222802451767e-06,
      "loss": 0.62434214,
      "num_input_tokens_seen": 102578070,
      "step": 4750,
      "time_per_iteration": 3.1377086639404297
    },
    {
      "auxiliary_loss_clip": 0.0115196,
      "auxiliary_loss_mlp": 0.0102977,
      "balance_loss_clip": 1.04867482,
      "balance_loss_mlp": 1.02245104,
      "epoch": 0.5712739734263211,
      "flos": 24717494396160.0,
      "grad_norm": 1.6649042227474369,
      "language_loss": 0.7530148,
      "learning_rate": 1.6374567084988561e-06,
      "loss": 0.77483213,
      "num_input_tokens_seen": 102599255,
      "step": 4751,
      "time_per_iteration": 2.54168701171875
    },
    {
      "auxiliary_loss_clip": 0.01150128,
      "auxiliary_loss_mlp": 0.0102702,
      "balance_loss_clip": 1.0527513,
      "balance_loss_mlp": 1.01900673,
      "epoch": 0.5713942163169603,
      "flos": 26578169792640.0,
      "grad_norm": 1.598190291869178,
      "language_loss": 0.76777369,
      "learning_rate": 1.6366906695434738e-06,
      "loss": 0.78954518,
      "num_input_tokens_seen": 102621775,
      "step": 4752,
      "time_per_iteration": 2.579773187637329
    },
    {
      "auxiliary_loss_clip": 0.0116089,
      "auxiliary_loss_mlp": 0.01032092,
      "balance_loss_clip": 1.05299926,
      "balance_loss_mlp": 1.02511501,
      "epoch": 0.5715144592075994,
      "flos": 21142443697920.0,
      "grad_norm": 1.781591038237632,
      "language_loss": 0.85552299,
      "learning_rate": 1.6359246857018275e-06,
      "loss": 0.87745285,
      "num_input_tokens_seen": 102639305,
      "step": 4753,
      "time_per_iteration": 2.4952282905578613
    },
    {
      "auxiliary_loss_clip": 0.01111512,
      "auxiliary_loss_mlp": 0.01025702,
      "balance_loss_clip": 1.04175806,
      "balance_loss_mlp": 1.01840043,
      "epoch": 0.5716347020982384,
      "flos": 23330265189120.0,
      "grad_norm": 1.8804661356654193,
      "language_loss": 0.78113353,
      "learning_rate": 1.6351587570901178e-06,
      "loss": 0.80250573,
      "num_input_tokens_seen": 102659430,
      "step": 4754,
      "time_per_iteration": 2.6202831268310547
    },
    {
      "auxiliary_loss_clip": 0.0113038,
      "auxiliary_loss_mlp": 0.01029163,
      "balance_loss_clip": 1.0492487,
      "balance_loss_mlp": 1.02137303,
      "epoch": 0.5717549449888776,
      "flos": 17009555806080.0,
      "grad_norm": 2.698012580418275,
      "language_loss": 0.76012558,
      "learning_rate": 1.634392883824534e-06,
      "loss": 0.781721,
      "num_input_tokens_seen": 102671430,
      "step": 4755,
      "time_per_iteration": 2.5178325176239014
    },
    {
      "auxiliary_loss_clip": 0.0111582,
      "auxiliary_loss_mlp": 0.01031547,
      "balance_loss_clip": 1.04433823,
      "balance_loss_mlp": 1.02371466,
      "epoch": 0.5718751878795166,
      "flos": 35518130922240.0,
      "grad_norm": 1.8204940130282183,
      "language_loss": 0.67399329,
      "learning_rate": 1.6336270660212595e-06,
      "loss": 0.695467,
      "num_input_tokens_seen": 102693025,
      "step": 4756,
      "time_per_iteration": 2.7197060585021973
    },
    {
      "auxiliary_loss_clip": 0.01141803,
      "auxiliary_loss_mlp": 0.01029469,
      "balance_loss_clip": 1.05158997,
      "balance_loss_mlp": 1.02105033,
      "epoch": 0.5719954307701557,
      "flos": 38613989255040.0,
      "grad_norm": 2.056452172171135,
      "language_loss": 0.65513343,
      "learning_rate": 1.6328613037964676e-06,
      "loss": 0.67684615,
      "num_input_tokens_seen": 102716090,
      "step": 4757,
      "time_per_iteration": 2.668653726577759
    },
    {
      "auxiliary_loss_clip": 0.01156821,
      "auxiliary_loss_mlp": 0.01028702,
      "balance_loss_clip": 1.0486939,
      "balance_loss_mlp": 1.02082849,
      "epoch": 0.5721156736607949,
      "flos": 20631111638400.0,
      "grad_norm": 1.7234097491464406,
      "language_loss": 0.67769587,
      "learning_rate": 1.6320955972663241e-06,
      "loss": 0.69955111,
      "num_input_tokens_seen": 102735685,
      "step": 4758,
      "time_per_iteration": 2.5086021423339844
    },
    {
      "auxiliary_loss_clip": 0.01157148,
      "auxiliary_loss_mlp": 0.01027811,
      "balance_loss_clip": 1.04796815,
      "balance_loss_mlp": 1.02002943,
      "epoch": 0.5722359165514339,
      "flos": 37415076076800.0,
      "grad_norm": 1.5412204856880487,
      "language_loss": 0.64868259,
      "learning_rate": 1.6313299465469857e-06,
      "loss": 0.67053211,
      "num_input_tokens_seen": 102758415,
      "step": 4759,
      "time_per_iteration": 2.6277527809143066
    },
    {
      "auxiliary_loss_clip": 0.01156048,
      "auxiliary_loss_mlp": 0.0103488,
      "balance_loss_clip": 1.05017495,
      "balance_loss_mlp": 1.02714288,
      "epoch": 0.572356159442073,
      "flos": 21972877205760.0,
      "grad_norm": 2.686810003290098,
      "language_loss": 0.7934559,
      "learning_rate": 1.6305643517546014e-06,
      "loss": 0.8153652,
      "num_input_tokens_seen": 102773795,
      "step": 4760,
      "time_per_iteration": 2.480984687805176
    },
    {
      "auxiliary_loss_clip": 0.01172289,
      "auxiliary_loss_mlp": 0.01030474,
      "balance_loss_clip": 1.05287194,
      "balance_loss_mlp": 1.02345872,
      "epoch": 0.5724764023327121,
      "flos": 19135540033920.0,
      "grad_norm": 1.764084811537368,
      "language_loss": 0.84211975,
      "learning_rate": 1.629798813005311e-06,
      "loss": 0.86414737,
      "num_input_tokens_seen": 102793515,
      "step": 4761,
      "time_per_iteration": 3.2491726875305176
    },
    {
      "auxiliary_loss_clip": 0.01119048,
      "auxiliary_loss_mlp": 0.01023115,
      "balance_loss_clip": 1.04783571,
      "balance_loss_mlp": 1.01566744,
      "epoch": 0.5725966452233512,
      "flos": 22819759142400.0,
      "grad_norm": 1.9445446213251805,
      "language_loss": 0.70825076,
      "learning_rate": 1.6290333304152473e-06,
      "loss": 0.72967243,
      "num_input_tokens_seen": 102813390,
      "step": 4762,
      "time_per_iteration": 2.610839366912842
    },
    {
      "auxiliary_loss_clip": 0.01140527,
      "auxiliary_loss_mlp": 0.01030485,
      "balance_loss_clip": 1.05040097,
      "balance_loss_mlp": 1.02258182,
      "epoch": 0.5727168881139902,
      "flos": 41496610498560.0,
      "grad_norm": 1.6940130316478839,
      "language_loss": 0.57042205,
      "learning_rate": 1.6282679041005314e-06,
      "loss": 0.59213221,
      "num_input_tokens_seen": 102838980,
      "step": 4763,
      "time_per_iteration": 2.7282886505126953
    },
    {
      "auxiliary_loss_clip": 0.01136855,
      "auxiliary_loss_mlp": 0.01027128,
      "balance_loss_clip": 1.04588389,
      "balance_loss_mlp": 1.01975834,
      "epoch": 0.5728371310046293,
      "flos": 14647675985280.0,
      "grad_norm": 3.961083090736645,
      "language_loss": 0.87047613,
      "learning_rate": 1.6275025341772789e-06,
      "loss": 0.89211607,
      "num_input_tokens_seen": 102855285,
      "step": 4764,
      "time_per_iteration": 2.505280017852783
    },
    {
      "auxiliary_loss_clip": 0.01143504,
      "auxiliary_loss_mlp": 0.01029875,
      "balance_loss_clip": 1.04676807,
      "balance_loss_mlp": 1.02166152,
      "epoch": 0.5729573738952685,
      "flos": 21506613736320.0,
      "grad_norm": 5.618618334090883,
      "language_loss": 0.82081956,
      "learning_rate": 1.626737220761596e-06,
      "loss": 0.84255338,
      "num_input_tokens_seen": 102872750,
      "step": 4765,
      "time_per_iteration": 2.552645683288574
    },
    {
      "auxiliary_loss_clip": 0.01154585,
      "auxiliary_loss_mlp": 0.01036999,
      "balance_loss_clip": 1.05006874,
      "balance_loss_mlp": 1.02928042,
      "epoch": 0.5730776167859075,
      "flos": 23621680229760.0,
      "grad_norm": 1.8654010787597635,
      "language_loss": 0.7898705,
      "learning_rate": 1.62597196396958e-06,
      "loss": 0.81178629,
      "num_input_tokens_seen": 102890920,
      "step": 4766,
      "time_per_iteration": 3.280918836593628
    },
    {
      "auxiliary_loss_clip": 0.01156227,
      "auxiliary_loss_mlp": 0.010289,
      "balance_loss_clip": 1.04863846,
      "balance_loss_mlp": 1.02112126,
      "epoch": 0.5731978596765466,
      "flos": 25739224761600.0,
      "grad_norm": 1.8773122771276927,
      "language_loss": 0.8566817,
      "learning_rate": 1.6252067639173197e-06,
      "loss": 0.87853301,
      "num_input_tokens_seen": 102912830,
      "step": 4767,
      "time_per_iteration": 2.546895742416382
    },
    {
      "auxiliary_loss_clip": 0.01160066,
      "auxiliary_loss_mlp": 0.01028992,
      "balance_loss_clip": 1.04966283,
      "balance_loss_mlp": 1.0213927,
      "epoch": 0.5733181025671857,
      "flos": 26359509749760.0,
      "grad_norm": 1.7463057671398208,
      "language_loss": 0.6995312,
      "learning_rate": 1.6244416207208956e-06,
      "loss": 0.72142178,
      "num_input_tokens_seen": 102933765,
      "step": 4768,
      "time_per_iteration": 2.5489680767059326
    },
    {
      "auxiliary_loss_clip": 0.01129432,
      "auxiliary_loss_mlp": 0.01026375,
      "balance_loss_clip": 1.04691076,
      "balance_loss_mlp": 1.0186559,
      "epoch": 0.5734383454578248,
      "flos": 29423874833280.0,
      "grad_norm": 1.7217540019677244,
      "language_loss": 0.73395759,
      "learning_rate": 1.6236765344963787e-06,
      "loss": 0.75551569,
      "num_input_tokens_seen": 102955025,
      "step": 4769,
      "time_per_iteration": 2.6353330612182617
    },
    {
      "auxiliary_loss_clip": 0.01145087,
      "auxiliary_loss_mlp": 0.01030171,
      "balance_loss_clip": 1.04959655,
      "balance_loss_mlp": 1.02301884,
      "epoch": 0.5735585883484638,
      "flos": 34969954487040.0,
      "grad_norm": 2.0155849524802276,
      "language_loss": 0.69119173,
      "learning_rate": 1.6229115053598322e-06,
      "loss": 0.71294433,
      "num_input_tokens_seen": 102976780,
      "step": 4770,
      "time_per_iteration": 2.6697425842285156
    },
    {
      "auxiliary_loss_clip": 0.01160832,
      "auxiliary_loss_mlp": 0.01036931,
      "balance_loss_clip": 1.05424094,
      "balance_loss_mlp": 1.02879238,
      "epoch": 0.573678831239103,
      "flos": 18770759464320.0,
      "grad_norm": 1.8125642122205157,
      "language_loss": 0.72255546,
      "learning_rate": 1.6221465334273108e-06,
      "loss": 0.74453306,
      "num_input_tokens_seen": 102995990,
      "step": 4771,
      "time_per_iteration": 3.292959213256836
    },
    {
      "auxiliary_loss_clip": 0.01134401,
      "auxiliary_loss_mlp": 0.0102348,
      "balance_loss_clip": 1.04693174,
      "balance_loss_mlp": 1.01647925,
      "epoch": 0.5737990741297421,
      "flos": 25702883176320.0,
      "grad_norm": 2.020893945632337,
      "language_loss": 0.61735672,
      "learning_rate": 1.6213816188148593e-06,
      "loss": 0.63893551,
      "num_input_tokens_seen": 103014695,
      "step": 4772,
      "time_per_iteration": 2.605311155319214
    },
    {
      "auxiliary_loss_clip": 0.01133305,
      "auxiliary_loss_mlp": 0.01029702,
      "balance_loss_clip": 1.04817891,
      "balance_loss_mlp": 1.02208745,
      "epoch": 0.5739193170203811,
      "flos": 27269234530560.0,
      "grad_norm": 1.759720230345461,
      "language_loss": 0.77431649,
      "learning_rate": 1.6206167616385162e-06,
      "loss": 0.79594654,
      "num_input_tokens_seen": 103035760,
      "step": 4773,
      "time_per_iteration": 2.5785958766937256
    },
    {
      "auxiliary_loss_clip": 0.01151543,
      "auxiliary_loss_mlp": 0.01028921,
      "balance_loss_clip": 1.05100727,
      "balance_loss_mlp": 1.02043962,
      "epoch": 0.5740395599110203,
      "flos": 12239721993600.0,
      "grad_norm": 2.5958423895791336,
      "language_loss": 0.74023098,
      "learning_rate": 1.6198519620143078e-06,
      "loss": 0.76203561,
      "num_input_tokens_seen": 103052915,
      "step": 4774,
      "time_per_iteration": 2.5263259410858154
    },
    {
      "auxiliary_loss_clip": 0.01135187,
      "auxiliary_loss_mlp": 0.01026265,
      "balance_loss_clip": 1.04890394,
      "balance_loss_mlp": 1.01860023,
      "epoch": 0.5741598028016593,
      "flos": 25921399564800.0,
      "grad_norm": 1.5435255837358124,
      "language_loss": 0.7831285,
      "learning_rate": 1.6190872200582546e-06,
      "loss": 0.80474305,
      "num_input_tokens_seen": 103074655,
      "step": 4775,
      "time_per_iteration": 2.62835955619812
    },
    {
      "auxiliary_loss_clip": 0.01140175,
      "auxiliary_loss_mlp": 0.00761628,
      "balance_loss_clip": 1.04755259,
      "balance_loss_mlp": 1.00020814,
      "epoch": 0.5742800456922984,
      "flos": 19244133826560.0,
      "grad_norm": 2.6474815931208657,
      "language_loss": 0.7825954,
      "learning_rate": 1.6183225358863676e-06,
      "loss": 0.80161345,
      "num_input_tokens_seen": 103091550,
      "step": 4776,
      "time_per_iteration": 2.5224947929382324
    },
    {
      "auxiliary_loss_clip": 0.01134854,
      "auxiliary_loss_mlp": 0.01026713,
      "balance_loss_clip": 1.04681802,
      "balance_loss_mlp": 1.01848793,
      "epoch": 0.5744002885829376,
      "flos": 30920487932160.0,
      "grad_norm": 4.105652291156933,
      "language_loss": 0.71387446,
      "learning_rate": 1.617557909614648e-06,
      "loss": 0.7354902,
      "num_input_tokens_seen": 103110985,
      "step": 4777,
      "time_per_iteration": 2.6222987174987793
    },
    {
      "auxiliary_loss_clip": 0.01127871,
      "auxiliary_loss_mlp": 0.01030873,
      "balance_loss_clip": 1.04680073,
      "balance_loss_mlp": 1.02333939,
      "epoch": 0.5745205314735766,
      "flos": 23840017050240.0,
      "grad_norm": 1.8215922031128746,
      "language_loss": 0.86272371,
      "learning_rate": 1.6167933413590899e-06,
      "loss": 0.88431108,
      "num_input_tokens_seen": 103129890,
      "step": 4778,
      "time_per_iteration": 2.5919148921966553
    },
    {
      "auxiliary_loss_clip": 0.01158553,
      "auxiliary_loss_mlp": 0.01025437,
      "balance_loss_clip": 1.05003905,
      "balance_loss_mlp": 1.01781368,
      "epoch": 0.5746407743642157,
      "flos": 12311902373760.0,
      "grad_norm": 2.269078951126678,
      "language_loss": 0.90704387,
      "learning_rate": 1.6160288312356773e-06,
      "loss": 0.92888373,
      "num_input_tokens_seen": 103147020,
      "step": 4779,
      "time_per_iteration": 2.493025541305542
    },
    {
      "auxiliary_loss_clip": 0.01162105,
      "auxiliary_loss_mlp": 0.01029699,
      "balance_loss_clip": 1.04978156,
      "balance_loss_mlp": 1.02209938,
      "epoch": 0.5747610172548548,
      "flos": 24133658734080.0,
      "grad_norm": 1.6131643340121986,
      "language_loss": 0.81732285,
      "learning_rate": 1.6152643793603857e-06,
      "loss": 0.83924091,
      "num_input_tokens_seen": 103167370,
      "step": 4780,
      "time_per_iteration": 2.5295186042785645
    },
    {
      "auxiliary_loss_clip": 0.01174452,
      "auxiliary_loss_mlp": 0.01033965,
      "balance_loss_clip": 1.0535208,
      "balance_loss_mlp": 1.02639556,
      "epoch": 0.5748812601454939,
      "flos": 25408451393280.0,
      "grad_norm": 1.6226286000650232,
      "language_loss": 0.87725425,
      "learning_rate": 1.6144999858491815e-06,
      "loss": 0.89933848,
      "num_input_tokens_seen": 103186000,
      "step": 4781,
      "time_per_iteration": 2.503038167953491
    },
    {
      "auxiliary_loss_clip": 0.01147616,
      "auxiliary_loss_mlp": 0.01025771,
      "balance_loss_clip": 1.04766583,
      "balance_loss_mlp": 1.01820755,
      "epoch": 0.575001503036133,
      "flos": 30624942827520.0,
      "grad_norm": 1.5980363231149146,
      "language_loss": 0.8568657,
      "learning_rate": 1.6137356508180232e-06,
      "loss": 0.87859964,
      "num_input_tokens_seen": 103207710,
      "step": 4782,
      "time_per_iteration": 2.6149351596832275
    },
    {
      "auxiliary_loss_clip": 0.0117066,
      "auxiliary_loss_mlp": 0.00761925,
      "balance_loss_clip": 1.04917169,
      "balance_loss_mlp": 1.0002104,
      "epoch": 0.5751217459267721,
      "flos": 21726566668800.0,
      "grad_norm": 1.7299858508468873,
      "language_loss": 0.81587791,
      "learning_rate": 1.6129713743828593e-06,
      "loss": 0.83520377,
      "num_input_tokens_seen": 103226720,
      "step": 4783,
      "time_per_iteration": 2.485017776489258
    },
    {
      "auxiliary_loss_clip": 0.01143442,
      "auxiliary_loss_mlp": 0.01030248,
      "balance_loss_clip": 1.04565346,
      "balance_loss_mlp": 1.02264857,
      "epoch": 0.5752419888174112,
      "flos": 21651620941440.0,
      "grad_norm": 1.4838722564552809,
      "language_loss": 0.75472128,
      "learning_rate": 1.6122071566596306e-06,
      "loss": 0.7764582,
      "num_input_tokens_seen": 103246995,
      "step": 4784,
      "time_per_iteration": 2.5426180362701416
    },
    {
      "auxiliary_loss_clip": 0.01160104,
      "auxiliary_loss_mlp": 0.01027385,
      "balance_loss_clip": 1.05040884,
      "balance_loss_mlp": 1.01951718,
      "epoch": 0.5753622317080502,
      "flos": 17775997234560.0,
      "grad_norm": 6.482085405431135,
      "language_loss": 0.83305538,
      "learning_rate": 1.6114429977642674e-06,
      "loss": 0.85493028,
      "num_input_tokens_seen": 103261500,
      "step": 4785,
      "time_per_iteration": 2.456416130065918
    },
    {
      "auxiliary_loss_clip": 0.01158002,
      "auxiliary_loss_mlp": 0.01029385,
      "balance_loss_clip": 1.05243027,
      "balance_loss_mlp": 1.02186251,
      "epoch": 0.5754824745986894,
      "flos": 19789616741760.0,
      "grad_norm": 7.606676463526228,
      "language_loss": 0.73793435,
      "learning_rate": 1.6106788978126926e-06,
      "loss": 0.75980824,
      "num_input_tokens_seen": 103280475,
      "step": 4786,
      "time_per_iteration": 2.5046398639678955
    },
    {
      "auxiliary_loss_clip": 0.01109222,
      "auxiliary_loss_mlp": 0.01027306,
      "balance_loss_clip": 1.04114509,
      "balance_loss_mlp": 1.01943839,
      "epoch": 0.5756027174893285,
      "flos": 30985665160320.0,
      "grad_norm": 2.445406540617777,
      "language_loss": 0.78718388,
      "learning_rate": 1.6099148569208196e-06,
      "loss": 0.80854917,
      "num_input_tokens_seen": 103297695,
      "step": 4787,
      "time_per_iteration": 3.455181360244751
    },
    {
      "auxiliary_loss_clip": 0.0114394,
      "auxiliary_loss_mlp": 0.0103374,
      "balance_loss_clip": 1.05181098,
      "balance_loss_mlp": 1.02570844,
      "epoch": 0.5757229603799675,
      "flos": 28546864364160.0,
      "grad_norm": 2.1091661787231413,
      "language_loss": 0.62842584,
      "learning_rate": 1.6091508752045523e-06,
      "loss": 0.65020263,
      "num_input_tokens_seen": 103318575,
      "step": 4788,
      "time_per_iteration": 2.5940284729003906
    },
    {
      "auxiliary_loss_clip": 0.01122593,
      "auxiliary_loss_mlp": 0.01032837,
      "balance_loss_clip": 1.04350877,
      "balance_loss_mlp": 1.02476633,
      "epoch": 0.5758432032706067,
      "flos": 22999024944000.0,
      "grad_norm": 1.5457673126035152,
      "language_loss": 0.86420023,
      "learning_rate": 1.608386952779787e-06,
      "loss": 0.88575453,
      "num_input_tokens_seen": 103337945,
      "step": 4789,
      "time_per_iteration": 2.6013150215148926
    },
    {
      "auxiliary_loss_clip": 0.01148094,
      "auxiliary_loss_mlp": 0.0102869,
      "balance_loss_clip": 1.04843998,
      "balance_loss_mlp": 1.0214417,
      "epoch": 0.5759634461612457,
      "flos": 25739727552000.0,
      "grad_norm": 1.5506805590535169,
      "language_loss": 0.74750948,
      "learning_rate": 1.6076230897624098e-06,
      "loss": 0.76927722,
      "num_input_tokens_seen": 103360150,
      "step": 4790,
      "time_per_iteration": 2.6182539463043213
    },
    {
      "auxiliary_loss_clip": 0.01158484,
      "auxiliary_loss_mlp": 0.01030395,
      "balance_loss_clip": 1.04729056,
      "balance_loss_mlp": 1.02258086,
      "epoch": 0.5760836890518848,
      "flos": 30591761639040.0,
      "grad_norm": 1.9088375603250827,
      "language_loss": 0.7758975,
      "learning_rate": 1.6068592862682974e-06,
      "loss": 0.79778636,
      "num_input_tokens_seen": 103378305,
      "step": 4791,
      "time_per_iteration": 2.575953960418701
    },
    {
      "auxiliary_loss_clip": 0.01148226,
      "auxiliary_loss_mlp": 0.01031449,
      "balance_loss_clip": 1.04884803,
      "balance_loss_mlp": 1.02446985,
      "epoch": 0.576203931942524,
      "flos": 36538963447680.0,
      "grad_norm": 2.204847724069221,
      "language_loss": 0.73596662,
      "learning_rate": 1.6060955424133187e-06,
      "loss": 0.75776345,
      "num_input_tokens_seen": 103399230,
      "step": 4792,
      "time_per_iteration": 2.6583011150360107
    },
    {
      "auxiliary_loss_clip": 0.01158691,
      "auxiliary_loss_mlp": 0.01027771,
      "balance_loss_clip": 1.0502764,
      "balance_loss_mlp": 1.02022171,
      "epoch": 0.576324174833163,
      "flos": 25516937445120.0,
      "grad_norm": 1.6171293810970002,
      "language_loss": 0.89616859,
      "learning_rate": 1.6053318583133332e-06,
      "loss": 0.91803324,
      "num_input_tokens_seen": 103420100,
      "step": 4793,
      "time_per_iteration": 3.374431610107422
    },
    {
      "auxiliary_loss_clip": 0.01156572,
      "auxiliary_loss_mlp": 0.0102472,
      "balance_loss_clip": 1.04942513,
      "balance_loss_mlp": 1.01738226,
      "epoch": 0.5764444177238021,
      "flos": 25119262995840.0,
      "grad_norm": 2.322522423522114,
      "language_loss": 0.75269282,
      "learning_rate": 1.6045682340841907e-06,
      "loss": 0.77450567,
      "num_input_tokens_seen": 103439025,
      "step": 4794,
      "time_per_iteration": 2.539273262023926
    },
    {
      "auxiliary_loss_clip": 0.01032689,
      "auxiliary_loss_mlp": 0.00751221,
      "balance_loss_clip": 1.01359463,
      "balance_loss_mlp": 1.00021029,
      "epoch": 0.5765646606144411,
      "flos": 62212687758720.0,
      "grad_norm": 0.7508870940778991,
      "language_loss": 0.57995141,
      "learning_rate": 1.6038046698417336e-06,
      "loss": 0.59779054,
      "num_input_tokens_seen": 103499920,
      "step": 4795,
      "time_per_iteration": 3.1307826042175293
    },
    {
      "auxiliary_loss_clip": 0.01155554,
      "auxiliary_loss_mlp": 0.01027849,
      "balance_loss_clip": 1.04715753,
      "balance_loss_mlp": 1.01985025,
      "epoch": 0.5766849035050803,
      "flos": 25118760205440.0,
      "grad_norm": 1.8083449014192836,
      "language_loss": 0.68772602,
      "learning_rate": 1.6030411657017919e-06,
      "loss": 0.7095601,
      "num_input_tokens_seen": 103519575,
      "step": 4796,
      "time_per_iteration": 2.5229616165161133
    },
    {
      "auxiliary_loss_clip": 0.01148146,
      "auxiliary_loss_mlp": 0.01031643,
      "balance_loss_clip": 1.04665375,
      "balance_loss_mlp": 1.02414465,
      "epoch": 0.5768051463957193,
      "flos": 15991093578240.0,
      "grad_norm": 1.7008754443343417,
      "language_loss": 0.84494591,
      "learning_rate": 1.6022777217801903e-06,
      "loss": 0.86674374,
      "num_input_tokens_seen": 103536530,
      "step": 4797,
      "time_per_iteration": 3.297555685043335
    },
    {
      "auxiliary_loss_clip": 0.01130484,
      "auxiliary_loss_mlp": 0.01029586,
      "balance_loss_clip": 1.04904699,
      "balance_loss_mlp": 1.02223372,
      "epoch": 0.5769253892863584,
      "flos": 22163635359360.0,
      "grad_norm": 1.9085876191566582,
      "language_loss": 0.7395792,
      "learning_rate": 1.601514338192742e-06,
      "loss": 0.7611798,
      "num_input_tokens_seen": 103556460,
      "step": 4798,
      "time_per_iteration": 2.5574562549591064
    },
    {
      "auxiliary_loss_clip": 0.01165067,
      "auxiliary_loss_mlp": 0.01027318,
      "balance_loss_clip": 1.04760957,
      "balance_loss_mlp": 1.02040946,
      "epoch": 0.5770456321769976,
      "flos": 22856388036480.0,
      "grad_norm": 2.0559547223549575,
      "language_loss": 0.71279776,
      "learning_rate": 1.6007510150552514e-06,
      "loss": 0.73472154,
      "num_input_tokens_seen": 103574520,
      "step": 4799,
      "time_per_iteration": 2.484485149383545
    },
    {
      "auxiliary_loss_clip": 0.0115935,
      "auxiliary_loss_mlp": 0.01031114,
      "balance_loss_clip": 1.04719055,
      "balance_loss_mlp": 1.02308846,
      "epoch": 0.5771658750676366,
      "flos": 46353672489600.0,
      "grad_norm": 1.5152514541315234,
      "language_loss": 0.61992913,
      "learning_rate": 1.599987752483515e-06,
      "loss": 0.64183378,
      "num_input_tokens_seen": 103598965,
      "step": 4800,
      "time_per_iteration": 2.732713460922241
    },
    {
      "auxiliary_loss_clip": 0.01123285,
      "auxiliary_loss_mlp": 0.01031384,
      "balance_loss_clip": 1.04452109,
      "balance_loss_mlp": 1.02354646,
      "epoch": 0.5772861179582757,
      "flos": 22159972172160.0,
      "grad_norm": 1.6078098071599773,
      "language_loss": 0.67803788,
      "learning_rate": 1.5992245505933184e-06,
      "loss": 0.6995846,
      "num_input_tokens_seen": 103618665,
      "step": 4801,
      "time_per_iteration": 2.557635545730591
    },
    {
      "auxiliary_loss_clip": 0.01172842,
      "auxiliary_loss_mlp": 0.01029738,
      "balance_loss_clip": 1.0515095,
      "balance_loss_mlp": 1.02215004,
      "epoch": 0.5774063608489148,
      "flos": 31248926916480.0,
      "grad_norm": 2.210489589614168,
      "language_loss": 0.70919347,
      "learning_rate": 1.5984614095004388e-06,
      "loss": 0.73121929,
      "num_input_tokens_seen": 103639800,
      "step": 4802,
      "time_per_iteration": 2.5752646923065186
    },
    {
      "auxiliary_loss_clip": 0.01150268,
      "auxiliary_loss_mlp": 0.01025872,
      "balance_loss_clip": 1.04750919,
      "balance_loss_mlp": 1.01792026,
      "epoch": 0.5775266037395539,
      "flos": 22527123039360.0,
      "grad_norm": 2.3473591381145793,
      "language_loss": 0.8086977,
      "learning_rate": 1.5976983293206438e-06,
      "loss": 0.83045912,
      "num_input_tokens_seen": 103655605,
      "step": 4803,
      "time_per_iteration": 2.488098382949829
    },
    {
      "auxiliary_loss_clip": 0.0113933,
      "auxiliary_loss_mlp": 0.01025847,
      "balance_loss_clip": 1.04440355,
      "balance_loss_mlp": 1.01837826,
      "epoch": 0.577646846630193,
      "flos": 21068790860160.0,
      "grad_norm": 1.9298818234721835,
      "language_loss": 0.71243626,
      "learning_rate": 1.5969353101696928e-06,
      "loss": 0.73408806,
      "num_input_tokens_seen": 103674045,
      "step": 4804,
      "time_per_iteration": 2.5387024879455566
    },
    {
      "auxiliary_loss_clip": 0.01156367,
      "auxiliary_loss_mlp": 0.0103026,
      "balance_loss_clip": 1.04830194,
      "balance_loss_mlp": 1.02350962,
      "epoch": 0.5777670895208321,
      "flos": 29714284293120.0,
      "grad_norm": 1.89261819296346,
      "language_loss": 0.79934031,
      "learning_rate": 1.5961723521633341e-06,
      "loss": 0.82120657,
      "num_input_tokens_seen": 103695285,
      "step": 4805,
      "time_per_iteration": 2.57186222076416
    },
    {
      "auxiliary_loss_clip": 0.01136441,
      "auxiliary_loss_mlp": 0.01027689,
      "balance_loss_clip": 1.04608333,
      "balance_loss_mlp": 1.02005982,
      "epoch": 0.5778873324114712,
      "flos": 19500428344320.0,
      "grad_norm": 2.3827984269904277,
      "language_loss": 0.90970594,
      "learning_rate": 1.5954094554173097e-06,
      "loss": 0.93134725,
      "num_input_tokens_seen": 103713275,
      "step": 4806,
      "time_per_iteration": 2.5221855640411377
    },
    {
      "auxiliary_loss_clip": 0.01148218,
      "auxiliary_loss_mlp": 0.01033268,
      "balance_loss_clip": 1.04969299,
      "balance_loss_mlp": 1.02640474,
      "epoch": 0.5780075753021102,
      "flos": 14136846716160.0,
      "grad_norm": 2.0704251206657345,
      "language_loss": 0.79254264,
      "learning_rate": 1.5946466200473482e-06,
      "loss": 0.81435752,
      "num_input_tokens_seen": 103731185,
      "step": 4807,
      "time_per_iteration": 2.5368432998657227
    },
    {
      "auxiliary_loss_clip": 0.01144792,
      "auxiliary_loss_mlp": 0.01026137,
      "balance_loss_clip": 1.04562306,
      "balance_loss_mlp": 1.01945543,
      "epoch": 0.5781278181927494,
      "flos": 15262178883840.0,
      "grad_norm": 1.8272160797844164,
      "language_loss": 0.83588547,
      "learning_rate": 1.5938838461691723e-06,
      "loss": 0.85759479,
      "num_input_tokens_seen": 103748095,
      "step": 4808,
      "time_per_iteration": 2.531243085861206
    },
    {
      "auxiliary_loss_clip": 0.01173008,
      "auxiliary_loss_mlp": 0.01030651,
      "balance_loss_clip": 1.05305696,
      "balance_loss_mlp": 1.02258635,
      "epoch": 0.5782480610833884,
      "flos": 16726831856640.0,
      "grad_norm": 2.5545174271581406,
      "language_loss": 0.83047283,
      "learning_rate": 1.593121133898494e-06,
      "loss": 0.85250938,
      "num_input_tokens_seen": 103765300,
      "step": 4809,
      "time_per_iteration": 2.433091878890991
    },
    {
      "auxiliary_loss_clip": 0.01165392,
      "auxiliary_loss_mlp": 0.01029475,
      "balance_loss_clip": 1.05145574,
      "balance_loss_mlp": 1.02232242,
      "epoch": 0.5783683039740275,
      "flos": 25482140144640.0,
      "grad_norm": 2.014364029489039,
      "language_loss": 0.79380512,
      "learning_rate": 1.592358483351016e-06,
      "loss": 0.81575382,
      "num_input_tokens_seen": 103785475,
      "step": 4810,
      "time_per_iteration": 2.5464024543762207
    },
    {
      "auxiliary_loss_clip": 0.01154574,
      "auxiliary_loss_mlp": 0.01023304,
      "balance_loss_clip": 1.04908109,
      "balance_loss_mlp": 1.01630938,
      "epoch": 0.5784885468646667,
      "flos": 18405835240320.0,
      "grad_norm": 1.8647481767233427,
      "language_loss": 0.72746277,
      "learning_rate": 1.5915958946424326e-06,
      "loss": 0.74924159,
      "num_input_tokens_seen": 103804160,
      "step": 4811,
      "time_per_iteration": 2.468224048614502
    },
    {
      "auxiliary_loss_clip": 0.01129735,
      "auxiliary_loss_mlp": 0.00761382,
      "balance_loss_clip": 1.04620695,
      "balance_loss_mlp": 1.0001837,
      "epoch": 0.5786087897553057,
      "flos": 46100717936640.0,
      "grad_norm": 1.9540912341668029,
      "language_loss": 0.74391639,
      "learning_rate": 1.5908333678884271e-06,
      "loss": 0.76282758,
      "num_input_tokens_seen": 103830580,
      "step": 4812,
      "time_per_iteration": 2.806713104248047
    },
    {
      "auxiliary_loss_clip": 0.01156702,
      "auxiliary_loss_mlp": 0.01027954,
      "balance_loss_clip": 1.05041099,
      "balance_loss_mlp": 1.02034843,
      "epoch": 0.5787290326459448,
      "flos": 12385950261120.0,
      "grad_norm": 1.9489988764100805,
      "language_loss": 0.74197894,
      "learning_rate": 1.5900709032046743e-06,
      "loss": 0.76382554,
      "num_input_tokens_seen": 103848655,
      "step": 4813,
      "time_per_iteration": 3.2448208332061768
    },
    {
      "auxiliary_loss_clip": 0.01143786,
      "auxiliary_loss_mlp": 0.01022891,
      "balance_loss_clip": 1.05307364,
      "balance_loss_mlp": 1.01608109,
      "epoch": 0.5788492755365839,
      "flos": 23290332243840.0,
      "grad_norm": 2.564992613876135,
      "language_loss": 0.78184879,
      "learning_rate": 1.5893085007068391e-06,
      "loss": 0.80351555,
      "num_input_tokens_seen": 103866215,
      "step": 4814,
      "time_per_iteration": 2.542616605758667
    },
    {
      "auxiliary_loss_clip": 0.01132306,
      "auxiliary_loss_mlp": 0.01026974,
      "balance_loss_clip": 1.04533434,
      "balance_loss_mlp": 1.01972568,
      "epoch": 0.578969518427223,
      "flos": 24061047390720.0,
      "grad_norm": 1.9074492796256006,
      "language_loss": 0.70961332,
      "learning_rate": 1.5885461605105786e-06,
      "loss": 0.73120612,
      "num_input_tokens_seen": 103887815,
      "step": 4815,
      "time_per_iteration": 2.554541826248169
    },
    {
      "auxiliary_loss_clip": 0.01141717,
      "auxiliary_loss_mlp": 0.01030461,
      "balance_loss_clip": 1.04683793,
      "balance_loss_mlp": 1.02363038,
      "epoch": 0.579089761317862,
      "flos": 21871825269120.0,
      "grad_norm": 2.0329790301598085,
      "language_loss": 0.76677567,
      "learning_rate": 1.5877838827315375e-06,
      "loss": 0.78849745,
      "num_input_tokens_seen": 103906360,
      "step": 4816,
      "time_per_iteration": 2.5364270210266113
    },
    {
      "auxiliary_loss_clip": 0.01172415,
      "auxiliary_loss_mlp": 0.0102824,
      "balance_loss_clip": 1.0522095,
      "balance_loss_mlp": 1.02042294,
      "epoch": 0.5792100042085012,
      "flos": 22929681738240.0,
      "grad_norm": 1.6853634445819305,
      "language_loss": 0.69763494,
      "learning_rate": 1.587021667485355e-06,
      "loss": 0.71964145,
      "num_input_tokens_seen": 103925730,
      "step": 4817,
      "time_per_iteration": 2.4639716148376465
    },
    {
      "auxiliary_loss_clip": 0.01146064,
      "auxiliary_loss_mlp": 0.01027758,
      "balance_loss_clip": 1.0469135,
      "balance_loss_mlp": 1.02052164,
      "epoch": 0.5793302470991403,
      "flos": 21470056669440.0,
      "grad_norm": 1.7439349154987231,
      "language_loss": 0.78544033,
      "learning_rate": 1.5862595148876559e-06,
      "loss": 0.8071785,
      "num_input_tokens_seen": 103945835,
      "step": 4818,
      "time_per_iteration": 2.5266549587249756
    },
    {
      "auxiliary_loss_clip": 0.01117661,
      "auxiliary_loss_mlp": 0.01030268,
      "balance_loss_clip": 1.04496408,
      "balance_loss_mlp": 1.02254057,
      "epoch": 0.5794504899897793,
      "flos": 12711013367040.0,
      "grad_norm": 2.023591617017957,
      "language_loss": 0.76428401,
      "learning_rate": 1.58549742505406e-06,
      "loss": 0.78576338,
      "num_input_tokens_seen": 103960580,
      "step": 4819,
      "time_per_iteration": 3.3362550735473633
    },
    {
      "auxiliary_loss_clip": 0.01169103,
      "auxiliary_loss_mlp": 0.01025329,
      "balance_loss_clip": 1.04916465,
      "balance_loss_mlp": 1.01790774,
      "epoch": 0.5795707328804185,
      "flos": 14867054300160.0,
      "grad_norm": 2.1719049137597732,
      "language_loss": 0.75719601,
      "learning_rate": 1.5847353981001747e-06,
      "loss": 0.77914035,
      "num_input_tokens_seen": 103977760,
      "step": 4820,
      "time_per_iteration": 2.438805103302002
    },
    {
      "auxiliary_loss_clip": 0.01135454,
      "auxiliary_loss_mlp": 0.01035913,
      "balance_loss_clip": 1.04419243,
      "balance_loss_mlp": 1.02860856,
      "epoch": 0.5796909757710575,
      "flos": 36430046432640.0,
      "grad_norm": 1.5077667256079406,
      "language_loss": 0.69952196,
      "learning_rate": 1.5839734341415993e-06,
      "loss": 0.72123563,
      "num_input_tokens_seen": 103999960,
      "step": 4821,
      "time_per_iteration": 2.675907611846924
    },
    {
      "auxiliary_loss_clip": 0.01151313,
      "auxiliary_loss_mlp": 0.01027659,
      "balance_loss_clip": 1.0524509,
      "balance_loss_mlp": 1.02097738,
      "epoch": 0.5798112186616966,
      "flos": 23039891642880.0,
      "grad_norm": 2.0868016052317797,
      "language_loss": 0.76419449,
      "learning_rate": 1.5832115332939238e-06,
      "loss": 0.78598416,
      "num_input_tokens_seen": 104018400,
      "step": 4822,
      "time_per_iteration": 2.509261131286621
    },
    {
      "auxiliary_loss_clip": 0.01159483,
      "auxiliary_loss_mlp": 0.01029529,
      "balance_loss_clip": 1.05097675,
      "balance_loss_mlp": 1.02208734,
      "epoch": 0.5799314615523358,
      "flos": 16652604401280.0,
      "grad_norm": 1.764672533787325,
      "language_loss": 0.74410248,
      "learning_rate": 1.5824496956727272e-06,
      "loss": 0.76599258,
      "num_input_tokens_seen": 104035605,
      "step": 4823,
      "time_per_iteration": 3.309842109680176
    },
    {
      "auxiliary_loss_clip": 0.01140366,
      "auxiliary_loss_mlp": 0.01023892,
      "balance_loss_clip": 1.04566479,
      "balance_loss_mlp": 1.01716018,
      "epoch": 0.5800517044429748,
      "flos": 20485673470080.0,
      "grad_norm": 1.7541143132296766,
      "language_loss": 0.72825897,
      "learning_rate": 1.5816879213935797e-06,
      "loss": 0.74990153,
      "num_input_tokens_seen": 104054415,
      "step": 4824,
      "time_per_iteration": 2.5237081050872803
    },
    {
      "auxiliary_loss_clip": 0.01155229,
      "auxiliary_loss_mlp": 0.01029107,
      "balance_loss_clip": 1.05122352,
      "balance_loss_mlp": 1.02204943,
      "epoch": 0.5801719473336139,
      "flos": 31538258968320.0,
      "grad_norm": 1.4357325508714724,
      "language_loss": 0.79430044,
      "learning_rate": 1.5809262105720416e-06,
      "loss": 0.81614375,
      "num_input_tokens_seen": 104075455,
      "step": 4825,
      "time_per_iteration": 2.590930223464966
    },
    {
      "auxiliary_loss_clip": 0.01170588,
      "auxiliary_loss_mlp": 0.0102699,
      "balance_loss_clip": 1.05077457,
      "balance_loss_mlp": 1.01974797,
      "epoch": 0.580292190224253,
      "flos": 20375966355840.0,
      "grad_norm": 1.5103004928313333,
      "language_loss": 0.79248887,
      "learning_rate": 1.5801645633236644e-06,
      "loss": 0.81446469,
      "num_input_tokens_seen": 104096440,
      "step": 4826,
      "time_per_iteration": 2.4867396354675293
    },
    {
      "auxiliary_loss_clip": 0.01137819,
      "auxiliary_loss_mlp": 0.01026146,
      "balance_loss_clip": 1.04640448,
      "balance_loss_mlp": 1.0185498,
      "epoch": 0.5804124331148921,
      "flos": 26615373304320.0,
      "grad_norm": 1.7971999051742023,
      "language_loss": 0.77424788,
      "learning_rate": 1.579402979763989e-06,
      "loss": 0.79588759,
      "num_input_tokens_seen": 104116775,
      "step": 4827,
      "time_per_iteration": 2.5885190963745117
    },
    {
      "auxiliary_loss_clip": 0.01111822,
      "auxiliary_loss_mlp": 0.01023958,
      "balance_loss_clip": 1.04587209,
      "balance_loss_mlp": 1.0168891,
      "epoch": 0.5805326760055312,
      "flos": 13478496289920.0,
      "grad_norm": 1.9931154863154255,
      "language_loss": 0.81127322,
      "learning_rate": 1.578641460008548e-06,
      "loss": 0.83263099,
      "num_input_tokens_seen": 104134510,
      "step": 4828,
      "time_per_iteration": 2.589778184890747
    },
    {
      "auxiliary_loss_clip": 0.01155845,
      "auxiliary_loss_mlp": 0.01022634,
      "balance_loss_clip": 1.0506146,
      "balance_loss_mlp": 1.01590729,
      "epoch": 0.5806529188961702,
      "flos": 12091374823680.0,
      "grad_norm": 2.151955386236936,
      "language_loss": 0.67818069,
      "learning_rate": 1.5778800041728613e-06,
      "loss": 0.69996548,
      "num_input_tokens_seen": 104150800,
      "step": 4829,
      "time_per_iteration": 2.475308418273926
    },
    {
      "auxiliary_loss_clip": 0.01155533,
      "auxiliary_loss_mlp": 0.01022477,
      "balance_loss_clip": 1.05131602,
      "balance_loss_mlp": 1.01598907,
      "epoch": 0.5807731617868094,
      "flos": 26214107495040.0,
      "grad_norm": 1.4840353178405017,
      "language_loss": 0.66206712,
      "learning_rate": 1.577118612372443e-06,
      "loss": 0.68384719,
      "num_input_tokens_seen": 104172640,
      "step": 4830,
      "time_per_iteration": 2.563689708709717
    },
    {
      "auxiliary_loss_clip": 0.01139842,
      "auxiliary_loss_mlp": 0.00761613,
      "balance_loss_clip": 1.04609966,
      "balance_loss_mlp": 1.0001992,
      "epoch": 0.5808934046774484,
      "flos": 37962139190400.0,
      "grad_norm": 1.804626099858776,
      "language_loss": 0.7054882,
      "learning_rate": 1.5763572847227943e-06,
      "loss": 0.7245028,
      "num_input_tokens_seen": 104193525,
      "step": 4831,
      "time_per_iteration": 2.68552303314209
    },
    {
      "auxiliary_loss_clip": 0.01153853,
      "auxiliary_loss_mlp": 0.01021301,
      "balance_loss_clip": 1.04659534,
      "balance_loss_mlp": 1.01490188,
      "epoch": 0.5810136475680875,
      "flos": 20485853038080.0,
      "grad_norm": 1.7830408934785524,
      "language_loss": 0.81329179,
      "learning_rate": 1.5755960213394091e-06,
      "loss": 0.83504337,
      "num_input_tokens_seen": 104210625,
      "step": 4832,
      "time_per_iteration": 2.511032819747925
    },
    {
      "auxiliary_loss_clip": 0.01129885,
      "auxiliary_loss_mlp": 0.01026376,
      "balance_loss_clip": 1.04633427,
      "balance_loss_mlp": 1.01909828,
      "epoch": 0.5811338904587267,
      "flos": 17530153574400.0,
      "grad_norm": 1.736493882798787,
      "language_loss": 0.7818318,
      "learning_rate": 1.5748348223377703e-06,
      "loss": 0.80339444,
      "num_input_tokens_seen": 104228180,
      "step": 4833,
      "time_per_iteration": 2.550215482711792
    },
    {
      "auxiliary_loss_clip": 0.01140542,
      "auxiliary_loss_mlp": 0.0102773,
      "balance_loss_clip": 1.04910111,
      "balance_loss_mlp": 1.02049422,
      "epoch": 0.5812541333493657,
      "flos": 19458017360640.0,
      "grad_norm": 1.6290558694031396,
      "language_loss": 0.77882922,
      "learning_rate": 1.5740736878333507e-06,
      "loss": 0.80051196,
      "num_input_tokens_seen": 104246020,
      "step": 4834,
      "time_per_iteration": 2.517512798309326
    },
    {
      "auxiliary_loss_clip": 0.01145884,
      "auxiliary_loss_mlp": 0.01027227,
      "balance_loss_clip": 1.04747295,
      "balance_loss_mlp": 1.02022052,
      "epoch": 0.5813743762400048,
      "flos": 20594949621120.0,
      "grad_norm": 2.6649396195790223,
      "language_loss": 0.7789712,
      "learning_rate": 1.5733126179416143e-06,
      "loss": 0.80070233,
      "num_input_tokens_seen": 104260505,
      "step": 4835,
      "time_per_iteration": 2.5310263633728027
    },
    {
      "auxiliary_loss_clip": 0.01156111,
      "auxiliary_loss_mlp": 0.01029591,
      "balance_loss_clip": 1.05075371,
      "balance_loss_mlp": 1.02220941,
      "epoch": 0.5814946191306439,
      "flos": 33178227246720.0,
      "grad_norm": 2.203887264801454,
      "language_loss": 0.72258753,
      "learning_rate": 1.5725516127780137e-06,
      "loss": 0.74444455,
      "num_input_tokens_seen": 104282640,
      "step": 4836,
      "time_per_iteration": 2.614704132080078
    },
    {
      "auxiliary_loss_clip": 0.01162972,
      "auxiliary_loss_mlp": 0.0102463,
      "balance_loss_clip": 1.0497992,
      "balance_loss_mlp": 1.01742387,
      "epoch": 0.581614862021283,
      "flos": 16143283503360.0,
      "grad_norm": 2.0437911840252108,
      "language_loss": 0.88016498,
      "learning_rate": 1.5717906724579943e-06,
      "loss": 0.90204096,
      "num_input_tokens_seen": 104299700,
      "step": 4837,
      "time_per_iteration": 2.4780101776123047
    },
    {
      "auxiliary_loss_clip": 0.0113424,
      "auxiliary_loss_mlp": 0.01025279,
      "balance_loss_clip": 1.04548025,
      "balance_loss_mlp": 1.01865363,
      "epoch": 0.581735104911922,
      "flos": 33802642298880.0,
      "grad_norm": 2.1670632958790486,
      "language_loss": 0.68286425,
      "learning_rate": 1.571029797096989e-06,
      "loss": 0.70445943,
      "num_input_tokens_seen": 104320805,
      "step": 4838,
      "time_per_iteration": 2.670537233352661
    },
    {
      "auxiliary_loss_clip": 0.01168804,
      "auxiliary_loss_mlp": 0.01023649,
      "balance_loss_clip": 1.05059958,
      "balance_loss_mlp": 1.01640701,
      "epoch": 0.5818553478025612,
      "flos": 23331163029120.0,
      "grad_norm": 1.7255514903728824,
      "language_loss": 0.78619874,
      "learning_rate": 1.570268986810423e-06,
      "loss": 0.80812323,
      "num_input_tokens_seen": 104340700,
      "step": 4839,
      "time_per_iteration": 3.247474193572998
    },
    {
      "auxiliary_loss_clip": 0.0114179,
      "auxiliary_loss_mlp": 0.01028851,
      "balance_loss_clip": 1.04893041,
      "balance_loss_mlp": 1.02171588,
      "epoch": 0.5819755906932003,
      "flos": 20996143603200.0,
      "grad_norm": 1.8612386448662739,
      "language_loss": 0.74944997,
      "learning_rate": 1.5695082417137096e-06,
      "loss": 0.77115643,
      "num_input_tokens_seen": 104358575,
      "step": 4840,
      "time_per_iteration": 2.5304155349731445
    },
    {
      "auxiliary_loss_clip": 0.01139625,
      "auxiliary_loss_mlp": 0.01026164,
      "balance_loss_clip": 1.04583526,
      "balance_loss_mlp": 1.01907134,
      "epoch": 0.5820958335838393,
      "flos": 21431668008960.0,
      "grad_norm": 1.5054334857171527,
      "language_loss": 0.74980873,
      "learning_rate": 1.5687475619222539e-06,
      "loss": 0.77146661,
      "num_input_tokens_seen": 104378530,
      "step": 4841,
      "time_per_iteration": 2.5731375217437744
    },
    {
      "auxiliary_loss_clip": 0.01134502,
      "auxiliary_loss_mlp": 0.01032783,
      "balance_loss_clip": 1.0439682,
      "balance_loss_mlp": 1.02535069,
      "epoch": 0.5822160764744785,
      "flos": 17967473660160.0,
      "grad_norm": 2.705172044043733,
      "language_loss": 0.73709035,
      "learning_rate": 1.5679869475514496e-06,
      "loss": 0.75876319,
      "num_input_tokens_seen": 104395465,
      "step": 4842,
      "time_per_iteration": 2.553636074066162
    },
    {
      "auxiliary_loss_clip": 0.01160128,
      "auxiliary_loss_mlp": 0.01028431,
      "balance_loss_clip": 1.05244398,
      "balance_loss_mlp": 1.02094173,
      "epoch": 0.5823363193651175,
      "flos": 23033858158080.0,
      "grad_norm": 1.9946328162978828,
      "language_loss": 0.81211591,
      "learning_rate": 1.567226398716682e-06,
      "loss": 0.83400154,
      "num_input_tokens_seen": 104415380,
      "step": 4843,
      "time_per_iteration": 2.5081214904785156
    },
    {
      "auxiliary_loss_clip": 0.0114744,
      "auxiliary_loss_mlp": 0.0102792,
      "balance_loss_clip": 1.04703403,
      "balance_loss_mlp": 1.02057028,
      "epoch": 0.5824565622557566,
      "flos": 32891840110080.0,
      "grad_norm": 1.855424445451852,
      "language_loss": 0.62014365,
      "learning_rate": 1.566465915533326e-06,
      "loss": 0.6418972,
      "num_input_tokens_seen": 104437410,
      "step": 4844,
      "time_per_iteration": 2.643021583557129
    },
    {
      "auxiliary_loss_clip": 0.01153604,
      "auxiliary_loss_mlp": 0.0103041,
      "balance_loss_clip": 1.04874587,
      "balance_loss_mlp": 1.02296507,
      "epoch": 0.5825768051463958,
      "flos": 22229674513920.0,
      "grad_norm": 1.9382944751951445,
      "language_loss": 0.88139904,
      "learning_rate": 1.5657054981167458e-06,
      "loss": 0.90323919,
      "num_input_tokens_seen": 104456305,
      "step": 4845,
      "time_per_iteration": 3.2869231700897217
    },
    {
      "auxiliary_loss_clip": 0.0115611,
      "auxiliary_loss_mlp": 0.01026086,
      "balance_loss_clip": 1.05037045,
      "balance_loss_mlp": 1.01892161,
      "epoch": 0.5826970480370348,
      "flos": 28001561016960.0,
      "grad_norm": 2.0815856332815392,
      "language_loss": 0.67584372,
      "learning_rate": 1.5649451465822965e-06,
      "loss": 0.69766569,
      "num_input_tokens_seen": 104477695,
      "step": 4846,
      "time_per_iteration": 2.611398220062256
    },
    {
      "auxiliary_loss_clip": 0.01113912,
      "auxiliary_loss_mlp": 0.01029048,
      "balance_loss_clip": 1.04682064,
      "balance_loss_mlp": 1.02206492,
      "epoch": 0.5828172909276739,
      "flos": 17858053854720.0,
      "grad_norm": 1.6453948273032635,
      "language_loss": 0.8337611,
      "learning_rate": 1.5641848610453218e-06,
      "loss": 0.85519069,
      "num_input_tokens_seen": 104496355,
      "step": 4847,
      "time_per_iteration": 2.560307264328003
    },
    {
      "auxiliary_loss_clip": 0.01154072,
      "auxiliary_loss_mlp": 0.01026838,
      "balance_loss_clip": 1.05106735,
      "balance_loss_mlp": 1.01930428,
      "epoch": 0.582937533818313,
      "flos": 19865244827520.0,
      "grad_norm": 1.9090144995555804,
      "language_loss": 0.85794461,
      "learning_rate": 1.563424641621158e-06,
      "loss": 0.87975371,
      "num_input_tokens_seen": 104515535,
      "step": 4848,
      "time_per_iteration": 2.5104312896728516
    },
    {
      "auxiliary_loss_clip": 0.01146396,
      "auxiliary_loss_mlp": 0.01024782,
      "balance_loss_clip": 1.04857326,
      "balance_loss_mlp": 1.01637816,
      "epoch": 0.5830577767089521,
      "flos": 26870734068480.0,
      "grad_norm": 2.2221618015082427,
      "language_loss": 0.69987863,
      "learning_rate": 1.5626644884251282e-06,
      "loss": 0.7215904,
      "num_input_tokens_seen": 104535055,
      "step": 4849,
      "time_per_iteration": 3.3997108936309814
    },
    {
      "auxiliary_loss_clip": 0.01169659,
      "auxiliary_loss_mlp": 0.01024323,
      "balance_loss_clip": 1.05113292,
      "balance_loss_mlp": 1.01707792,
      "epoch": 0.5831780195995911,
      "flos": 25298205575040.0,
      "grad_norm": 1.74805622058221,
      "language_loss": 0.8809036,
      "learning_rate": 1.5619044015725488e-06,
      "loss": 0.90284348,
      "num_input_tokens_seen": 104554745,
      "step": 4850,
      "time_per_iteration": 2.4941868782043457
    },
    {
      "auxiliary_loss_clip": 0.01178758,
      "auxiliary_loss_mlp": 0.01031044,
      "balance_loss_clip": 1.05652821,
      "balance_loss_mlp": 1.02315271,
      "epoch": 0.5832982624902303,
      "flos": 14756988049920.0,
      "grad_norm": 2.2189850639961706,
      "language_loss": 0.87270528,
      "learning_rate": 1.5611443811787224e-06,
      "loss": 0.89480329,
      "num_input_tokens_seen": 104568870,
      "step": 4851,
      "time_per_iteration": 2.434347629547119
    },
    {
      "auxiliary_loss_clip": 0.0115671,
      "auxiliary_loss_mlp": 0.01028124,
      "balance_loss_clip": 1.05247247,
      "balance_loss_mlp": 1.02157044,
      "epoch": 0.5834185053808694,
      "flos": 20444555376000.0,
      "grad_norm": 2.168894311170541,
      "language_loss": 0.69712985,
      "learning_rate": 1.560384427358945e-06,
      "loss": 0.71897817,
      "num_input_tokens_seen": 104588415,
      "step": 4852,
      "time_per_iteration": 2.489314317703247
    },
    {
      "auxiliary_loss_clip": 0.01135257,
      "auxiliary_loss_mlp": 0.01027829,
      "balance_loss_clip": 1.04399776,
      "balance_loss_mlp": 1.02024138,
      "epoch": 0.5835387482715084,
      "flos": 27200394115200.0,
      "grad_norm": 1.4829691165411758,
      "language_loss": 0.73171854,
      "learning_rate": 1.5596245402284998e-06,
      "loss": 0.75334936,
      "num_input_tokens_seen": 104611940,
      "step": 4853,
      "time_per_iteration": 2.5984067916870117
    },
    {
      "auxiliary_loss_clip": 0.01159588,
      "auxiliary_loss_mlp": 0.01028992,
      "balance_loss_clip": 1.05269468,
      "balance_loss_mlp": 1.02154756,
      "epoch": 0.5836589911621476,
      "flos": 16654615562880.0,
      "grad_norm": 1.7514720525333853,
      "language_loss": 0.81907368,
      "learning_rate": 1.5588647199026619e-06,
      "loss": 0.84095943,
      "num_input_tokens_seen": 104629675,
      "step": 4854,
      "time_per_iteration": 2.4673664569854736
    },
    {
      "auxiliary_loss_clip": 0.01175494,
      "auxiliary_loss_mlp": 0.0102628,
      "balance_loss_clip": 1.05386484,
      "balance_loss_mlp": 1.01910067,
      "epoch": 0.5837792340527866,
      "flos": 20446817932800.0,
      "grad_norm": 2.0695717459997294,
      "language_loss": 0.87748593,
      "learning_rate": 1.5581049664966956e-06,
      "loss": 0.89950371,
      "num_input_tokens_seen": 104647435,
      "step": 4855,
      "time_per_iteration": 2.442946672439575
    },
    {
      "auxiliary_loss_clip": 0.01011156,
      "auxiliary_loss_mlp": 0.01002434,
      "balance_loss_clip": 1.0102632,
      "balance_loss_mlp": 1.00127149,
      "epoch": 0.5838994769434257,
      "flos": 65995480765440.0,
      "grad_norm": 0.9904263493673009,
      "language_loss": 0.65204132,
      "learning_rate": 1.5573452801258545e-06,
      "loss": 0.6721772,
      "num_input_tokens_seen": 104694605,
      "step": 4856,
      "time_per_iteration": 3.0067784786224365
    },
    {
      "auxiliary_loss_clip": 0.01164211,
      "auxiliary_loss_mlp": 0.01033255,
      "balance_loss_clip": 1.05262136,
      "balance_loss_mlp": 1.02583992,
      "epoch": 0.5840197198340649,
      "flos": 21470523546240.0,
      "grad_norm": 1.9455751269205175,
      "language_loss": 0.63345504,
      "learning_rate": 1.5565856609053824e-06,
      "loss": 0.65542966,
      "num_input_tokens_seen": 104713400,
      "step": 4857,
      "time_per_iteration": 2.505228042602539
    },
    {
      "auxiliary_loss_clip": 0.01171769,
      "auxiliary_loss_mlp": 0.01030231,
      "balance_loss_clip": 1.05277121,
      "balance_loss_mlp": 1.02264333,
      "epoch": 0.5841399627247039,
      "flos": 19135144984320.0,
      "grad_norm": 38.60943201965022,
      "language_loss": 0.79598427,
      "learning_rate": 1.5558261089505127e-06,
      "loss": 0.81800425,
      "num_input_tokens_seen": 104732130,
      "step": 4858,
      "time_per_iteration": 2.449418306350708
    },
    {
      "auxiliary_loss_clip": 0.01157949,
      "auxiliary_loss_mlp": 0.01029704,
      "balance_loss_clip": 1.05342555,
      "balance_loss_mlp": 1.02258992,
      "epoch": 0.584260205615343,
      "flos": 26425692558720.0,
      "grad_norm": 1.8760304064627622,
      "language_loss": 0.80200237,
      "learning_rate": 1.5550666243764697e-06,
      "loss": 0.82387888,
      "num_input_tokens_seen": 104750290,
      "step": 4859,
      "time_per_iteration": 2.5435678958892822
    },
    {
      "auxiliary_loss_clip": 0.01156838,
      "auxiliary_loss_mlp": 0.01027127,
      "balance_loss_clip": 1.05095088,
      "balance_loss_mlp": 1.01989126,
      "epoch": 0.584380448505982,
      "flos": 13881809174400.0,
      "grad_norm": 2.1883210242533027,
      "language_loss": 0.77089959,
      "learning_rate": 1.554307207298465e-06,
      "loss": 0.79273921,
      "num_input_tokens_seen": 104768550,
      "step": 4860,
      "time_per_iteration": 2.467388868331909
    },
    {
      "auxiliary_loss_clip": 0.01174378,
      "auxiliary_loss_mlp": 0.01033908,
      "balance_loss_clip": 1.05320644,
      "balance_loss_mlp": 1.02593541,
      "epoch": 0.5845006913966212,
      "flos": 21543709507200.0,
      "grad_norm": 3.880492985685576,
      "language_loss": 0.78467155,
      "learning_rate": 1.553547857831704e-06,
      "loss": 0.80675441,
      "num_input_tokens_seen": 104785060,
      "step": 4861,
      "time_per_iteration": 2.479334831237793
    },
    {
      "auxiliary_loss_clip": 0.01060401,
      "auxiliary_loss_mlp": 0.0100135,
      "balance_loss_clip": 1.01187062,
      "balance_loss_mlp": 1.0002594,
      "epoch": 0.5846209342872603,
      "flos": 58375452712320.0,
      "grad_norm": 1.5086142326966725,
      "language_loss": 0.64268041,
      "learning_rate": 1.5527885760913771e-06,
      "loss": 0.66329801,
      "num_input_tokens_seen": 104834950,
      "step": 4862,
      "time_per_iteration": 2.901493787765503
    },
    {
      "auxiliary_loss_clip": 0.0114583,
      "auxiliary_loss_mlp": 0.01031549,
      "balance_loss_clip": 1.05089855,
      "balance_loss_mlp": 1.02503991,
      "epoch": 0.5847411771778993,
      "flos": 18588045957120.0,
      "grad_norm": 1.781349329639557,
      "language_loss": 0.76385748,
      "learning_rate": 1.552029362192668e-06,
      "loss": 0.7856313,
      "num_input_tokens_seen": 104854210,
      "step": 4863,
      "time_per_iteration": 2.5201826095581055
    },
    {
      "auxiliary_loss_clip": 0.01127093,
      "auxiliary_loss_mlp": 0.01032495,
      "balance_loss_clip": 1.04727793,
      "balance_loss_mlp": 1.02537227,
      "epoch": 0.5848614200685385,
      "flos": 24240780069120.0,
      "grad_norm": 1.9242041392750093,
      "language_loss": 0.72349977,
      "learning_rate": 1.5512702162507478e-06,
      "loss": 0.74509567,
      "num_input_tokens_seen": 104874525,
      "step": 4864,
      "time_per_iteration": 2.594768762588501
    },
    {
      "auxiliary_loss_clip": 0.01040852,
      "auxiliary_loss_mlp": 0.01001173,
      "balance_loss_clip": 1.01238108,
      "balance_loss_mlp": 1.0001297,
      "epoch": 0.5849816629591775,
      "flos": 71660245933440.0,
      "grad_norm": 1.1278782501392024,
      "language_loss": 0.55818605,
      "learning_rate": 1.5505111383807792e-06,
      "loss": 0.57860625,
      "num_input_tokens_seen": 104937195,
      "step": 4865,
      "time_per_iteration": 3.936511516571045
    },
    {
      "auxiliary_loss_clip": 0.01116636,
      "auxiliary_loss_mlp": 0.01022055,
      "balance_loss_clip": 1.04273975,
      "balance_loss_mlp": 1.01515293,
      "epoch": 0.5851019058498166,
      "flos": 23802095266560.0,
      "grad_norm": 1.7591303506194025,
      "language_loss": 0.80455691,
      "learning_rate": 1.5497521286979138e-06,
      "loss": 0.82594383,
      "num_input_tokens_seen": 104957435,
      "step": 4866,
      "time_per_iteration": 2.6219053268432617
    },
    {
      "auxiliary_loss_clip": 0.01134632,
      "auxiliary_loss_mlp": 0.0102679,
      "balance_loss_clip": 1.04861307,
      "balance_loss_mlp": 1.01936293,
      "epoch": 0.5852221487404557,
      "flos": 24388516707840.0,
      "grad_norm": 2.385620986582856,
      "language_loss": 0.74255443,
      "learning_rate": 1.5489931873172927e-06,
      "loss": 0.76416862,
      "num_input_tokens_seen": 104978755,
      "step": 4867,
      "time_per_iteration": 2.5996220111846924
    },
    {
      "auxiliary_loss_clip": 0.0108198,
      "auxiliary_loss_mlp": 0.010258,
      "balance_loss_clip": 1.03809953,
      "balance_loss_mlp": 1.01817071,
      "epoch": 0.5853423916310948,
      "flos": 27271425260160.0,
      "grad_norm": 1.6898278836169136,
      "language_loss": 0.7918334,
      "learning_rate": 1.5482343143540467e-06,
      "loss": 0.81291115,
      "num_input_tokens_seen": 105000020,
      "step": 4868,
      "time_per_iteration": 2.6843981742858887
    },
    {
      "auxiliary_loss_clip": 0.01128419,
      "auxiliary_loss_mlp": 0.00760852,
      "balance_loss_clip": 1.04610157,
      "balance_loss_mlp": 1.00020313,
      "epoch": 0.5854626345217339,
      "flos": 11983786611840.0,
      "grad_norm": 1.9137062332089865,
      "language_loss": 0.82431877,
      "learning_rate": 1.547475509923295e-06,
      "loss": 0.84321153,
      "num_input_tokens_seen": 105017060,
      "step": 4869,
      "time_per_iteration": 2.5343446731567383
    },
    {
      "auxiliary_loss_clip": 0.0101965,
      "auxiliary_loss_mlp": 0.00999801,
      "balance_loss_clip": 1.01192594,
      "balance_loss_mlp": 0.99870473,
      "epoch": 0.585582877412373,
      "flos": 64342335173760.0,
      "grad_norm": 0.7328556917102086,
      "language_loss": 0.56109107,
      "learning_rate": 1.5467167741401495e-06,
      "loss": 0.5812856,
      "num_input_tokens_seen": 105078540,
      "step": 4870,
      "time_per_iteration": 3.1773455142974854
    },
    {
      "auxiliary_loss_clip": 0.01140969,
      "auxiliary_loss_mlp": 0.01029339,
      "balance_loss_clip": 1.04522038,
      "balance_loss_mlp": 1.02212405,
      "epoch": 0.5857031203030121,
      "flos": 17011926103680.0,
      "grad_norm": 1.980483207224238,
      "language_loss": 0.7140227,
      "learning_rate": 1.5459581071197083e-06,
      "loss": 0.73572576,
      "num_input_tokens_seen": 105094200,
      "step": 4871,
      "time_per_iteration": 3.297919511795044
    },
    {
      "auxiliary_loss_clip": 0.01162515,
      "auxiliary_loss_mlp": 0.01029342,
      "balance_loss_clip": 1.05389869,
      "balance_loss_mlp": 1.02180839,
      "epoch": 0.5858233631936511,
      "flos": 20885682303360.0,
      "grad_norm": 2.0890385543943912,
      "language_loss": 0.83430851,
      "learning_rate": 1.5451995089770624e-06,
      "loss": 0.85622716,
      "num_input_tokens_seen": 105113985,
      "step": 4872,
      "time_per_iteration": 2.509453058242798
    },
    {
      "auxiliary_loss_clip": 0.01172632,
      "auxiliary_loss_mlp": 0.01023995,
      "balance_loss_clip": 1.05307484,
      "balance_loss_mlp": 1.01723254,
      "epoch": 0.5859436060842903,
      "flos": 23191902000000.0,
      "grad_norm": 1.3105149828120086,
      "language_loss": 0.72061276,
      "learning_rate": 1.5444409798272885e-06,
      "loss": 0.74257898,
      "num_input_tokens_seen": 105138075,
      "step": 4873,
      "time_per_iteration": 2.5764496326446533
    },
    {
      "auxiliary_loss_clip": 0.01131271,
      "auxiliary_loss_mlp": 0.01027395,
      "balance_loss_clip": 1.04639709,
      "balance_loss_mlp": 1.01955712,
      "epoch": 0.5860638489749294,
      "flos": 22492648961280.0,
      "grad_norm": 1.750430969542323,
      "language_loss": 0.8051101,
      "learning_rate": 1.543682519785456e-06,
      "loss": 0.82669675,
      "num_input_tokens_seen": 105156555,
      "step": 4874,
      "time_per_iteration": 2.5816845893859863
    },
    {
      "auxiliary_loss_clip": 0.01142874,
      "auxiliary_loss_mlp": 0.01026422,
      "balance_loss_clip": 1.04789639,
      "balance_loss_mlp": 1.01961482,
      "epoch": 0.5861840918655684,
      "flos": 17566243764480.0,
      "grad_norm": 2.4378225309203603,
      "language_loss": 0.80179513,
      "learning_rate": 1.5429241289666219e-06,
      "loss": 0.823488,
      "num_input_tokens_seen": 105174055,
      "step": 4875,
      "time_per_iteration": 4.095436096191406
    },
    {
      "auxiliary_loss_clip": 0.0113479,
      "auxiliary_loss_mlp": 0.01031047,
      "balance_loss_clip": 1.04615784,
      "balance_loss_mlp": 1.02365017,
      "epoch": 0.5863043347562076,
      "flos": 25556152118400.0,
      "grad_norm": 1.9970828984536362,
      "language_loss": 0.69383073,
      "learning_rate": 1.5421658074858342e-06,
      "loss": 0.71548915,
      "num_input_tokens_seen": 105192160,
      "step": 4876,
      "time_per_iteration": 2.5670664310455322
    },
    {
      "auxiliary_loss_clip": 0.01138719,
      "auxiliary_loss_mlp": 0.01032843,
      "balance_loss_clip": 1.04754591,
      "balance_loss_mlp": 1.0252552,
      "epoch": 0.5864245776468466,
      "flos": 20667525050880.0,
      "grad_norm": 2.182059164787899,
      "language_loss": 0.66334164,
      "learning_rate": 1.5414075554581298e-06,
      "loss": 0.68505728,
      "num_input_tokens_seen": 105210205,
      "step": 4877,
      "time_per_iteration": 2.533712387084961
    },
    {
      "auxiliary_loss_clip": 0.01172528,
      "auxiliary_loss_mlp": 0.01027847,
      "balance_loss_clip": 1.05179,
      "balance_loss_mlp": 1.02023566,
      "epoch": 0.5865448205374857,
      "flos": 28913907490560.0,
      "grad_norm": 2.367192826429621,
      "language_loss": 0.78074396,
      "learning_rate": 1.5406493729985348e-06,
      "loss": 0.80274773,
      "num_input_tokens_seen": 105229400,
      "step": 4878,
      "time_per_iteration": 2.516009569168091
    },
    {
      "auxiliary_loss_clip": 0.01116542,
      "auxiliary_loss_mlp": 0.00760687,
      "balance_loss_clip": 1.04660869,
      "balance_loss_mlp": 1.00016594,
      "epoch": 0.5866650634281249,
      "flos": 25842575168640.0,
      "grad_norm": 2.222934135137192,
      "language_loss": 0.71407384,
      "learning_rate": 1.5398912602220644e-06,
      "loss": 0.73284614,
      "num_input_tokens_seen": 105248675,
      "step": 4879,
      "time_per_iteration": 2.646860122680664
    },
    {
      "auxiliary_loss_clip": 0.0112232,
      "auxiliary_loss_mlp": 0.01028019,
      "balance_loss_clip": 1.04613137,
      "balance_loss_mlp": 1.02063704,
      "epoch": 0.5867853063187639,
      "flos": 17052325925760.0,
      "grad_norm": 4.128134555490838,
      "language_loss": 0.79115534,
      "learning_rate": 1.539133217243724e-06,
      "loss": 0.81265867,
      "num_input_tokens_seen": 105265695,
      "step": 4880,
      "time_per_iteration": 2.557936906814575
    },
    {
      "auxiliary_loss_clip": 0.01136185,
      "auxiliary_loss_mlp": 0.01030464,
      "balance_loss_clip": 1.04588652,
      "balance_loss_mlp": 1.02208996,
      "epoch": 0.586905549209403,
      "flos": 24645026707200.0,
      "grad_norm": 4.2087026675962536,
      "language_loss": 0.76247954,
      "learning_rate": 1.5383752441785081e-06,
      "loss": 0.78414607,
      "num_input_tokens_seen": 105284920,
      "step": 4881,
      "time_per_iteration": 2.606729030609131
    },
    {
      "auxiliary_loss_clip": 0.01164171,
      "auxiliary_loss_mlp": 0.01031154,
      "balance_loss_clip": 1.05215085,
      "balance_loss_mlp": 1.02350116,
      "epoch": 0.5870257921000421,
      "flos": 14720538723840.0,
      "grad_norm": 2.2060155545073927,
      "language_loss": 0.85456866,
      "learning_rate": 1.5376173411414003e-06,
      "loss": 0.87652194,
      "num_input_tokens_seen": 105302960,
      "step": 4882,
      "time_per_iteration": 2.474823236465454
    },
    {
      "auxiliary_loss_clip": 0.01143543,
      "auxiliary_loss_mlp": 0.01029326,
      "balance_loss_clip": 1.0452373,
      "balance_loss_mlp": 1.02216446,
      "epoch": 0.5871460349906812,
      "flos": 23914998691200.0,
      "grad_norm": 1.884649094622077,
      "language_loss": 0.78933263,
      "learning_rate": 1.5368595082473753e-06,
      "loss": 0.81106132,
      "num_input_tokens_seen": 105321260,
      "step": 4883,
      "time_per_iteration": 2.5613725185394287
    },
    {
      "auxiliary_loss_clip": 0.0115608,
      "auxiliary_loss_mlp": 0.01022139,
      "balance_loss_clip": 1.04795837,
      "balance_loss_mlp": 1.01514721,
      "epoch": 0.5872662778813202,
      "flos": 22164174063360.0,
      "grad_norm": 1.7293067771025656,
      "language_loss": 0.78209245,
      "learning_rate": 1.5361017456113935e-06,
      "loss": 0.80387461,
      "num_input_tokens_seen": 105341610,
      "step": 4884,
      "time_per_iteration": 2.5066311359405518
    },
    {
      "auxiliary_loss_clip": 0.01157438,
      "auxiliary_loss_mlp": 0.01027175,
      "balance_loss_clip": 1.04807043,
      "balance_loss_mlp": 1.01930165,
      "epoch": 0.5873865207719594,
      "flos": 18441925430400.0,
      "grad_norm": 2.019789961321854,
      "language_loss": 0.85467267,
      "learning_rate": 1.5353440533484085e-06,
      "loss": 0.87651879,
      "num_input_tokens_seen": 105360465,
      "step": 4885,
      "time_per_iteration": 2.512587070465088
    },
    {
      "auxiliary_loss_clip": 0.0114568,
      "auxiliary_loss_mlp": 0.01025608,
      "balance_loss_clip": 1.04821312,
      "balance_loss_mlp": 1.01761544,
      "epoch": 0.5875067636625985,
      "flos": 54015321427200.0,
      "grad_norm": 1.6979826261719433,
      "language_loss": 0.66163158,
      "learning_rate": 1.534586431573361e-06,
      "loss": 0.68334442,
      "num_input_tokens_seen": 105385405,
      "step": 4886,
      "time_per_iteration": 2.807114362716675
    },
    {
      "auxiliary_loss_clip": 0.01104595,
      "auxiliary_loss_mlp": 0.01031011,
      "balance_loss_clip": 1.04142225,
      "balance_loss_mlp": 1.0232203,
      "epoch": 0.5876270065532375,
      "flos": 27995707100160.0,
      "grad_norm": 1.6898003988532517,
      "language_loss": 0.7902602,
      "learning_rate": 1.5338288804011817e-06,
      "loss": 0.81161624,
      "num_input_tokens_seen": 105404905,
      "step": 4887,
      "time_per_iteration": 2.6721949577331543
    },
    {
      "auxiliary_loss_clip": 0.01138439,
      "auxiliary_loss_mlp": 0.0102674,
      "balance_loss_clip": 1.04510462,
      "balance_loss_mlp": 1.01922941,
      "epoch": 0.5877472494438767,
      "flos": 21361462876800.0,
      "grad_norm": 1.919845247092281,
      "language_loss": 0.71005845,
      "learning_rate": 1.533071399946791e-06,
      "loss": 0.73171031,
      "num_input_tokens_seen": 105423650,
      "step": 4888,
      "time_per_iteration": 2.518092393875122
    },
    {
      "auxiliary_loss_clip": 0.01142624,
      "auxiliary_loss_mlp": 0.01028185,
      "balance_loss_clip": 1.04807734,
      "balance_loss_mlp": 1.02096677,
      "epoch": 0.5878674923345157,
      "flos": 22383013674240.0,
      "grad_norm": 2.153065227257642,
      "language_loss": 0.57670295,
      "learning_rate": 1.5323139903250977e-06,
      "loss": 0.59841108,
      "num_input_tokens_seen": 105444255,
      "step": 4889,
      "time_per_iteration": 2.5541951656341553
    },
    {
      "auxiliary_loss_clip": 0.01148484,
      "auxiliary_loss_mlp": 0.0102826,
      "balance_loss_clip": 1.05215096,
      "balance_loss_mlp": 1.02079153,
      "epoch": 0.5879877352251548,
      "flos": 21868664872320.0,
      "grad_norm": 1.4786514369628596,
      "language_loss": 0.77139497,
      "learning_rate": 1.5315566516510002e-06,
      "loss": 0.79316235,
      "num_input_tokens_seen": 105462425,
      "step": 4890,
      "time_per_iteration": 2.535137891769409
    },
    {
      "auxiliary_loss_clip": 0.01172306,
      "auxiliary_loss_mlp": 0.01025783,
      "balance_loss_clip": 1.05195308,
      "balance_loss_mlp": 1.01837087,
      "epoch": 0.5881079781157939,
      "flos": 17493811989120.0,
      "grad_norm": 1.6585637110155793,
      "language_loss": 0.67828214,
      "learning_rate": 1.5307993840393857e-06,
      "loss": 0.70026302,
      "num_input_tokens_seen": 105480505,
      "step": 4891,
      "time_per_iteration": 3.297316789627075
    },
    {
      "auxiliary_loss_clip": 0.01169822,
      "auxiliary_loss_mlp": 0.0102721,
      "balance_loss_clip": 1.0504024,
      "balance_loss_mlp": 1.02044749,
      "epoch": 0.588228221006433,
      "flos": 22601853285120.0,
      "grad_norm": 1.858044534124567,
      "language_loss": 0.80612922,
      "learning_rate": 1.530042187605132e-06,
      "loss": 0.82809949,
      "num_input_tokens_seen": 105499760,
      "step": 4892,
      "time_per_iteration": 2.4902749061584473
    },
    {
      "auxiliary_loss_clip": 0.0115904,
      "auxiliary_loss_mlp": 0.00760005,
      "balance_loss_clip": 1.05031109,
      "balance_loss_mlp": 1.00013566,
      "epoch": 0.5883484638970721,
      "flos": 26176939896960.0,
      "grad_norm": 1.4063164877226104,
      "language_loss": 0.84168983,
      "learning_rate": 1.5292850624631044e-06,
      "loss": 0.86088026,
      "num_input_tokens_seen": 105521955,
      "step": 4893,
      "time_per_iteration": 2.5578320026397705
    },
    {
      "auxiliary_loss_clip": 0.01151644,
      "auxiliary_loss_mlp": 0.01030909,
      "balance_loss_clip": 1.04943359,
      "balance_loss_mlp": 1.02373266,
      "epoch": 0.5884687067877111,
      "flos": 30443737691520.0,
      "grad_norm": 1.8820540599268687,
      "language_loss": 0.80111945,
      "learning_rate": 1.5285280087281593e-06,
      "loss": 0.822945,
      "num_input_tokens_seen": 105542685,
      "step": 4894,
      "time_per_iteration": 2.5754315853118896
    },
    {
      "auxiliary_loss_clip": 0.01042978,
      "auxiliary_loss_mlp": 0.01008085,
      "balance_loss_clip": 1.01260996,
      "balance_loss_mlp": 1.00676811,
      "epoch": 0.5885889496783503,
      "flos": 70507550580480.0,
      "grad_norm": 0.6482687509409895,
      "language_loss": 0.56631756,
      "learning_rate": 1.5277710265151398e-06,
      "loss": 0.58682823,
      "num_input_tokens_seen": 105612165,
      "step": 4895,
      "time_per_iteration": 3.283946990966797
    },
    {
      "auxiliary_loss_clip": 0.01156842,
      "auxiliary_loss_mlp": 0.01029021,
      "balance_loss_clip": 1.0489471,
      "balance_loss_mlp": 1.02111745,
      "epoch": 0.5887091925689893,
      "flos": 19098767485440.0,
      "grad_norm": 2.920746522994618,
      "language_loss": 0.77364838,
      "learning_rate": 1.5270141159388803e-06,
      "loss": 0.79550707,
      "num_input_tokens_seen": 105629185,
      "step": 4896,
      "time_per_iteration": 2.4842166900634766
    },
    {
      "auxiliary_loss_clip": 0.01169999,
      "auxiliary_loss_mlp": 0.01029543,
      "balance_loss_clip": 1.05002594,
      "balance_loss_mlp": 1.02151442,
      "epoch": 0.5888294354596284,
      "flos": 23294282739840.0,
      "grad_norm": 1.6388209494567,
      "language_loss": 0.80499077,
      "learning_rate": 1.526257277114203e-06,
      "loss": 0.82698613,
      "num_input_tokens_seen": 105650260,
      "step": 4897,
      "time_per_iteration": 2.4983887672424316
    },
    {
      "auxiliary_loss_clip": 0.01141942,
      "auxiliary_loss_mlp": 0.01025302,
      "balance_loss_clip": 1.04969406,
      "balance_loss_mlp": 1.01808071,
      "epoch": 0.5889496783502676,
      "flos": 21981532383360.0,
      "grad_norm": 1.8278107779294945,
      "language_loss": 0.79342842,
      "learning_rate": 1.5255005101559201e-06,
      "loss": 0.81510085,
      "num_input_tokens_seen": 105667870,
      "step": 4898,
      "time_per_iteration": 3.324629545211792
    },
    {
      "auxiliary_loss_clip": 0.01157671,
      "auxiliary_loss_mlp": 0.01031649,
      "balance_loss_clip": 1.04963076,
      "balance_loss_mlp": 1.02480328,
      "epoch": 0.5890699212409066,
      "flos": 21685233093120.0,
      "grad_norm": 1.8676212334019406,
      "language_loss": 0.76633704,
      "learning_rate": 1.524743815178833e-06,
      "loss": 0.7882303,
      "num_input_tokens_seen": 105685830,
      "step": 4899,
      "time_per_iteration": 2.5018045902252197
    },
    {
      "auxiliary_loss_clip": 0.01143417,
      "auxiliary_loss_mlp": 0.01030543,
      "balance_loss_clip": 1.04561651,
      "balance_loss_mlp": 1.02351832,
      "epoch": 0.5891901641315457,
      "flos": 19464553635840.0,
      "grad_norm": 1.8049702374308343,
      "language_loss": 0.8092925,
      "learning_rate": 1.5239871922977315e-06,
      "loss": 0.8310321,
      "num_input_tokens_seen": 105705745,
      "step": 4900,
      "time_per_iteration": 2.5529839992523193
    },
    {
      "auxiliary_loss_clip": 0.01142887,
      "auxiliary_loss_mlp": 0.010274,
      "balance_loss_clip": 1.04764831,
      "balance_loss_mlp": 1.02028966,
      "epoch": 0.5893104070221848,
      "flos": 19609884063360.0,
      "grad_norm": 1.6703471734701048,
      "language_loss": 0.89424366,
      "learning_rate": 1.523230641627394e-06,
      "loss": 0.9159466,
      "num_input_tokens_seen": 105724730,
      "step": 4901,
      "time_per_iteration": 4.175079107284546
    },
    {
      "auxiliary_loss_clip": 0.0111598,
      "auxiliary_loss_mlp": 0.01028995,
      "balance_loss_clip": 1.04138231,
      "balance_loss_mlp": 1.02192271,
      "epoch": 0.5894306499128239,
      "flos": 29060063930880.0,
      "grad_norm": 1.8356308475899945,
      "language_loss": 0.72687101,
      "learning_rate": 1.5224741632825888e-06,
      "loss": 0.74832076,
      "num_input_tokens_seen": 105744920,
      "step": 4902,
      "time_per_iteration": 2.662339687347412
    },
    {
      "auxiliary_loss_clip": 0.01176102,
      "auxiliary_loss_mlp": 0.01028912,
      "balance_loss_clip": 1.0546,
      "balance_loss_mlp": 1.02137244,
      "epoch": 0.589550892803463,
      "flos": 42298890721920.0,
      "grad_norm": 1.7053620440638568,
      "language_loss": 0.69285369,
      "learning_rate": 1.521717757378074e-06,
      "loss": 0.71490377,
      "num_input_tokens_seen": 105765465,
      "step": 4903,
      "time_per_iteration": 2.6617751121520996
    },
    {
      "auxiliary_loss_clip": 0.01163631,
      "auxiliary_loss_mlp": 0.01030485,
      "balance_loss_clip": 1.0514946,
      "balance_loss_mlp": 1.0225457,
      "epoch": 0.5896711356941021,
      "flos": 14137062197760.0,
      "grad_norm": 1.9114130241308882,
      "language_loss": 0.69196153,
      "learning_rate": 1.5209614240285943e-06,
      "loss": 0.71390265,
      "num_input_tokens_seen": 105783120,
      "step": 4904,
      "time_per_iteration": 2.483524799346924
    },
    {
      "auxiliary_loss_clip": 0.01171919,
      "auxiliary_loss_mlp": 0.00761188,
      "balance_loss_clip": 1.05192935,
      "balance_loss_mlp": 1.00017202,
      "epoch": 0.5897913785847412,
      "flos": 17201355454080.0,
      "grad_norm": 1.9247187646404937,
      "language_loss": 0.84784108,
      "learning_rate": 1.520205163348887e-06,
      "loss": 0.86717212,
      "num_input_tokens_seen": 105801055,
      "step": 4905,
      "time_per_iteration": 2.478612184524536
    },
    {
      "auxiliary_loss_clip": 0.01036183,
      "auxiliary_loss_mlp": 0.01000649,
      "balance_loss_clip": 1.01423967,
      "balance_loss_mlp": 0.99936724,
      "epoch": 0.5899116214753802,
      "flos": 48794164202880.0,
      "grad_norm": 0.7258831705606584,
      "language_loss": 0.5692696,
      "learning_rate": 1.519448975453674e-06,
      "loss": 0.58963799,
      "num_input_tokens_seen": 105856155,
      "step": 4906,
      "time_per_iteration": 3.023139238357544
    },
    {
      "auxiliary_loss_clip": 0.01164859,
      "auxiliary_loss_mlp": 0.0076138,
      "balance_loss_clip": 1.05655003,
      "balance_loss_mlp": 1.00018668,
      "epoch": 0.5900318643660194,
      "flos": 21103659987840.0,
      "grad_norm": 2.1740961821739986,
      "language_loss": 0.75565284,
      "learning_rate": 1.5186928604576696e-06,
      "loss": 0.77491522,
      "num_input_tokens_seen": 105873350,
      "step": 4907,
      "time_per_iteration": 2.497197389602661
    },
    {
      "auxiliary_loss_clip": 0.01144302,
      "auxiliary_loss_mlp": 0.01031222,
      "balance_loss_clip": 1.0468775,
      "balance_loss_mlp": 1.02392614,
      "epoch": 0.5901521072566585,
      "flos": 21178390233600.0,
      "grad_norm": 2.0793663183361346,
      "language_loss": 0.77203268,
      "learning_rate": 1.5179368184755752e-06,
      "loss": 0.7937879,
      "num_input_tokens_seen": 105891435,
      "step": 4908,
      "time_per_iteration": 2.5226454734802246
    },
    {
      "auxiliary_loss_clip": 0.01142521,
      "auxiliary_loss_mlp": 0.01022716,
      "balance_loss_clip": 1.04854357,
      "balance_loss_mlp": 1.0159986,
      "epoch": 0.5902723501472975,
      "flos": 20225967160320.0,
      "grad_norm": 1.6045343990891583,
      "language_loss": 0.82569194,
      "learning_rate": 1.5171808496220821e-06,
      "loss": 0.84734428,
      "num_input_tokens_seen": 105910190,
      "step": 4909,
      "time_per_iteration": 2.543941020965576
    },
    {
      "auxiliary_loss_clip": 0.0114885,
      "auxiliary_loss_mlp": 0.01029688,
      "balance_loss_clip": 1.04810178,
      "balance_loss_mlp": 1.02239537,
      "epoch": 0.5903925930379367,
      "flos": 22964407211520.0,
      "grad_norm": 1.7016647151417654,
      "language_loss": 0.81568986,
      "learning_rate": 1.5164249540118708e-06,
      "loss": 0.83747518,
      "num_input_tokens_seen": 105929315,
      "step": 4910,
      "time_per_iteration": 2.560776948928833
    },
    {
      "auxiliary_loss_clip": 0.01106361,
      "auxiliary_loss_mlp": 0.0102794,
      "balance_loss_clip": 1.04239416,
      "balance_loss_mlp": 1.02046847,
      "epoch": 0.5905128359285757,
      "flos": 23367720096000.0,
      "grad_norm": 1.541881026419899,
      "language_loss": 0.83108616,
      "learning_rate": 1.5156691317596093e-06,
      "loss": 0.85242915,
      "num_input_tokens_seen": 105950740,
      "step": 4911,
      "time_per_iteration": 2.6541121006011963
    },
    {
      "auxiliary_loss_clip": 0.01157263,
      "auxiliary_loss_mlp": 0.00760177,
      "balance_loss_clip": 1.04877138,
      "balance_loss_mlp": 1.00019825,
      "epoch": 0.5906330788192148,
      "flos": 28032335994240.0,
      "grad_norm": 2.0532310463698433,
      "language_loss": 0.6626054,
      "learning_rate": 1.5149133829799556e-06,
      "loss": 0.6817798,
      "num_input_tokens_seen": 105968735,
      "step": 4912,
      "time_per_iteration": 2.57694149017334
    },
    {
      "auxiliary_loss_clip": 0.01148009,
      "auxiliary_loss_mlp": 0.0102452,
      "balance_loss_clip": 1.04926372,
      "balance_loss_mlp": 1.01772475,
      "epoch": 0.590753321709854,
      "flos": 18477943793280.0,
      "grad_norm": 2.137545618444139,
      "language_loss": 0.80904931,
      "learning_rate": 1.5141577077875556e-06,
      "loss": 0.83077461,
      "num_input_tokens_seen": 105986060,
      "step": 4913,
      "time_per_iteration": 2.5129330158233643
    },
    {
      "auxiliary_loss_clip": 0.01160855,
      "auxiliary_loss_mlp": 0.01025352,
      "balance_loss_clip": 1.05124164,
      "balance_loss_mlp": 1.01846802,
      "epoch": 0.590873564600493,
      "flos": 16873706568960.0,
      "grad_norm": 1.8266482896711302,
      "language_loss": 0.72268641,
      "learning_rate": 1.5134021062970451e-06,
      "loss": 0.7445485,
      "num_input_tokens_seen": 106004440,
      "step": 4914,
      "time_per_iteration": 2.5013575553894043
    },
    {
      "auxiliary_loss_clip": 0.01121393,
      "auxiliary_loss_mlp": 0.01033111,
      "balance_loss_clip": 1.04867816,
      "balance_loss_mlp": 1.02569056,
      "epoch": 0.5909938074911321,
      "flos": 13516166678400.0,
      "grad_norm": 1.918777874394368,
      "language_loss": 0.81061244,
      "learning_rate": 1.5126465786230483e-06,
      "loss": 0.83215749,
      "num_input_tokens_seen": 106021215,
      "step": 4915,
      "time_per_iteration": 2.5504186153411865
    },
    {
      "auxiliary_loss_clip": 0.01171692,
      "auxiliary_loss_mlp": 0.01028517,
      "balance_loss_clip": 1.05151987,
      "balance_loss_mlp": 1.02131629,
      "epoch": 0.5911140503817712,
      "flos": 26024067613440.0,
      "grad_norm": 1.6988681741621354,
      "language_loss": 0.82101023,
      "learning_rate": 1.5118911248801787e-06,
      "loss": 0.84301233,
      "num_input_tokens_seen": 106039225,
      "step": 4916,
      "time_per_iteration": 2.52847957611084
    },
    {
      "auxiliary_loss_clip": 0.01153985,
      "auxiliary_loss_mlp": 0.01030794,
      "balance_loss_clip": 1.04909873,
      "balance_loss_mlp": 1.02406716,
      "epoch": 0.5912342932724103,
      "flos": 23258731253760.0,
      "grad_norm": 2.8385331723413287,
      "language_loss": 0.79594541,
      "learning_rate": 1.5111357451830364e-06,
      "loss": 0.81779313,
      "num_input_tokens_seen": 106057920,
      "step": 4917,
      "time_per_iteration": 3.2237794399261475
    },
    {
      "auxiliary_loss_clip": 0.01156829,
      "auxiliary_loss_mlp": 0.0102918,
      "balance_loss_clip": 1.04910982,
      "balance_loss_mlp": 1.02286792,
      "epoch": 0.5913545361630493,
      "flos": 19573039687680.0,
      "grad_norm": 1.8364539734887542,
      "language_loss": 0.71392214,
      "learning_rate": 1.5103804396462131e-06,
      "loss": 0.73578227,
      "num_input_tokens_seen": 106077855,
      "step": 4918,
      "time_per_iteration": 2.522834062576294
    },
    {
      "auxiliary_loss_clip": 0.01158961,
      "auxiliary_loss_mlp": 0.01029614,
      "balance_loss_clip": 1.04752111,
      "balance_loss_mlp": 1.02168667,
      "epoch": 0.5914747790536885,
      "flos": 26213532877440.0,
      "grad_norm": 2.2966766142768775,
      "language_loss": 0.79817551,
      "learning_rate": 1.5096252083842877e-06,
      "loss": 0.82006127,
      "num_input_tokens_seen": 106097065,
      "step": 4919,
      "time_per_iteration": 2.529672384262085
    },
    {
      "auxiliary_loss_clip": 0.01154649,
      "auxiliary_loss_mlp": 0.01027937,
      "balance_loss_clip": 1.0469259,
      "balance_loss_mlp": 1.02026558,
      "epoch": 0.5915950219443276,
      "flos": 27417545786880.0,
      "grad_norm": 1.7225577234607803,
      "language_loss": 0.85396218,
      "learning_rate": 1.5088700515118285e-06,
      "loss": 0.87578803,
      "num_input_tokens_seen": 106116385,
      "step": 4920,
      "time_per_iteration": 2.5839152336120605
    },
    {
      "auxiliary_loss_clip": 0.01126847,
      "auxiliary_loss_mlp": 0.01033015,
      "balance_loss_clip": 1.04669333,
      "balance_loss_mlp": 1.02526963,
      "epoch": 0.5917152648349666,
      "flos": 21907879545600.0,
      "grad_norm": 1.650192691341449,
      "language_loss": 0.66642636,
      "learning_rate": 1.508114969143392e-06,
      "loss": 0.688025,
      "num_input_tokens_seen": 106136370,
      "step": 4921,
      "time_per_iteration": 2.5942749977111816
    },
    {
      "auxiliary_loss_clip": 0.01140232,
      "auxiliary_loss_mlp": 0.01025666,
      "balance_loss_clip": 1.04396558,
      "balance_loss_mlp": 1.01908588,
      "epoch": 0.5918355077256057,
      "flos": 28109185142400.0,
      "grad_norm": 1.5231740737225965,
      "language_loss": 0.77531517,
      "learning_rate": 1.5073599613935238e-06,
      "loss": 0.79697418,
      "num_input_tokens_seen": 106158490,
      "step": 4922,
      "time_per_iteration": 2.6173675060272217
    },
    {
      "auxiliary_loss_clip": 0.01141613,
      "auxiliary_loss_mlp": 0.01030431,
      "balance_loss_clip": 1.04745555,
      "balance_loss_mlp": 1.02308774,
      "epoch": 0.5919557506162448,
      "flos": 28183807647360.0,
      "grad_norm": 1.871509948582554,
      "language_loss": 0.57680422,
      "learning_rate": 1.5066050283767574e-06,
      "loss": 0.59852463,
      "num_input_tokens_seen": 106179170,
      "step": 4923,
      "time_per_iteration": 2.5840277671813965
    },
    {
      "auxiliary_loss_clip": 0.0113438,
      "auxiliary_loss_mlp": 0.0102445,
      "balance_loss_clip": 1.04604959,
      "balance_loss_mlp": 1.01751459,
      "epoch": 0.5920759935068839,
      "flos": 12094355652480.0,
      "grad_norm": 1.8455383663932154,
      "language_loss": 0.82635844,
      "learning_rate": 1.505850170207616e-06,
      "loss": 0.84794682,
      "num_input_tokens_seen": 106196035,
      "step": 4924,
      "time_per_iteration": 3.261245012283325
    },
    {
      "auxiliary_loss_clip": 0.01142211,
      "auxiliary_loss_mlp": 0.01025952,
      "balance_loss_clip": 1.04627705,
      "balance_loss_mlp": 1.0190289,
      "epoch": 0.592196236397523,
      "flos": 29424772673280.0,
      "grad_norm": 1.9221368445915812,
      "language_loss": 0.78357834,
      "learning_rate": 1.505095387000611e-06,
      "loss": 0.80525994,
      "num_input_tokens_seen": 106218335,
      "step": 4925,
      "time_per_iteration": 2.61132550239563
    },
    {
      "auxiliary_loss_clip": 0.01134717,
      "auxiliary_loss_mlp": 0.01024046,
      "balance_loss_clip": 1.04788673,
      "balance_loss_mlp": 1.01723325,
      "epoch": 0.5923164792881621,
      "flos": 24384709866240.0,
      "grad_norm": 2.072709458397679,
      "language_loss": 0.74224973,
      "learning_rate": 1.504340678870242e-06,
      "loss": 0.7638374,
      "num_input_tokens_seen": 106236550,
      "step": 4926,
      "time_per_iteration": 2.555328369140625
    },
    {
      "auxiliary_loss_clip": 0.01154651,
      "auxiliary_loss_mlp": 0.01026527,
      "balance_loss_clip": 1.048069,
      "balance_loss_mlp": 1.01861119,
      "epoch": 0.5924367221788012,
      "flos": 24024238928640.0,
      "grad_norm": 1.9015707812745677,
      "language_loss": 0.89757496,
      "learning_rate": 1.5035860459309989e-06,
      "loss": 0.91938674,
      "num_input_tokens_seen": 106254265,
      "step": 4927,
      "time_per_iteration": 4.063888788223267
    },
    {
      "auxiliary_loss_clip": 0.01139904,
      "auxiliary_loss_mlp": 0.01031259,
      "balance_loss_clip": 1.04792452,
      "balance_loss_mlp": 1.02351642,
      "epoch": 0.5925569650694402,
      "flos": 26870590414080.0,
      "grad_norm": 2.0267810449516404,
      "language_loss": 0.6329267,
      "learning_rate": 1.5028314882973568e-06,
      "loss": 0.65463841,
      "num_input_tokens_seen": 106274670,
      "step": 4928,
      "time_per_iteration": 2.571357250213623
    },
    {
      "auxiliary_loss_clip": 0.01143082,
      "auxiliary_loss_mlp": 0.01026323,
      "balance_loss_clip": 1.04780102,
      "balance_loss_mlp": 1.01880956,
      "epoch": 0.5926772079600794,
      "flos": 22302788647680.0,
      "grad_norm": 1.867633896170128,
      "language_loss": 0.84454769,
      "learning_rate": 1.502077006083783e-06,
      "loss": 0.86624175,
      "num_input_tokens_seen": 106293330,
      "step": 4929,
      "time_per_iteration": 2.5418643951416016
    },
    {
      "auxiliary_loss_clip": 0.01158414,
      "auxiliary_loss_mlp": 0.00760299,
      "balance_loss_clip": 1.05166495,
      "balance_loss_mlp": 1.00017071,
      "epoch": 0.5927974508507184,
      "flos": 19865244827520.0,
      "grad_norm": 1.7674895595216809,
      "language_loss": 0.767335,
      "learning_rate": 1.5013225994047315e-06,
      "loss": 0.78652209,
      "num_input_tokens_seen": 106310960,
      "step": 4930,
      "time_per_iteration": 2.5073676109313965
    },
    {
      "auxiliary_loss_clip": 0.01158604,
      "auxiliary_loss_mlp": 0.00760461,
      "balance_loss_clip": 1.05113912,
      "balance_loss_mlp": 1.00013125,
      "epoch": 0.5929176937413575,
      "flos": 15776743167360.0,
      "grad_norm": 1.8447548684730877,
      "language_loss": 0.81064022,
      "learning_rate": 1.5005682683746452e-06,
      "loss": 0.82983088,
      "num_input_tokens_seen": 106329475,
      "step": 4931,
      "time_per_iteration": 2.541358470916748
    },
    {
      "auxiliary_loss_clip": 0.01157569,
      "auxiliary_loss_mlp": 0.01028442,
      "balance_loss_clip": 1.05161786,
      "balance_loss_mlp": 1.02112269,
      "epoch": 0.5930379366319967,
      "flos": 17601472028160.0,
      "grad_norm": 2.2507506290493606,
      "language_loss": 0.72738862,
      "learning_rate": 1.4998140131079553e-06,
      "loss": 0.74924874,
      "num_input_tokens_seen": 106345565,
      "step": 4932,
      "time_per_iteration": 2.460134744644165
    },
    {
      "auxiliary_loss_clip": 0.01097771,
      "auxiliary_loss_mlp": 0.00760554,
      "balance_loss_clip": 1.04351926,
      "balance_loss_mlp": 1.00014257,
      "epoch": 0.5931581795226357,
      "flos": 17704283731200.0,
      "grad_norm": 1.7621857394681002,
      "language_loss": 0.73521745,
      "learning_rate": 1.4990598337190821e-06,
      "loss": 0.75380069,
      "num_input_tokens_seen": 106361920,
      "step": 4933,
      "time_per_iteration": 2.6184725761413574
    },
    {
      "auxiliary_loss_clip": 0.01170578,
      "auxiliary_loss_mlp": 0.00760694,
      "balance_loss_clip": 1.05137968,
      "balance_loss_mlp": 1.00011766,
      "epoch": 0.5932784224132748,
      "flos": 24280102483200.0,
      "grad_norm": 1.6769025440705971,
      "language_loss": 0.6779772,
      "learning_rate": 1.4983057303224338e-06,
      "loss": 0.69728994,
      "num_input_tokens_seen": 106381735,
      "step": 4934,
      "time_per_iteration": 2.5065150260925293
    },
    {
      "auxiliary_loss_clip": 0.01114978,
      "auxiliary_loss_mlp": 0.01027065,
      "balance_loss_clip": 1.04522574,
      "balance_loss_mlp": 1.02006137,
      "epoch": 0.5933986653039139,
      "flos": 22926700909440.0,
      "grad_norm": 1.602730917779371,
      "language_loss": 0.87558407,
      "learning_rate": 1.4975517030324072e-06,
      "loss": 0.89700449,
      "num_input_tokens_seen": 106399745,
      "step": 4935,
      "time_per_iteration": 2.6094958782196045
    },
    {
      "auxiliary_loss_clip": 0.01063837,
      "auxiliary_loss_mlp": 0.00751368,
      "balance_loss_clip": 1.01494527,
      "balance_loss_mlp": 1.00017822,
      "epoch": 0.593518908194553,
      "flos": 71121730256640.0,
      "grad_norm": 0.7814877092464576,
      "language_loss": 0.61851883,
      "learning_rate": 1.4967977519633882e-06,
      "loss": 0.63667083,
      "num_input_tokens_seen": 106457205,
      "step": 4936,
      "time_per_iteration": 3.1946847438812256
    },
    {
      "auxiliary_loss_clip": 0.01128936,
      "auxiliary_loss_mlp": 0.01032334,
      "balance_loss_clip": 1.04664278,
      "balance_loss_mlp": 1.02470791,
      "epoch": 0.593639151085192,
      "flos": 20448649526400.0,
      "grad_norm": 1.814706649215006,
      "language_loss": 0.78118503,
      "learning_rate": 1.4960438772297494e-06,
      "loss": 0.80279779,
      "num_input_tokens_seen": 106474250,
      "step": 4937,
      "time_per_iteration": 2.5947492122650146
    },
    {
      "auxiliary_loss_clip": 0.0114614,
      "auxiliary_loss_mlp": 0.01028455,
      "balance_loss_clip": 1.04608941,
      "balance_loss_mlp": 1.02064085,
      "epoch": 0.5937593939758312,
      "flos": 30883428074880.0,
      "grad_norm": 2.1796965352311806,
      "language_loss": 0.73670471,
      "learning_rate": 1.495290078945855e-06,
      "loss": 0.75845063,
      "num_input_tokens_seen": 106494015,
      "step": 4938,
      "time_per_iteration": 2.66733455657959
    },
    {
      "auxiliary_loss_clip": 0.0116917,
      "auxiliary_loss_mlp": 0.01026495,
      "balance_loss_clip": 1.05083203,
      "balance_loss_mlp": 1.01936936,
      "epoch": 0.5938796368664703,
      "flos": 36898069668480.0,
      "grad_norm": 1.7260550078194208,
      "language_loss": 0.74305558,
      "learning_rate": 1.4945363572260529e-06,
      "loss": 0.76501215,
      "num_input_tokens_seen": 106515010,
      "step": 4939,
      "time_per_iteration": 2.665654420852661
    },
    {
      "auxiliary_loss_clip": 0.01153699,
      "auxiliary_loss_mlp": 0.01021097,
      "balance_loss_clip": 1.04753351,
      "balance_loss_mlp": 1.01372373,
      "epoch": 0.5939998797571093,
      "flos": 23842926051840.0,
      "grad_norm": 2.0561318723523363,
      "language_loss": 0.68349516,
      "learning_rate": 1.4937827121846845e-06,
      "loss": 0.70524311,
      "num_input_tokens_seen": 106535265,
      "step": 4940,
      "time_per_iteration": 2.5404906272888184
    },
    {
      "auxiliary_loss_clip": 0.01120281,
      "auxiliary_loss_mlp": 0.01032336,
      "balance_loss_clip": 1.04523039,
      "balance_loss_mlp": 1.02506471,
      "epoch": 0.5941201226477485,
      "flos": 25191407462400.0,
      "grad_norm": 1.4974181291318573,
      "language_loss": 0.7343905,
      "learning_rate": 1.4930291439360755e-06,
      "loss": 0.75591671,
      "num_input_tokens_seen": 106557830,
      "step": 4941,
      "time_per_iteration": 2.6068835258483887
    },
    {
      "auxiliary_loss_clip": 0.01157658,
      "auxiliary_loss_mlp": 0.010319,
      "balance_loss_clip": 1.05024934,
      "balance_loss_mlp": 1.02437174,
      "epoch": 0.5942403655383875,
      "flos": 22418996123520.0,
      "grad_norm": 1.8020657355664922,
      "language_loss": 0.79431498,
      "learning_rate": 1.4922756525945427e-06,
      "loss": 0.81621057,
      "num_input_tokens_seen": 106577140,
      "step": 4942,
      "time_per_iteration": 2.5212016105651855
    },
    {
      "auxiliary_loss_clip": 0.0105153,
      "auxiliary_loss_mlp": 0.01011245,
      "balance_loss_clip": 1.01549101,
      "balance_loss_mlp": 1.01012468,
      "epoch": 0.5943606084290266,
      "flos": 67629310796160.0,
      "grad_norm": 0.9416388444901403,
      "language_loss": 0.59596115,
      "learning_rate": 1.4915222382743894e-06,
      "loss": 0.61658889,
      "num_input_tokens_seen": 106635975,
      "step": 4943,
      "time_per_iteration": 3.88228702545166
    },
    {
      "auxiliary_loss_clip": 0.01157404,
      "auxiliary_loss_mlp": 0.01036981,
      "balance_loss_clip": 1.05152106,
      "balance_loss_mlp": 1.0292089,
      "epoch": 0.5944808513196658,
      "flos": 18223157646720.0,
      "grad_norm": 2.5368862216964527,
      "language_loss": 0.71964741,
      "learning_rate": 1.4907689010899085e-06,
      "loss": 0.74159127,
      "num_input_tokens_seen": 106653555,
      "step": 4944,
      "time_per_iteration": 2.477567434310913
    },
    {
      "auxiliary_loss_clip": 0.01141904,
      "auxiliary_loss_mlp": 0.01024615,
      "balance_loss_clip": 1.04816771,
      "balance_loss_mlp": 1.01770401,
      "epoch": 0.5946010942103048,
      "flos": 24790824011520.0,
      "grad_norm": 1.838468174874239,
      "language_loss": 0.62560189,
      "learning_rate": 1.4900156411553804e-06,
      "loss": 0.64726704,
      "num_input_tokens_seen": 106673385,
      "step": 4945,
      "time_per_iteration": 2.577545166015625
    },
    {
      "auxiliary_loss_clip": 0.01146134,
      "auxiliary_loss_mlp": 0.0103382,
      "balance_loss_clip": 1.04975331,
      "balance_loss_mlp": 1.02671194,
      "epoch": 0.5947213371009439,
      "flos": 15231619388160.0,
      "grad_norm": 2.042271934224919,
      "language_loss": 0.85566735,
      "learning_rate": 1.4892624585850739e-06,
      "loss": 0.87746692,
      "num_input_tokens_seen": 106691740,
      "step": 4946,
      "time_per_iteration": 2.5260186195373535
    },
    {
      "auxiliary_loss_clip": 0.01173348,
      "auxiliary_loss_mlp": 0.010263,
      "balance_loss_clip": 1.05173409,
      "balance_loss_mlp": 1.01920664,
      "epoch": 0.594841579991583,
      "flos": 25848069949440.0,
      "grad_norm": 2.160239026251684,
      "language_loss": 0.79617047,
      "learning_rate": 1.4885093534932465e-06,
      "loss": 0.81816697,
      "num_input_tokens_seen": 106709705,
      "step": 4947,
      "time_per_iteration": 2.499295711517334
    },
    {
      "auxiliary_loss_clip": 0.01141825,
      "auxiliary_loss_mlp": 0.01032633,
      "balance_loss_clip": 1.05122542,
      "balance_loss_mlp": 1.02522981,
      "epoch": 0.5949618228822221,
      "flos": 23981109672960.0,
      "grad_norm": 1.9371950946161733,
      "language_loss": 0.71124041,
      "learning_rate": 1.4877563259941433e-06,
      "loss": 0.73298502,
      "num_input_tokens_seen": 106727560,
      "step": 4948,
      "time_per_iteration": 2.559952735900879
    },
    {
      "auxiliary_loss_clip": 0.01164304,
      "auxiliary_loss_mlp": 0.01026008,
      "balance_loss_clip": 1.05129421,
      "balance_loss_mlp": 1.01818752,
      "epoch": 0.5950820657728612,
      "flos": 40547491476480.0,
      "grad_norm": 2.198893873823859,
      "language_loss": 0.67517966,
      "learning_rate": 1.4870033762019988e-06,
      "loss": 0.69708276,
      "num_input_tokens_seen": 106747725,
      "step": 4949,
      "time_per_iteration": 3.4141390323638916
    },
    {
      "auxiliary_loss_clip": 0.01141284,
      "auxiliary_loss_mlp": 0.01029389,
      "balance_loss_clip": 1.04707515,
      "balance_loss_mlp": 1.02142,
      "epoch": 0.5952023086635003,
      "flos": 23184467884800.0,
      "grad_norm": 1.583804077719472,
      "language_loss": 0.7349515,
      "learning_rate": 1.4862505042310334e-06,
      "loss": 0.75665814,
      "num_input_tokens_seen": 106767010,
      "step": 4950,
      "time_per_iteration": 2.5252623558044434
    },
    {
      "auxiliary_loss_clip": 0.01137435,
      "auxiliary_loss_mlp": 0.01030251,
      "balance_loss_clip": 1.04847646,
      "balance_loss_mlp": 1.02300572,
      "epoch": 0.5953225515541394,
      "flos": 33653289548160.0,
      "grad_norm": 1.6381361841671558,
      "language_loss": 0.69407034,
      "learning_rate": 1.4854977101954587e-06,
      "loss": 0.71574718,
      "num_input_tokens_seen": 106789230,
      "step": 4951,
      "time_per_iteration": 2.635918140411377
    },
    {
      "auxiliary_loss_clip": 0.01157373,
      "auxiliary_loss_mlp": 0.01027982,
      "balance_loss_clip": 1.04728937,
      "balance_loss_mlp": 1.02013826,
      "epoch": 0.5954427944447784,
      "flos": 24459619680000.0,
      "grad_norm": 1.8205106592082994,
      "language_loss": 0.85983253,
      "learning_rate": 1.4847449942094716e-06,
      "loss": 0.88168603,
      "num_input_tokens_seen": 106808110,
      "step": 4952,
      "time_per_iteration": 2.5185632705688477
    },
    {
      "auxiliary_loss_clip": 0.0114235,
      "auxiliary_loss_mlp": 0.01026679,
      "balance_loss_clip": 1.04882193,
      "balance_loss_mlp": 1.01894832,
      "epoch": 0.5955630373354175,
      "flos": 18551848026240.0,
      "grad_norm": 3.048823204584803,
      "language_loss": 0.86300039,
      "learning_rate": 1.4839923563872598e-06,
      "loss": 0.88469064,
      "num_input_tokens_seen": 106826650,
      "step": 4953,
      "time_per_iteration": 4.145396947860718
    },
    {
      "auxiliary_loss_clip": 0.01126167,
      "auxiliary_loss_mlp": 0.01026338,
      "balance_loss_clip": 1.04654002,
      "balance_loss_mlp": 1.01912594,
      "epoch": 0.5956832802260567,
      "flos": 19791699730560.0,
      "grad_norm": 1.7019358353226637,
      "language_loss": 0.75969315,
      "learning_rate": 1.483239796842997e-06,
      "loss": 0.78121823,
      "num_input_tokens_seen": 106844680,
      "step": 4954,
      "time_per_iteration": 2.559460401535034
    },
    {
      "auxiliary_loss_clip": 0.01127558,
      "auxiliary_loss_mlp": 0.01025659,
      "balance_loss_clip": 1.04823875,
      "balance_loss_mlp": 1.01881075,
      "epoch": 0.5958035231166957,
      "flos": 19750868945280.0,
      "grad_norm": 1.6026827236600416,
      "language_loss": 0.83781356,
      "learning_rate": 1.4824873156908462e-06,
      "loss": 0.85934579,
      "num_input_tokens_seen": 106862605,
      "step": 4955,
      "time_per_iteration": 2.557255268096924
    },
    {
      "auxiliary_loss_clip": 0.01159569,
      "auxiliary_loss_mlp": 0.00761244,
      "balance_loss_clip": 1.05209899,
      "balance_loss_mlp": 1.00013292,
      "epoch": 0.5959237660073348,
      "flos": 21652806090240.0,
      "grad_norm": 1.4970469937253992,
      "language_loss": 0.75319445,
      "learning_rate": 1.4817349130449584e-06,
      "loss": 0.77240258,
      "num_input_tokens_seen": 106882325,
      "step": 4956,
      "time_per_iteration": 2.527223587036133
    },
    {
      "auxiliary_loss_clip": 0.01156108,
      "auxiliary_loss_mlp": 0.01029058,
      "balance_loss_clip": 1.05039203,
      "balance_loss_mlp": 1.02220654,
      "epoch": 0.5960440088979739,
      "flos": 21171207513600.0,
      "grad_norm": 3.9239504211231364,
      "language_loss": 0.82929188,
      "learning_rate": 1.4809825890194717e-06,
      "loss": 0.8511436,
      "num_input_tokens_seen": 106900995,
      "step": 4957,
      "time_per_iteration": 2.5119004249572754
    },
    {
      "auxiliary_loss_clip": 0.01136639,
      "auxiliary_loss_mlp": 0.01028765,
      "balance_loss_clip": 1.04599643,
      "balance_loss_mlp": 1.02211905,
      "epoch": 0.596164251788613,
      "flos": 14757526753920.0,
      "grad_norm": 1.7245731850533845,
      "language_loss": 0.77006006,
      "learning_rate": 1.4802303437285139e-06,
      "loss": 0.79171413,
      "num_input_tokens_seen": 106918265,
      "step": 4958,
      "time_per_iteration": 2.5098612308502197
    },
    {
      "auxiliary_loss_clip": 0.01143317,
      "auxiliary_loss_mlp": 0.01029356,
      "balance_loss_clip": 1.04729402,
      "balance_loss_mlp": 1.02117825,
      "epoch": 0.596284494679252,
      "flos": 20485924865280.0,
      "grad_norm": 2.3946136020528463,
      "language_loss": 0.8006283,
      "learning_rate": 1.4794781772861994e-06,
      "loss": 0.82235503,
      "num_input_tokens_seen": 106934760,
      "step": 4959,
      "time_per_iteration": 2.5287961959838867
    },
    {
      "auxiliary_loss_clip": 0.01142608,
      "auxiliary_loss_mlp": 0.00760745,
      "balance_loss_clip": 1.04820704,
      "balance_loss_mlp": 1.00014782,
      "epoch": 0.5964047375698912,
      "flos": 31212262108800.0,
      "grad_norm": 1.890362889090565,
      "language_loss": 0.66995776,
      "learning_rate": 1.4787260898066324e-06,
      "loss": 0.68899131,
      "num_input_tokens_seen": 106954760,
      "step": 4960,
      "time_per_iteration": 2.6273903846740723
    },
    {
      "auxiliary_loss_clip": 0.01167735,
      "auxiliary_loss_mlp": 0.01027823,
      "balance_loss_clip": 1.05070078,
      "balance_loss_mlp": 1.02033341,
      "epoch": 0.5965249804605303,
      "flos": 27483620855040.0,
      "grad_norm": 1.9634626220985516,
      "language_loss": 0.86040682,
      "learning_rate": 1.4779740814039023e-06,
      "loss": 0.88236243,
      "num_input_tokens_seen": 106974845,
      "step": 4961,
      "time_per_iteration": 2.498137950897217
    },
    {
      "auxiliary_loss_clip": 0.01172877,
      "auxiliary_loss_mlp": 0.0102209,
      "balance_loss_clip": 1.05240273,
      "balance_loss_mlp": 1.01456165,
      "epoch": 0.5966452233511693,
      "flos": 30773936442240.0,
      "grad_norm": 1.8253998573973973,
      "language_loss": 0.68060118,
      "learning_rate": 1.4772221521920894e-06,
      "loss": 0.70255083,
      "num_input_tokens_seen": 106994870,
      "step": 4962,
      "time_per_iteration": 2.5359127521514893
    },
    {
      "auxiliary_loss_clip": 0.01142013,
      "auxiliary_loss_mlp": 0.01029354,
      "balance_loss_clip": 1.04916692,
      "balance_loss_mlp": 1.02213025,
      "epoch": 0.5967654662418085,
      "flos": 25481170477440.0,
      "grad_norm": 2.4609260596742106,
      "language_loss": 0.7405647,
      "learning_rate": 1.4764703022852598e-06,
      "loss": 0.76227832,
      "num_input_tokens_seen": 107015390,
      "step": 4963,
      "time_per_iteration": 2.5565099716186523
    },
    {
      "auxiliary_loss_clip": 0.01090469,
      "auxiliary_loss_mlp": 0.01026487,
      "balance_loss_clip": 1.04414678,
      "balance_loss_mlp": 1.01960576,
      "epoch": 0.5968857091324475,
      "flos": 19099126621440.0,
      "grad_norm": 1.7404933431817045,
      "language_loss": 0.76792175,
      "learning_rate": 1.4757185317974696e-06,
      "loss": 0.78909129,
      "num_input_tokens_seen": 107033775,
      "step": 4964,
      "time_per_iteration": 2.603670835494995
    },
    {
      "auxiliary_loss_clip": 0.01155156,
      "auxiliary_loss_mlp": 0.01026795,
      "balance_loss_clip": 1.04808283,
      "balance_loss_mlp": 1.01965427,
      "epoch": 0.5970059520230866,
      "flos": 23692711374720.0,
      "grad_norm": 2.0228160139703224,
      "language_loss": 0.70779198,
      "learning_rate": 1.474966840842761e-06,
      "loss": 0.72961152,
      "num_input_tokens_seen": 107053355,
      "step": 4965,
      "time_per_iteration": 2.5000007152557373
    },
    {
      "auxiliary_loss_clip": 0.01159763,
      "auxiliary_loss_mlp": 0.01032595,
      "balance_loss_clip": 1.05043602,
      "balance_loss_mlp": 1.02515078,
      "epoch": 0.5971261949137258,
      "flos": 23185545292800.0,
      "grad_norm": 2.2901797523001735,
      "language_loss": 0.86947185,
      "learning_rate": 1.4742152295351655e-06,
      "loss": 0.89139545,
      "num_input_tokens_seen": 107072510,
      "step": 4966,
      "time_per_iteration": 2.5368471145629883
    },
    {
      "auxiliary_loss_clip": 0.01157867,
      "auxiliary_loss_mlp": 0.00761567,
      "balance_loss_clip": 1.05103004,
      "balance_loss_mlp": 1.00017929,
      "epoch": 0.5972464378043648,
      "flos": 20557710195840.0,
      "grad_norm": 3.2015555167182295,
      "language_loss": 0.64401215,
      "learning_rate": 1.4734636979887016e-06,
      "loss": 0.66320658,
      "num_input_tokens_seen": 107089970,
      "step": 4967,
      "time_per_iteration": 2.498842239379883
    },
    {
      "auxiliary_loss_clip": 0.01133442,
      "auxiliary_loss_mlp": 0.01032199,
      "balance_loss_clip": 1.04586828,
      "balance_loss_mlp": 1.02433431,
      "epoch": 0.5973666806950039,
      "flos": 29387030457600.0,
      "grad_norm": 2.058990998833509,
      "language_loss": 0.90192312,
      "learning_rate": 1.4727122463173755e-06,
      "loss": 0.92357957,
      "num_input_tokens_seen": 107108500,
      "step": 4968,
      "time_per_iteration": 2.5988411903381348
    },
    {
      "auxiliary_loss_clip": 0.01144814,
      "auxiliary_loss_mlp": 0.01026865,
      "balance_loss_clip": 1.05053282,
      "balance_loss_mlp": 1.0193069,
      "epoch": 0.597486923585643,
      "flos": 22273522041600.0,
      "grad_norm": 1.8513019254147054,
      "language_loss": 0.64295685,
      "learning_rate": 1.471960874635183e-06,
      "loss": 0.66467363,
      "num_input_tokens_seen": 107128060,
      "step": 4969,
      "time_per_iteration": 3.3159897327423096
    },
    {
      "auxiliary_loss_clip": 0.01139291,
      "auxiliary_loss_mlp": 0.01027003,
      "balance_loss_clip": 1.04659724,
      "balance_loss_mlp": 1.01949906,
      "epoch": 0.5976071664762821,
      "flos": 13772461196160.0,
      "grad_norm": 2.02812293871986,
      "language_loss": 0.70650601,
      "learning_rate": 1.4712095830561055e-06,
      "loss": 0.72816896,
      "num_input_tokens_seen": 107146550,
      "step": 4970,
      "time_per_iteration": 2.5073142051696777
    },
    {
      "auxiliary_loss_clip": 0.0114389,
      "auxiliary_loss_mlp": 0.01029279,
      "balance_loss_clip": 1.04644704,
      "balance_loss_mlp": 1.022138,
      "epoch": 0.5977274093669211,
      "flos": 19098623831040.0,
      "grad_norm": 2.5243872349622967,
      "language_loss": 0.81205362,
      "learning_rate": 1.4704583716941147e-06,
      "loss": 0.8337853,
      "num_input_tokens_seen": 107165415,
      "step": 4971,
      "time_per_iteration": 2.4951910972595215
    },
    {
      "auxiliary_loss_clip": 0.01150524,
      "auxiliary_loss_mlp": 0.0103683,
      "balance_loss_clip": 1.05081582,
      "balance_loss_mlp": 1.02932584,
      "epoch": 0.5978476522575603,
      "flos": 20376002269440.0,
      "grad_norm": 1.8454770498227142,
      "language_loss": 0.72436142,
      "learning_rate": 1.4697072406631672e-06,
      "loss": 0.74623495,
      "num_input_tokens_seen": 107185320,
      "step": 4972,
      "time_per_iteration": 2.505258798599243
    },
    {
      "auxiliary_loss_clip": 0.01120945,
      "auxiliary_loss_mlp": 0.01032547,
      "balance_loss_clip": 1.04874778,
      "balance_loss_mlp": 1.02506077,
      "epoch": 0.5979678951481994,
      "flos": 29023147728000.0,
      "grad_norm": 1.6355659600866272,
      "language_loss": 0.72527552,
      "learning_rate": 1.4689561900772097e-06,
      "loss": 0.74681044,
      "num_input_tokens_seen": 107205380,
      "step": 4973,
      "time_per_iteration": 2.650524616241455
    },
    {
      "auxiliary_loss_clip": 0.01143368,
      "auxiliary_loss_mlp": 0.01029942,
      "balance_loss_clip": 1.0474031,
      "balance_loss_mlp": 1.0225687,
      "epoch": 0.5980881380388384,
      "flos": 17967689141760.0,
      "grad_norm": 2.1245241270444715,
      "language_loss": 0.72083718,
      "learning_rate": 1.4682052200501758e-06,
      "loss": 0.74257028,
      "num_input_tokens_seen": 107222585,
      "step": 4974,
      "time_per_iteration": 2.5049898624420166
    },
    {
      "auxiliary_loss_clip": 0.0116929,
      "auxiliary_loss_mlp": 0.01028519,
      "balance_loss_clip": 1.05021095,
      "balance_loss_mlp": 1.02127731,
      "epoch": 0.5982083809294776,
      "flos": 22962827013120.0,
      "grad_norm": 1.7102658490633584,
      "language_loss": 0.79901171,
      "learning_rate": 1.4674543306959876e-06,
      "loss": 0.82098985,
      "num_input_tokens_seen": 107242055,
      "step": 4975,
      "time_per_iteration": 3.2376315593719482
    },
    {
      "auxiliary_loss_clip": 0.01148235,
      "auxiliary_loss_mlp": 0.01029737,
      "balance_loss_clip": 1.04861844,
      "balance_loss_mlp": 1.02203608,
      "epoch": 0.5983286238201166,
      "flos": 20991941712000.0,
      "grad_norm": 2.3119467174979382,
      "language_loss": 0.84080279,
      "learning_rate": 1.4667035221285535e-06,
      "loss": 0.86258256,
      "num_input_tokens_seen": 107259695,
      "step": 4976,
      "time_per_iteration": 2.516244888305664
    },
    {
      "auxiliary_loss_clip": 0.01153378,
      "auxiliary_loss_mlp": 0.01024775,
      "balance_loss_clip": 1.05044377,
      "balance_loss_mlp": 1.0174613,
      "epoch": 0.5984488667107557,
      "flos": 28183448511360.0,
      "grad_norm": 1.9461037709054763,
      "language_loss": 0.74233413,
      "learning_rate": 1.4659527944617715e-06,
      "loss": 0.76411569,
      "num_input_tokens_seen": 107279640,
      "step": 4977,
      "time_per_iteration": 2.5352916717529297
    },
    {
      "auxiliary_loss_clip": 0.01095291,
      "auxiliary_loss_mlp": 0.01030104,
      "balance_loss_clip": 1.04244733,
      "balance_loss_mlp": 1.02262056,
      "epoch": 0.5985691096013949,
      "flos": 16471794314880.0,
      "grad_norm": 1.6866192473436747,
      "language_loss": 0.76199692,
      "learning_rate": 1.465202147809526e-06,
      "loss": 0.78325087,
      "num_input_tokens_seen": 107298135,
      "step": 4978,
      "time_per_iteration": 3.4044220447540283
    },
    {
      "auxiliary_loss_clip": 0.0117308,
      "auxiliary_loss_mlp": 0.01027398,
      "balance_loss_clip": 1.05270672,
      "balance_loss_mlp": 1.02005494,
      "epoch": 0.5986893524920339,
      "flos": 26719046933760.0,
      "grad_norm": 1.919778392226231,
      "language_loss": 0.76026136,
      "learning_rate": 1.4644515822856888e-06,
      "loss": 0.78226614,
      "num_input_tokens_seen": 107316570,
      "step": 4979,
      "time_per_iteration": 3.2520570755004883
    },
    {
      "auxiliary_loss_clip": 0.01032236,
      "auxiliary_loss_mlp": 0.01003381,
      "balance_loss_clip": 1.01329887,
      "balance_loss_mlp": 1.0023793,
      "epoch": 0.598809595382673,
      "flos": 61608061100160.0,
      "grad_norm": 0.7592408506814904,
      "language_loss": 0.56544816,
      "learning_rate": 1.4637010980041215e-06,
      "loss": 0.58580428,
      "num_input_tokens_seen": 107378680,
      "step": 4980,
      "time_per_iteration": 3.1571218967437744
    },
    {
      "auxiliary_loss_clip": 0.01172562,
      "auxiliary_loss_mlp": 0.01025678,
      "balance_loss_clip": 1.05263638,
      "balance_loss_mlp": 1.01809669,
      "epoch": 0.5989298382733121,
      "flos": 11801719549440.0,
      "grad_norm": 2.063031892929405,
      "language_loss": 0.89316475,
      "learning_rate": 1.4629506950786707e-06,
      "loss": 0.91514719,
      "num_input_tokens_seen": 107394860,
      "step": 4981,
      "time_per_iteration": 2.464160442352295
    },
    {
      "auxiliary_loss_clip": 0.01063521,
      "auxiliary_loss_mlp": 0.01002464,
      "balance_loss_clip": 1.01471531,
      "balance_loss_mlp": 1.00145102,
      "epoch": 0.5990500811639512,
      "flos": 60025800021120.0,
      "grad_norm": 1.1930710231793182,
      "language_loss": 0.56083143,
      "learning_rate": 1.4622003736231733e-06,
      "loss": 0.58149129,
      "num_input_tokens_seen": 107453850,
      "step": 4982,
      "time_per_iteration": 3.131883144378662
    },
    {
      "auxiliary_loss_clip": 0.01152956,
      "auxiliary_loss_mlp": 0.01037829,
      "balance_loss_clip": 1.04883218,
      "balance_loss_mlp": 1.03052187,
      "epoch": 0.5991703240545903,
      "flos": 18222726683520.0,
      "grad_norm": 1.7603372419566368,
      "language_loss": 0.80887842,
      "learning_rate": 1.461450133751451e-06,
      "loss": 0.83078635,
      "num_input_tokens_seen": 107471920,
      "step": 4983,
      "time_per_iteration": 2.450629711151123
    },
    {
      "auxiliary_loss_clip": 0.01156172,
      "auxiliary_loss_mlp": 0.01028753,
      "balance_loss_clip": 1.04892278,
      "balance_loss_mlp": 1.02164268,
      "epoch": 0.5992905669452293,
      "flos": 27709894581120.0,
      "grad_norm": 1.7210128952502044,
      "language_loss": 0.75731546,
      "learning_rate": 1.4606999755773153e-06,
      "loss": 0.77916467,
      "num_input_tokens_seen": 107493125,
      "step": 4984,
      "time_per_iteration": 2.5277669429779053
    },
    {
      "auxiliary_loss_clip": 0.01172326,
      "auxiliary_loss_mlp": 0.01025991,
      "balance_loss_clip": 1.05404556,
      "balance_loss_mlp": 1.01892805,
      "epoch": 0.5994108098358685,
      "flos": 20449008662400.0,
      "grad_norm": 2.083681758306004,
      "language_loss": 0.81989276,
      "learning_rate": 1.4599498992145643e-06,
      "loss": 0.84187591,
      "num_input_tokens_seen": 107513150,
      "step": 4985,
      "time_per_iteration": 2.4487643241882324
    },
    {
      "auxiliary_loss_clip": 0.01149392,
      "auxiliary_loss_mlp": 0.00760888,
      "balance_loss_clip": 1.05087125,
      "balance_loss_mlp": 1.00016153,
      "epoch": 0.5995310527265075,
      "flos": 22269966595200.0,
      "grad_norm": 2.028073502215373,
      "language_loss": 0.70679224,
      "learning_rate": 1.4591999047769846e-06,
      "loss": 0.72589505,
      "num_input_tokens_seen": 107532005,
      "step": 4986,
      "time_per_iteration": 2.5288710594177246
    },
    {
      "auxiliary_loss_clip": 0.01097341,
      "auxiliary_loss_mlp": 0.01027345,
      "balance_loss_clip": 1.0415684,
      "balance_loss_mlp": 1.02018321,
      "epoch": 0.5996512956171466,
      "flos": 18916951818240.0,
      "grad_norm": 2.382431305698634,
      "language_loss": 0.75121605,
      "learning_rate": 1.4584499923783486e-06,
      "loss": 0.77246296,
      "num_input_tokens_seen": 107550585,
      "step": 4987,
      "time_per_iteration": 2.586296796798706
    },
    {
      "auxiliary_loss_clip": 0.01142265,
      "auxiliary_loss_mlp": 0.01029596,
      "balance_loss_clip": 1.04800951,
      "balance_loss_mlp": 1.02257431,
      "epoch": 0.5997715385077858,
      "flos": 15370916330880.0,
      "grad_norm": 1.7222629669012672,
      "language_loss": 0.76159251,
      "learning_rate": 1.457700162132419e-06,
      "loss": 0.78331107,
      "num_input_tokens_seen": 107567575,
      "step": 4988,
      "time_per_iteration": 2.5300979614257812
    },
    {
      "auxiliary_loss_clip": 0.01112994,
      "auxiliary_loss_mlp": 0.01032018,
      "balance_loss_clip": 1.04701173,
      "balance_loss_mlp": 1.02472198,
      "epoch": 0.5998917813984248,
      "flos": 25264844818560.0,
      "grad_norm": 1.9571896677795666,
      "language_loss": 0.72550428,
      "learning_rate": 1.4569504141529433e-06,
      "loss": 0.74695438,
      "num_input_tokens_seen": 107585410,
      "step": 4989,
      "time_per_iteration": 2.648343324661255
    },
    {
      "auxiliary_loss_clip": 0.01152732,
      "auxiliary_loss_mlp": 0.01030156,
      "balance_loss_clip": 1.04880786,
      "balance_loss_mlp": 1.02306938,
      "epoch": 0.6000120242890639,
      "flos": 22054502862720.0,
      "grad_norm": 2.2076063931375547,
      "language_loss": 0.71788228,
      "learning_rate": 1.456200748553658e-06,
      "loss": 0.73971117,
      "num_input_tokens_seen": 107603405,
      "step": 4990,
      "time_per_iteration": 2.5250959396362305
    },
    {
      "auxiliary_loss_clip": 0.01173983,
      "auxiliary_loss_mlp": 0.0103102,
      "balance_loss_clip": 1.05403042,
      "balance_loss_mlp": 1.02368569,
      "epoch": 0.600132267179703,
      "flos": 29863421562240.0,
      "grad_norm": 1.521462865616661,
      "language_loss": 0.78798807,
      "learning_rate": 1.455451165448287e-06,
      "loss": 0.81003809,
      "num_input_tokens_seen": 107626060,
      "step": 4991,
      "time_per_iteration": 2.5222954750061035
    },
    {
      "auxiliary_loss_clip": 0.01140372,
      "auxiliary_loss_mlp": 0.01028986,
      "balance_loss_clip": 1.04875445,
      "balance_loss_mlp": 1.02141857,
      "epoch": 0.6002525100703421,
      "flos": 25045358762880.0,
      "grad_norm": 2.2631740119076538,
      "language_loss": 0.73336339,
      "learning_rate": 1.4547016649505407e-06,
      "loss": 0.75505698,
      "num_input_tokens_seen": 107644070,
      "step": 4992,
      "time_per_iteration": 2.541682720184326
    },
    {
      "auxiliary_loss_clip": 0.0112388,
      "auxiliary_loss_mlp": 0.01029327,
      "balance_loss_clip": 1.04208529,
      "balance_loss_mlp": 1.02195406,
      "epoch": 0.6003727529609811,
      "flos": 20849592113280.0,
      "grad_norm": 2.330119454792163,
      "language_loss": 0.85161555,
      "learning_rate": 1.4539522471741193e-06,
      "loss": 0.87314761,
      "num_input_tokens_seen": 107661495,
      "step": 4993,
      "time_per_iteration": 2.5848255157470703
    },
    {
      "auxiliary_loss_clip": 0.01160248,
      "auxiliary_loss_mlp": 0.01031748,
      "balance_loss_clip": 1.04946947,
      "balance_loss_mlp": 1.02332568,
      "epoch": 0.6004929958516203,
      "flos": 15594604277760.0,
      "grad_norm": 2.1527565979297476,
      "language_loss": 0.70737916,
      "learning_rate": 1.4532029122327067e-06,
      "loss": 0.72929907,
      "num_input_tokens_seen": 107678280,
      "step": 4994,
      "time_per_iteration": 2.4901325702667236
    },
    {
      "auxiliary_loss_clip": 0.01124907,
      "auxiliary_loss_mlp": 0.0102944,
      "balance_loss_clip": 1.04954314,
      "balance_loss_mlp": 1.02223945,
      "epoch": 0.6006132387422594,
      "flos": 21763267390080.0,
      "grad_norm": 1.8854270442171732,
      "language_loss": 0.75616729,
      "learning_rate": 1.4524536602399783e-06,
      "loss": 0.7777108,
      "num_input_tokens_seen": 107697370,
      "step": 4995,
      "time_per_iteration": 3.395029067993164
    },
    {
      "auxiliary_loss_clip": 0.01140542,
      "auxiliary_loss_mlp": 0.01031567,
      "balance_loss_clip": 1.05063426,
      "balance_loss_mlp": 1.0244863,
      "epoch": 0.6007334816328984,
      "flos": 22858542852480.0,
      "grad_norm": 1.5318933953903326,
      "language_loss": 0.77445042,
      "learning_rate": 1.4517044913095938e-06,
      "loss": 0.79617155,
      "num_input_tokens_seen": 107717790,
      "step": 4996,
      "time_per_iteration": 2.5938239097595215
    },
    {
      "auxiliary_loss_clip": 0.01160631,
      "auxiliary_loss_mlp": 0.01027004,
      "balance_loss_clip": 1.05207086,
      "balance_loss_mlp": 1.01940131,
      "epoch": 0.6008537245235376,
      "flos": 28324577047680.0,
      "grad_norm": 1.8716186995571915,
      "language_loss": 0.81571144,
      "learning_rate": 1.4509554055552022e-06,
      "loss": 0.83758777,
      "num_input_tokens_seen": 107738020,
      "step": 4997,
      "time_per_iteration": 2.545639991760254
    },
    {
      "auxiliary_loss_clip": 0.01140713,
      "auxiliary_loss_mlp": 0.01030419,
      "balance_loss_clip": 1.04765427,
      "balance_loss_mlp": 1.02289057,
      "epoch": 0.6009739674141766,
      "flos": 20886113266560.0,
      "grad_norm": 2.724066863539614,
      "language_loss": 0.84136361,
      "learning_rate": 1.450206403090439e-06,
      "loss": 0.86307502,
      "num_input_tokens_seen": 107756215,
      "step": 4998,
      "time_per_iteration": 2.5296237468719482
    },
    {
      "auxiliary_loss_clip": 0.01155399,
      "auxiliary_loss_mlp": 0.01029711,
      "balance_loss_clip": 1.05153394,
      "balance_loss_mlp": 1.02196836,
      "epoch": 0.6010942103048157,
      "flos": 20481004702080.0,
      "grad_norm": 2.2014417986606305,
      "language_loss": 0.86197782,
      "learning_rate": 1.4494574840289274e-06,
      "loss": 0.883829,
      "num_input_tokens_seen": 107773330,
      "step": 4999,
      "time_per_iteration": 2.4913082122802734
    },
    {
      "auxiliary_loss_clip": 0.01161842,
      "auxiliary_loss_mlp": 0.01028612,
      "balance_loss_clip": 1.04896176,
      "balance_loss_mlp": 1.02065229,
      "epoch": 0.6012144531954549,
      "flos": 23805973935360.0,
      "grad_norm": 1.626210647302647,
      "language_loss": 0.73897898,
      "learning_rate": 1.4487086484842782e-06,
      "loss": 0.76088357,
      "num_input_tokens_seen": 107791975,
      "step": 5000,
      "time_per_iteration": 3.3269667625427246
    },
    {
      "auxiliary_loss_clip": 0.01172894,
      "auxiliary_loss_mlp": 0.01025834,
      "balance_loss_clip": 1.05255353,
      "balance_loss_mlp": 1.01861334,
      "epoch": 0.6013346960860939,
      "flos": 18988378012800.0,
      "grad_norm": 2.074818950457689,
      "language_loss": 0.60172033,
      "learning_rate": 1.4479598965700878e-06,
      "loss": 0.62370765,
      "num_input_tokens_seen": 107809240,
      "step": 5001,
      "time_per_iteration": 2.4390101432800293
    },
    {
      "auxiliary_loss_clip": 0.0112751,
      "auxiliary_loss_mlp": 0.01032741,
      "balance_loss_clip": 1.04602087,
      "balance_loss_mlp": 1.02540064,
      "epoch": 0.601454938976733,
      "flos": 24025316336640.0,
      "grad_norm": 2.260677376856249,
      "language_loss": 0.690153,
      "learning_rate": 1.4472112283999427e-06,
      "loss": 0.71175551,
      "num_input_tokens_seen": 107827895,
      "step": 5002,
      "time_per_iteration": 2.572288990020752
    },
    {
      "auxiliary_loss_clip": 0.0115176,
      "auxiliary_loss_mlp": 0.01030213,
      "balance_loss_clip": 1.04986107,
      "balance_loss_mlp": 1.0231235,
      "epoch": 0.6015751818673721,
      "flos": 26427129102720.0,
      "grad_norm": 1.8766889377158704,
      "language_loss": 0.69073755,
      "learning_rate": 1.4464626440874143e-06,
      "loss": 0.71255732,
      "num_input_tokens_seen": 107847010,
      "step": 5003,
      "time_per_iteration": 3.376095771789551
    },
    {
      "auxiliary_loss_clip": 0.01118402,
      "auxiliary_loss_mlp": 0.01026422,
      "balance_loss_clip": 1.04218197,
      "balance_loss_mlp": 1.01850653,
      "epoch": 0.6016954247580112,
      "flos": 13115260005120.0,
      "grad_norm": 2.592184211996956,
      "language_loss": 0.74434882,
      "learning_rate": 1.4457141437460636e-06,
      "loss": 0.76579702,
      "num_input_tokens_seen": 107864235,
      "step": 5004,
      "time_per_iteration": 2.562472105026245
    },
    {
      "auxiliary_loss_clip": 0.0114437,
      "auxiliary_loss_mlp": 0.01028972,
      "balance_loss_clip": 1.049739,
      "balance_loss_mlp": 1.02103555,
      "epoch": 0.6018156676486502,
      "flos": 23768447201280.0,
      "grad_norm": 1.6987704466959332,
      "language_loss": 0.73268712,
      "learning_rate": 1.444965727489436e-06,
      "loss": 0.75442052,
      "num_input_tokens_seen": 107883680,
      "step": 5005,
      "time_per_iteration": 3.356118679046631
    },
    {
      "auxiliary_loss_clip": 0.0112791,
      "auxiliary_loss_mlp": 0.0103137,
      "balance_loss_clip": 1.04429018,
      "balance_loss_mlp": 1.02459025,
      "epoch": 0.6019359105392894,
      "flos": 26469360518400.0,
      "grad_norm": 2.342467979886596,
      "language_loss": 0.63484031,
      "learning_rate": 1.444217395431066e-06,
      "loss": 0.65643311,
      "num_input_tokens_seen": 107906220,
      "step": 5006,
      "time_per_iteration": 2.6083390712738037
    },
    {
      "auxiliary_loss_clip": 0.01031577,
      "auxiliary_loss_mlp": 0.01001203,
      "balance_loss_clip": 1.01548362,
      "balance_loss_mlp": 1.00007617,
      "epoch": 0.6020561534299285,
      "flos": 69190849728000.0,
      "grad_norm": 0.8050056976344008,
      "language_loss": 0.55850816,
      "learning_rate": 1.4434691476844755e-06,
      "loss": 0.57883596,
      "num_input_tokens_seen": 107967195,
      "step": 5007,
      "time_per_iteration": 3.128286838531494
    },
    {
      "auxiliary_loss_clip": 0.01141738,
      "auxiliary_loss_mlp": 0.0102828,
      "balance_loss_clip": 1.04991913,
      "balance_loss_mlp": 1.02120495,
      "epoch": 0.6021763963205675,
      "flos": 21835304115840.0,
      "grad_norm": 2.2594764321591883,
      "language_loss": 0.66796839,
      "learning_rate": 1.4427209843631729e-06,
      "loss": 0.6896686,
      "num_input_tokens_seen": 107984245,
      "step": 5008,
      "time_per_iteration": 2.5236167907714844
    },
    {
      "auxiliary_loss_clip": 0.0116872,
      "auxiliary_loss_mlp": 0.00760807,
      "balance_loss_clip": 1.05076206,
      "balance_loss_mlp": 1.00015819,
      "epoch": 0.6022966392112067,
      "flos": 26578636669440.0,
      "grad_norm": 2.000628927777027,
      "language_loss": 0.80750167,
      "learning_rate": 1.4419729055806534e-06,
      "loss": 0.82679695,
      "num_input_tokens_seen": 108003680,
      "step": 5009,
      "time_per_iteration": 2.5175673961639404
    },
    {
      "auxiliary_loss_clip": 0.01141703,
      "auxiliary_loss_mlp": 0.00761072,
      "balance_loss_clip": 1.05177784,
      "balance_loss_mlp": 1.00015187,
      "epoch": 0.6024168821018457,
      "flos": 20703722981760.0,
      "grad_norm": 1.7071341725266582,
      "language_loss": 0.81984186,
      "learning_rate": 1.441224911450401e-06,
      "loss": 0.83886969,
      "num_input_tokens_seen": 108019635,
      "step": 5010,
      "time_per_iteration": 2.5113325119018555
    },
    {
      "auxiliary_loss_clip": 0.01162352,
      "auxiliary_loss_mlp": 0.01029791,
      "balance_loss_clip": 1.05160356,
      "balance_loss_mlp": 1.0223465,
      "epoch": 0.6025371249924848,
      "flos": 24680973242880.0,
      "grad_norm": 1.6305472870591722,
      "language_loss": 0.82173908,
      "learning_rate": 1.4404770020858851e-06,
      "loss": 0.84366059,
      "num_input_tokens_seen": 108039120,
      "step": 5011,
      "time_per_iteration": 2.527385711669922
    },
    {
      "auxiliary_loss_clip": 0.01150365,
      "auxiliary_loss_mlp": 0.01023982,
      "balance_loss_clip": 1.04833531,
      "balance_loss_mlp": 1.01695442,
      "epoch": 0.602657367883124,
      "flos": 25955801815680.0,
      "grad_norm": 1.7352037619373997,
      "language_loss": 0.85877842,
      "learning_rate": 1.439729177600563e-06,
      "loss": 0.88052189,
      "num_input_tokens_seen": 108059615,
      "step": 5012,
      "time_per_iteration": 2.5252771377563477
    },
    {
      "auxiliary_loss_clip": 0.01158304,
      "auxiliary_loss_mlp": 0.01027159,
      "balance_loss_clip": 1.05130601,
      "balance_loss_mlp": 1.01963401,
      "epoch": 0.602777610773763,
      "flos": 16690633925760.0,
      "grad_norm": 1.9211105861191367,
      "language_loss": 0.72947651,
      "learning_rate": 1.4389814381078793e-06,
      "loss": 0.75133115,
      "num_input_tokens_seen": 108078855,
      "step": 5013,
      "time_per_iteration": 2.5023303031921387
    },
    {
      "auxiliary_loss_clip": 0.0106547,
      "auxiliary_loss_mlp": 0.01036073,
      "balance_loss_clip": 1.04346442,
      "balance_loss_mlp": 1.0284673,
      "epoch": 0.6028978536644021,
      "flos": 13334243270400.0,
      "grad_norm": 1.913315598122271,
      "language_loss": 0.80133998,
      "learning_rate": 1.438233783721265e-06,
      "loss": 0.82235539,
      "num_input_tokens_seen": 108095020,
      "step": 5014,
      "time_per_iteration": 2.9828877449035645
    },
    {
      "auxiliary_loss_clip": 0.01139291,
      "auxiliary_loss_mlp": 0.01032058,
      "balance_loss_clip": 1.05010903,
      "balance_loss_mlp": 1.02455056,
      "epoch": 0.6030180965550412,
      "flos": 19644825018240.0,
      "grad_norm": 1.8961866703507466,
      "language_loss": 0.78032362,
      "learning_rate": 1.43748621455414e-06,
      "loss": 0.80203712,
      "num_input_tokens_seen": 108111455,
      "step": 5015,
      "time_per_iteration": 2.6738622188568115
    },
    {
      "auxiliary_loss_clip": 0.01140652,
      "auxiliary_loss_mlp": 0.01027358,
      "balance_loss_clip": 1.04843664,
      "balance_loss_mlp": 1.01984787,
      "epoch": 0.6031383394456803,
      "flos": 14458390289280.0,
      "grad_norm": 2.362249084316124,
      "language_loss": 0.80117351,
      "learning_rate": 1.4367387307199082e-06,
      "loss": 0.82285357,
      "num_input_tokens_seen": 108128305,
      "step": 5016,
      "time_per_iteration": 2.5071182250976562
    },
    {
      "auxiliary_loss_clip": 0.01149982,
      "auxiliary_loss_mlp": 0.01032171,
      "balance_loss_clip": 1.04597187,
      "balance_loss_mlp": 1.02495289,
      "epoch": 0.6032585823363193,
      "flos": 13917791623680.0,
      "grad_norm": 3.1658959208693584,
      "language_loss": 0.82427037,
      "learning_rate": 1.4359913323319632e-06,
      "loss": 0.84609193,
      "num_input_tokens_seen": 108145475,
      "step": 5017,
      "time_per_iteration": 2.473132848739624
    },
    {
      "auxiliary_loss_clip": 0.01091245,
      "auxiliary_loss_mlp": 0.01029435,
      "balance_loss_clip": 1.04056907,
      "balance_loss_mlp": 1.02199078,
      "epoch": 0.6033788252269584,
      "flos": 24353252530560.0,
      "grad_norm": 1.9468612562311294,
      "language_loss": 0.77517068,
      "learning_rate": 1.4352440195036847e-06,
      "loss": 0.79637748,
      "num_input_tokens_seen": 108165650,
      "step": 5018,
      "time_per_iteration": 2.657898187637329
    },
    {
      "auxiliary_loss_clip": 0.01088682,
      "auxiliary_loss_mlp": 0.01025708,
      "balance_loss_clip": 1.03953481,
      "balance_loss_mlp": 1.01827502,
      "epoch": 0.6034990681175976,
      "flos": 25521247077120.0,
      "grad_norm": 1.553084400981856,
      "language_loss": 0.79962242,
      "learning_rate": 1.4344967923484395e-06,
      "loss": 0.82076627,
      "num_input_tokens_seen": 108187620,
      "step": 5019,
      "time_per_iteration": 2.7157187461853027
    },
    {
      "auxiliary_loss_clip": 0.01154169,
      "auxiliary_loss_mlp": 0.0102469,
      "balance_loss_clip": 1.04833627,
      "balance_loss_mlp": 1.01684618,
      "epoch": 0.6036193110082366,
      "flos": 25958387594880.0,
      "grad_norm": 2.11648332857497,
      "language_loss": 0.71973741,
      "learning_rate": 1.433749650979581e-06,
      "loss": 0.74152601,
      "num_input_tokens_seen": 108207605,
      "step": 5020,
      "time_per_iteration": 3.544306993484497
    },
    {
      "auxiliary_loss_clip": 0.01130646,
      "auxiliary_loss_mlp": 0.01026528,
      "balance_loss_clip": 1.04597783,
      "balance_loss_mlp": 1.01919675,
      "epoch": 0.6037395538988757,
      "flos": 25593427457280.0,
      "grad_norm": 1.884362125407266,
      "language_loss": 0.68122274,
      "learning_rate": 1.433002595510451e-06,
      "loss": 0.70279443,
      "num_input_tokens_seen": 108226385,
      "step": 5021,
      "time_per_iteration": 2.6176869869232178
    },
    {
      "auxiliary_loss_clip": 0.01139528,
      "auxiliary_loss_mlp": 0.00761298,
      "balance_loss_clip": 1.04784632,
      "balance_loss_mlp": 1.00017428,
      "epoch": 0.6038597967895148,
      "flos": 17816253402240.0,
      "grad_norm": 1.8691836255799785,
      "language_loss": 0.72049999,
      "learning_rate": 1.4322556260543757e-06,
      "loss": 0.73950827,
      "num_input_tokens_seen": 108242960,
      "step": 5022,
      "time_per_iteration": 2.5375945568084717
    },
    {
      "auxiliary_loss_clip": 0.0103335,
      "auxiliary_loss_mlp": 0.01002663,
      "balance_loss_clip": 1.01370335,
      "balance_loss_mlp": 1.00154209,
      "epoch": 0.6039800396801539,
      "flos": 65169213235200.0,
      "grad_norm": 0.8912829236257772,
      "language_loss": 0.62697852,
      "learning_rate": 1.4315087427246703e-06,
      "loss": 0.64733863,
      "num_input_tokens_seen": 108296785,
      "step": 5023,
      "time_per_iteration": 3.0460522174835205
    },
    {
      "auxiliary_loss_clip": 0.01063874,
      "auxiliary_loss_mlp": 0.01000174,
      "balance_loss_clip": 1.01485348,
      "balance_loss_mlp": 0.9990595,
      "epoch": 0.604100282570793,
      "flos": 67386409073280.0,
      "grad_norm": 0.8752576712014443,
      "language_loss": 0.58510673,
      "learning_rate": 1.4307619456346372e-06,
      "loss": 0.6057471,
      "num_input_tokens_seen": 108341090,
      "step": 5024,
      "time_per_iteration": 2.781691312789917
    },
    {
      "auxiliary_loss_clip": 0.0115599,
      "auxiliary_loss_mlp": 0.01031215,
      "balance_loss_clip": 1.04696679,
      "balance_loss_mlp": 1.02350175,
      "epoch": 0.6042205254614321,
      "flos": 35297495631360.0,
      "grad_norm": 1.8392660669120915,
      "language_loss": 0.74428624,
      "learning_rate": 1.430015234897564e-06,
      "loss": 0.76615828,
      "num_input_tokens_seen": 108364370,
      "step": 5025,
      "time_per_iteration": 2.6645851135253906
    },
    {
      "auxiliary_loss_clip": 0.01168098,
      "auxiliary_loss_mlp": 0.00760665,
      "balance_loss_clip": 1.04810953,
      "balance_loss_mlp": 1.00020099,
      "epoch": 0.6043407683520712,
      "flos": 45658262206080.0,
      "grad_norm": 1.9419058352399683,
      "language_loss": 0.66698223,
      "learning_rate": 1.4292686106267274e-06,
      "loss": 0.68626982,
      "num_input_tokens_seen": 108387220,
      "step": 5026,
      "time_per_iteration": 2.6779115200042725
    },
    {
      "auxiliary_loss_clip": 0.01161443,
      "auxiliary_loss_mlp": 0.01029924,
      "balance_loss_clip": 1.05058587,
      "balance_loss_mlp": 1.0225209,
      "epoch": 0.6044610112427102,
      "flos": 16180020138240.0,
      "grad_norm": 1.7201349394366874,
      "language_loss": 0.76950133,
      "learning_rate": 1.4285220729353876e-06,
      "loss": 0.79141498,
      "num_input_tokens_seen": 108405760,
      "step": 5027,
      "time_per_iteration": 3.263826370239258
    },
    {
      "auxiliary_loss_clip": 0.01142944,
      "auxiliary_loss_mlp": 0.01028721,
      "balance_loss_clip": 1.04552758,
      "balance_loss_mlp": 1.0215416,
      "epoch": 0.6045812541333494,
      "flos": 13804062186240.0,
      "grad_norm": 1.9173193395582409,
      "language_loss": 0.77169096,
      "learning_rate": 1.4277756219367957e-06,
      "loss": 0.79340756,
      "num_input_tokens_seen": 108422785,
      "step": 5028,
      "time_per_iteration": 2.5185365676879883
    },
    {
      "auxiliary_loss_clip": 0.01134698,
      "auxiliary_loss_mlp": 0.01028397,
      "balance_loss_clip": 1.04797542,
      "balance_loss_mlp": 1.02133954,
      "epoch": 0.6047014970239885,
      "flos": 19975059682560.0,
      "grad_norm": 2.373948950005817,
      "language_loss": 0.79629922,
      "learning_rate": 1.4270292577441864e-06,
      "loss": 0.8179301,
      "num_input_tokens_seen": 108442290,
      "step": 5029,
      "time_per_iteration": 3.6376543045043945
    },
    {
      "auxiliary_loss_clip": 0.0115865,
      "auxiliary_loss_mlp": 0.01026968,
      "balance_loss_clip": 1.04741406,
      "balance_loss_mlp": 1.01924968,
      "epoch": 0.6048217399146275,
      "flos": 25337097025920.0,
      "grad_norm": 1.734029427462117,
      "language_loss": 0.71954924,
      "learning_rate": 1.4262829804707836e-06,
      "loss": 0.74140537,
      "num_input_tokens_seen": 108464280,
      "step": 5030,
      "time_per_iteration": 2.5502326488494873
    },
    {
      "auxiliary_loss_clip": 0.01156592,
      "auxiliary_loss_mlp": 0.01031136,
      "balance_loss_clip": 1.04770315,
      "balance_loss_mlp": 1.02317882,
      "epoch": 0.6049419828052667,
      "flos": 26030819370240.0,
      "grad_norm": 1.4065751727777187,
      "language_loss": 0.69818598,
      "learning_rate": 1.4255367902297958e-06,
      "loss": 0.72006327,
      "num_input_tokens_seen": 108485610,
      "step": 5031,
      "time_per_iteration": 3.2819416522979736
    },
    {
      "auxiliary_loss_clip": 0.01169679,
      "auxiliary_loss_mlp": 0.01026393,
      "balance_loss_clip": 1.05160785,
      "balance_loss_mlp": 1.01915073,
      "epoch": 0.6050622256959057,
      "flos": 14648106948480.0,
      "grad_norm": 2.0407381611507884,
      "language_loss": 0.78800255,
      "learning_rate": 1.4247906871344215e-06,
      "loss": 0.80996323,
      "num_input_tokens_seen": 108501005,
      "step": 5032,
      "time_per_iteration": 2.4371049404144287
    },
    {
      "auxiliary_loss_clip": 0.01137004,
      "auxiliary_loss_mlp": 0.01027924,
      "balance_loss_clip": 1.0445894,
      "balance_loss_mlp": 1.02084279,
      "epoch": 0.6051824685865448,
      "flos": 23331450337920.0,
      "grad_norm": 2.0528535245311135,
      "language_loss": 0.74954277,
      "learning_rate": 1.4240446712978415e-06,
      "loss": 0.77119207,
      "num_input_tokens_seen": 108519990,
      "step": 5033,
      "time_per_iteration": 2.5370421409606934
    },
    {
      "auxiliary_loss_clip": 0.01160855,
      "auxiliary_loss_mlp": 0.01028043,
      "balance_loss_clip": 1.04959178,
      "balance_loss_mlp": 1.02087569,
      "epoch": 0.605302711477184,
      "flos": 27563307177600.0,
      "grad_norm": 1.8582921908899248,
      "language_loss": 0.74431491,
      "learning_rate": 1.423298742833227e-06,
      "loss": 0.76620388,
      "num_input_tokens_seen": 108538650,
      "step": 5034,
      "time_per_iteration": 2.554076671600342
    },
    {
      "auxiliary_loss_clip": 0.01133233,
      "auxiliary_loss_mlp": 0.01026169,
      "balance_loss_clip": 1.04476261,
      "balance_loss_mlp": 1.01834321,
      "epoch": 0.605422954367823,
      "flos": 15154698412800.0,
      "grad_norm": 1.9074758464865238,
      "language_loss": 0.71623856,
      "learning_rate": 1.4225529018537352e-06,
      "loss": 0.73783255,
      "num_input_tokens_seen": 108554155,
      "step": 5035,
      "time_per_iteration": 2.5255210399627686
    },
    {
      "auxiliary_loss_clip": 0.01170124,
      "auxiliary_loss_mlp": 0.01029439,
      "balance_loss_clip": 1.05102158,
      "balance_loss_mlp": 1.02208924,
      "epoch": 0.6055431972584621,
      "flos": 27673912131840.0,
      "grad_norm": 1.5682219295038988,
      "language_loss": 0.7759645,
      "learning_rate": 1.4218071484725082e-06,
      "loss": 0.7979601,
      "num_input_tokens_seen": 108576275,
      "step": 5036,
      "time_per_iteration": 2.5155751705169678
    },
    {
      "auxiliary_loss_clip": 0.01140313,
      "auxiliary_loss_mlp": 0.01029841,
      "balance_loss_clip": 1.04978085,
      "balance_loss_mlp": 1.02194381,
      "epoch": 0.6056634401491012,
      "flos": 19387489006080.0,
      "grad_norm": 1.8116777868601297,
      "language_loss": 0.76052415,
      "learning_rate": 1.4210614828026786e-06,
      "loss": 0.78222573,
      "num_input_tokens_seen": 108594125,
      "step": 5037,
      "time_per_iteration": 2.5189146995544434
    },
    {
      "auxiliary_loss_clip": 0.01168178,
      "auxiliary_loss_mlp": 0.01027302,
      "balance_loss_clip": 1.04875851,
      "balance_loss_mlp": 1.01997662,
      "epoch": 0.6057836830397403,
      "flos": 24789459294720.0,
      "grad_norm": 1.4431975136755333,
      "language_loss": 0.74256539,
      "learning_rate": 1.4203159049573605e-06,
      "loss": 0.76452017,
      "num_input_tokens_seen": 108615360,
      "step": 5038,
      "time_per_iteration": 2.492875337600708
    },
    {
      "auxiliary_loss_clip": 0.01144783,
      "auxiliary_loss_mlp": 0.01032171,
      "balance_loss_clip": 1.04657054,
      "balance_loss_mlp": 1.02422619,
      "epoch": 0.6059039259303793,
      "flos": 20558248899840.0,
      "grad_norm": 2.1016629590022644,
      "language_loss": 0.86850607,
      "learning_rate": 1.4195704150496593e-06,
      "loss": 0.89027572,
      "num_input_tokens_seen": 108633075,
      "step": 5039,
      "time_per_iteration": 2.5013492107391357
    },
    {
      "auxiliary_loss_clip": 0.01141825,
      "auxiliary_loss_mlp": 0.01031541,
      "balance_loss_clip": 1.04902124,
      "balance_loss_mlp": 1.02423906,
      "epoch": 0.6060241688210185,
      "flos": 21069724613760.0,
      "grad_norm": 2.3025083398920865,
      "language_loss": 0.74200451,
      "learning_rate": 1.4188250131926639e-06,
      "loss": 0.76373816,
      "num_input_tokens_seen": 108651875,
      "step": 5040,
      "time_per_iteration": 2.5423552989959717
    },
    {
      "auxiliary_loss_clip": 0.01142352,
      "auxiliary_loss_mlp": 0.01028038,
      "balance_loss_clip": 1.04623497,
      "balance_loss_mlp": 1.0202837,
      "epoch": 0.6061444117116576,
      "flos": 16361081619840.0,
      "grad_norm": 1.787963858499802,
      "language_loss": 0.80430949,
      "learning_rate": 1.4180796994994525e-06,
      "loss": 0.82601345,
      "num_input_tokens_seen": 108669290,
      "step": 5041,
      "time_per_iteration": 2.4954183101654053
    },
    {
      "auxiliary_loss_clip": 0.01141024,
      "auxiliary_loss_mlp": 0.01024772,
      "balance_loss_clip": 1.04632294,
      "balance_loss_mlp": 1.01767302,
      "epoch": 0.6062646546022966,
      "flos": 21507296094720.0,
      "grad_norm": 1.928442226117961,
      "language_loss": 0.71961474,
      "learning_rate": 1.4173344740830877e-06,
      "loss": 0.74127269,
      "num_input_tokens_seen": 108688420,
      "step": 5042,
      "time_per_iteration": 2.5367302894592285
    },
    {
      "auxiliary_loss_clip": 0.01133586,
      "auxiliary_loss_mlp": 0.01032335,
      "balance_loss_clip": 1.0463922,
      "balance_loss_mlp": 1.02470541,
      "epoch": 0.6063848974929358,
      "flos": 38983151283840.0,
      "grad_norm": 1.626895413962969,
      "language_loss": 0.70770895,
      "learning_rate": 1.4165893370566206e-06,
      "loss": 0.72936809,
      "num_input_tokens_seen": 108712175,
      "step": 5043,
      "time_per_iteration": 2.680866003036499
    },
    {
      "auxiliary_loss_clip": 0.0115119,
      "auxiliary_loss_mlp": 0.01031736,
      "balance_loss_clip": 1.0475806,
      "balance_loss_mlp": 1.0240705,
      "epoch": 0.6065051403835748,
      "flos": 19646584784640.0,
      "grad_norm": 1.671285411578337,
      "language_loss": 0.77359647,
      "learning_rate": 1.4158442885330865e-06,
      "loss": 0.79542571,
      "num_input_tokens_seen": 108730745,
      "step": 5044,
      "time_per_iteration": 2.517709255218506
    },
    {
      "auxiliary_loss_clip": 0.01148676,
      "auxiliary_loss_mlp": 0.01027954,
      "balance_loss_clip": 1.046296,
      "balance_loss_mlp": 1.02058697,
      "epoch": 0.6066253832742139,
      "flos": 23513086437120.0,
      "grad_norm": 2.2595020639729397,
      "language_loss": 0.78854567,
      "learning_rate": 1.4150993286255094e-06,
      "loss": 0.81031191,
      "num_input_tokens_seen": 108749995,
      "step": 5045,
      "time_per_iteration": 2.5145297050476074
    },
    {
      "auxiliary_loss_clip": 0.01168425,
      "auxiliary_loss_mlp": 0.01025041,
      "balance_loss_clip": 1.04864192,
      "balance_loss_mlp": 1.01819873,
      "epoch": 0.6067456261648531,
      "flos": 19133708440320.0,
      "grad_norm": 2.3081722415663832,
      "language_loss": 0.7969625,
      "learning_rate": 1.4143544574468993e-06,
      "loss": 0.81889725,
      "num_input_tokens_seen": 108768355,
      "step": 5046,
      "time_per_iteration": 3.2486488819122314
    },
    {
      "auxiliary_loss_clip": 0.01153536,
      "auxiliary_loss_mlp": 0.01023674,
      "balance_loss_clip": 1.04908299,
      "balance_loss_mlp": 1.01647675,
      "epoch": 0.6068658690554921,
      "flos": 20520614424960.0,
      "grad_norm": 1.7980539147425736,
      "language_loss": 0.82365543,
      "learning_rate": 1.4136096751102523e-06,
      "loss": 0.84542757,
      "num_input_tokens_seen": 108786685,
      "step": 5047,
      "time_per_iteration": 2.4696905612945557
    },
    {
      "auxiliary_loss_clip": 0.01145681,
      "auxiliary_loss_mlp": 0.01033755,
      "balance_loss_clip": 1.04978609,
      "balance_loss_mlp": 1.02615893,
      "epoch": 0.6069861119461312,
      "flos": 27374560185600.0,
      "grad_norm": 2.2578256745616905,
      "language_loss": 0.83113116,
      "learning_rate": 1.4128649817285516e-06,
      "loss": 0.85292554,
      "num_input_tokens_seen": 108804820,
      "step": 5048,
      "time_per_iteration": 2.5480709075927734
    },
    {
      "auxiliary_loss_clip": 0.01144217,
      "auxiliary_loss_mlp": 0.01037267,
      "balance_loss_clip": 1.0462687,
      "balance_loss_mlp": 1.02821898,
      "epoch": 0.6071063548367702,
      "flos": 25626500904960.0,
      "grad_norm": 2.067967429811361,
      "language_loss": 0.63019472,
      "learning_rate": 1.412120377414766e-06,
      "loss": 0.65200955,
      "num_input_tokens_seen": 108825010,
      "step": 5049,
      "time_per_iteration": 2.569143295288086
    },
    {
      "auxiliary_loss_clip": 0.01169189,
      "auxiliary_loss_mlp": 0.01028863,
      "balance_loss_clip": 1.05025578,
      "balance_loss_mlp": 1.02163839,
      "epoch": 0.6072265977274094,
      "flos": 24460517520000.0,
      "grad_norm": 1.5601169010285756,
      "language_loss": 0.71248406,
      "learning_rate": 1.4113758622818522e-06,
      "loss": 0.73446453,
      "num_input_tokens_seen": 108845075,
      "step": 5050,
      "time_per_iteration": 2.473106622695923
    },
    {
      "auxiliary_loss_clip": 0.01147405,
      "auxiliary_loss_mlp": 0.00760056,
      "balance_loss_clip": 1.05008459,
      "balance_loss_mlp": 1.00016809,
      "epoch": 0.6073468406180484,
      "flos": 18149253413760.0,
      "grad_norm": 2.0102643311803594,
      "language_loss": 0.83410805,
      "learning_rate": 1.410631436442751e-06,
      "loss": 0.85318267,
      "num_input_tokens_seen": 108863870,
      "step": 5051,
      "time_per_iteration": 2.5103046894073486
    },
    {
      "auxiliary_loss_clip": 0.01162416,
      "auxiliary_loss_mlp": 0.01030772,
      "balance_loss_clip": 1.0504868,
      "balance_loss_mlp": 1.02297306,
      "epoch": 0.6074670835086875,
      "flos": 20697617669760.0,
      "grad_norm": 2.1185557217760107,
      "language_loss": 0.8626864,
      "learning_rate": 1.4098871000103936e-06,
      "loss": 0.88461822,
      "num_input_tokens_seen": 108882470,
      "step": 5052,
      "time_per_iteration": 3.259561061859131
    },
    {
      "auxiliary_loss_clip": 0.01141407,
      "auxiliary_loss_mlp": 0.01029747,
      "balance_loss_clip": 1.04696882,
      "balance_loss_mlp": 1.02248442,
      "epoch": 0.6075873263993267,
      "flos": 23769955572480.0,
      "grad_norm": 1.7385177174217754,
      "language_loss": 0.8265267,
      "learning_rate": 1.409142853097693e-06,
      "loss": 0.84823829,
      "num_input_tokens_seen": 108902710,
      "step": 5053,
      "time_per_iteration": 2.521294116973877
    },
    {
      "auxiliary_loss_clip": 0.01145355,
      "auxiliary_loss_mlp": 0.01032823,
      "balance_loss_clip": 1.04974174,
      "balance_loss_mlp": 1.02527738,
      "epoch": 0.6077075692899657,
      "flos": 24454484035200.0,
      "grad_norm": 1.8619629891851506,
      "language_loss": 0.7926302,
      "learning_rate": 1.408398695817553e-06,
      "loss": 0.81441194,
      "num_input_tokens_seen": 108919935,
      "step": 5054,
      "time_per_iteration": 2.537020206451416
    },
    {
      "auxiliary_loss_clip": 0.01142889,
      "auxiliary_loss_mlp": 0.01028864,
      "balance_loss_clip": 1.04665685,
      "balance_loss_mlp": 1.02068591,
      "epoch": 0.6078278121806048,
      "flos": 27382102041600.0,
      "grad_norm": 1.746642085765271,
      "language_loss": 0.69873393,
      "learning_rate": 1.4076546282828593e-06,
      "loss": 0.72045147,
      "num_input_tokens_seen": 108942790,
      "step": 5055,
      "time_per_iteration": 2.572387933731079
    },
    {
      "auxiliary_loss_clip": 0.01144373,
      "auxiliary_loss_mlp": 0.01024083,
      "balance_loss_clip": 1.0441277,
      "balance_loss_mlp": 1.01620579,
      "epoch": 0.6079480550712439,
      "flos": 38436447306240.0,
      "grad_norm": 2.2600734127843074,
      "language_loss": 0.66149127,
      "learning_rate": 1.4069106506064874e-06,
      "loss": 0.6831758,
      "num_input_tokens_seen": 108964215,
      "step": 5056,
      "time_per_iteration": 4.191065073013306
    },
    {
      "auxiliary_loss_clip": 0.01138103,
      "auxiliary_loss_mlp": 0.01023523,
      "balance_loss_clip": 1.04938555,
      "balance_loss_mlp": 1.01637018,
      "epoch": 0.608068297961883,
      "flos": 25336271013120.0,
      "grad_norm": 1.7104498192279372,
      "language_loss": 0.78463447,
      "learning_rate": 1.4061667629012989e-06,
      "loss": 0.80625081,
      "num_input_tokens_seen": 108984885,
      "step": 5057,
      "time_per_iteration": 2.570335865020752
    },
    {
      "auxiliary_loss_clip": 0.01134633,
      "auxiliary_loss_mlp": 0.01028496,
      "balance_loss_clip": 1.04867816,
      "balance_loss_mlp": 1.02150989,
      "epoch": 0.608188540852522,
      "flos": 24202463235840.0,
      "grad_norm": 2.2961106713848607,
      "language_loss": 0.83135146,
      "learning_rate": 1.40542296528014e-06,
      "loss": 0.85298276,
      "num_input_tokens_seen": 109004545,
      "step": 5058,
      "time_per_iteration": 2.5524842739105225
    },
    {
      "auxiliary_loss_clip": 0.01155536,
      "auxiliary_loss_mlp": 0.01030942,
      "balance_loss_clip": 1.04722738,
      "balance_loss_mlp": 1.02319336,
      "epoch": 0.6083087837431612,
      "flos": 21284146851840.0,
      "grad_norm": 1.9347432085659586,
      "language_loss": 0.76178789,
      "learning_rate": 1.4046792578558452e-06,
      "loss": 0.78365266,
      "num_input_tokens_seen": 109022440,
      "step": 5059,
      "time_per_iteration": 2.5003468990325928
    },
    {
      "auxiliary_loss_clip": 0.01136476,
      "auxiliary_loss_mlp": 0.0103122,
      "balance_loss_clip": 1.0461483,
      "balance_loss_mlp": 1.02392411,
      "epoch": 0.6084290266338003,
      "flos": 16471435178880.0,
      "grad_norm": 3.2171644880225685,
      "language_loss": 0.75511885,
      "learning_rate": 1.4039356407412325e-06,
      "loss": 0.7767958,
      "num_input_tokens_seen": 109035680,
      "step": 5060,
      "time_per_iteration": 2.4715845584869385
    },
    {
      "auxiliary_loss_clip": 0.01057504,
      "auxiliary_loss_mlp": 0.01006219,
      "balance_loss_clip": 1.01745939,
      "balance_loss_mlp": 1.00492573,
      "epoch": 0.6085492695244393,
      "flos": 66443574931200.0,
      "grad_norm": 0.7831537492150985,
      "language_loss": 0.57146549,
      "learning_rate": 1.40319211404911e-06,
      "loss": 0.59210277,
      "num_input_tokens_seen": 109090680,
      "step": 5061,
      "time_per_iteration": 3.066530227661133
    },
    {
      "auxiliary_loss_clip": 0.01171089,
      "auxiliary_loss_mlp": 0.01030953,
      "balance_loss_clip": 1.05088449,
      "balance_loss_mlp": 1.02378893,
      "epoch": 0.6086695124150785,
      "flos": 23618986709760.0,
      "grad_norm": 1.6985727293282236,
      "language_loss": 0.90548825,
      "learning_rate": 1.4024486778922691e-06,
      "loss": 0.92750871,
      "num_input_tokens_seen": 109108995,
      "step": 5062,
      "time_per_iteration": 2.544379472732544
    },
    {
      "auxiliary_loss_clip": 0.0114634,
      "auxiliary_loss_mlp": 0.01028583,
      "balance_loss_clip": 1.04569292,
      "balance_loss_mlp": 1.02160597,
      "epoch": 0.6087897553057176,
      "flos": 20157054917760.0,
      "grad_norm": 1.9023240487785429,
      "language_loss": 0.76764518,
      "learning_rate": 1.4017053323834884e-06,
      "loss": 0.78939438,
      "num_input_tokens_seen": 109128825,
      "step": 5063,
      "time_per_iteration": 2.530339002609253
    },
    {
      "auxiliary_loss_clip": 0.01142137,
      "auxiliary_loss_mlp": 0.01027493,
      "balance_loss_clip": 1.04512393,
      "balance_loss_mlp": 1.02019775,
      "epoch": 0.6089099981963566,
      "flos": 25482535194240.0,
      "grad_norm": 2.0078600141272247,
      "language_loss": 0.76283896,
      "learning_rate": 1.4009620776355333e-06,
      "loss": 0.78453529,
      "num_input_tokens_seen": 109150425,
      "step": 5064,
      "time_per_iteration": 2.5701327323913574
    },
    {
      "auxiliary_loss_clip": 0.01153238,
      "auxiliary_loss_mlp": 0.0103539,
      "balance_loss_clip": 1.04729366,
      "balance_loss_mlp": 1.02723002,
      "epoch": 0.6090302410869958,
      "flos": 25332895134720.0,
      "grad_norm": 1.756782841512221,
      "language_loss": 0.79081744,
      "learning_rate": 1.4002189137611553e-06,
      "loss": 0.81270373,
      "num_input_tokens_seen": 109169765,
      "step": 5065,
      "time_per_iteration": 2.532048225402832
    },
    {
      "auxiliary_loss_clip": 0.01153278,
      "auxiliary_loss_mlp": 0.01033108,
      "balance_loss_clip": 1.04774785,
      "balance_loss_mlp": 1.02531528,
      "epoch": 0.6091504839776348,
      "flos": 23987358639360.0,
      "grad_norm": 2.8675355436094887,
      "language_loss": 0.69467521,
      "learning_rate": 1.3994758408730901e-06,
      "loss": 0.71653914,
      "num_input_tokens_seen": 109188950,
      "step": 5066,
      "time_per_iteration": 2.525874614715576
    },
    {
      "auxiliary_loss_clip": 0.01143515,
      "auxiliary_loss_mlp": 0.01026742,
      "balance_loss_clip": 1.05069149,
      "balance_loss_mlp": 1.01947665,
      "epoch": 0.6092707268682739,
      "flos": 29643037666560.0,
      "grad_norm": 1.8470910208855524,
      "language_loss": 0.76441479,
      "learning_rate": 1.3987328590840629e-06,
      "loss": 0.78611743,
      "num_input_tokens_seen": 109209895,
      "step": 5067,
      "time_per_iteration": 2.63199782371521
    },
    {
      "auxiliary_loss_clip": 0.01150765,
      "auxiliary_loss_mlp": 0.01025956,
      "balance_loss_clip": 1.04621947,
      "balance_loss_mlp": 1.0185864,
      "epoch": 0.609390969758913,
      "flos": 24024957200640.0,
      "grad_norm": 1.721551124570284,
      "language_loss": 0.85964131,
      "learning_rate": 1.397989968506783e-06,
      "loss": 0.88140857,
      "num_input_tokens_seen": 109228905,
      "step": 5068,
      "time_per_iteration": 2.5091867446899414
    },
    {
      "auxiliary_loss_clip": 0.01175781,
      "auxiliary_loss_mlp": 0.01030597,
      "balance_loss_clip": 1.05303454,
      "balance_loss_mlp": 1.02341473,
      "epoch": 0.6095112126495521,
      "flos": 11102143288320.0,
      "grad_norm": 2.109023382822166,
      "language_loss": 0.72671509,
      "learning_rate": 1.3972471692539458e-06,
      "loss": 0.74877894,
      "num_input_tokens_seen": 109243620,
      "step": 5069,
      "time_per_iteration": 2.4198598861694336
    },
    {
      "auxiliary_loss_clip": 0.01139239,
      "auxiliary_loss_mlp": 0.01030885,
      "balance_loss_clip": 1.04768944,
      "balance_loss_mlp": 1.02317774,
      "epoch": 0.6096314555401912,
      "flos": 17265491187840.0,
      "grad_norm": 3.4164571385206086,
      "language_loss": 0.75351608,
      "learning_rate": 1.3965044614382348e-06,
      "loss": 0.77521729,
      "num_input_tokens_seen": 109259070,
      "step": 5070,
      "time_per_iteration": 2.4764938354492188
    },
    {
      "auxiliary_loss_clip": 0.01173107,
      "auxiliary_loss_mlp": 0.01027738,
      "balance_loss_clip": 1.05251002,
      "balance_loss_mlp": 1.01969767,
      "epoch": 0.6097516984308303,
      "flos": 21645910679040.0,
      "grad_norm": 2.487989062878312,
      "language_loss": 0.75452071,
      "learning_rate": 1.3957618451723162e-06,
      "loss": 0.77652919,
      "num_input_tokens_seen": 109275100,
      "step": 5071,
      "time_per_iteration": 2.442556619644165
    },
    {
      "auxiliary_loss_clip": 0.01142018,
      "auxiliary_loss_mlp": 0.01032854,
      "balance_loss_clip": 1.04707325,
      "balance_loss_mlp": 1.02551961,
      "epoch": 0.6098719413214694,
      "flos": 27199208966400.0,
      "grad_norm": 1.8373054300509908,
      "language_loss": 0.7177304,
      "learning_rate": 1.3950193205688457e-06,
      "loss": 0.73947912,
      "num_input_tokens_seen": 109294825,
      "step": 5072,
      "time_per_iteration": 3.3874645233154297
    },
    {
      "auxiliary_loss_clip": 0.01138344,
      "auxiliary_loss_mlp": 0.0103031,
      "balance_loss_clip": 1.04713941,
      "balance_loss_mlp": 1.02354479,
      "epoch": 0.6099921842121084,
      "flos": 20412954385920.0,
      "grad_norm": 1.7461668814778766,
      "language_loss": 0.83456469,
      "learning_rate": 1.3942768877404627e-06,
      "loss": 0.85625124,
      "num_input_tokens_seen": 109313790,
      "step": 5073,
      "time_per_iteration": 2.497920513153076
    },
    {
      "auxiliary_loss_clip": 0.0116795,
      "auxiliary_loss_mlp": 0.01027124,
      "balance_loss_clip": 1.04798901,
      "balance_loss_mlp": 1.01939309,
      "epoch": 0.6101124271027476,
      "flos": 23366139897600.0,
      "grad_norm": 1.9610123389150957,
      "language_loss": 0.73571491,
      "learning_rate": 1.393534546799795e-06,
      "loss": 0.75766563,
      "num_input_tokens_seen": 109333490,
      "step": 5074,
      "time_per_iteration": 2.49157452583313
    },
    {
      "auxiliary_loss_clip": 0.01134555,
      "auxiliary_loss_mlp": 0.01027311,
      "balance_loss_clip": 1.04720521,
      "balance_loss_mlp": 1.01961327,
      "epoch": 0.6102326699933867,
      "flos": 26687840993280.0,
      "grad_norm": 1.5186531643137702,
      "language_loss": 0.67445815,
      "learning_rate": 1.3927922978594536e-06,
      "loss": 0.69607681,
      "num_input_tokens_seen": 109354575,
      "step": 5075,
      "time_per_iteration": 2.561363935470581
    },
    {
      "auxiliary_loss_clip": 0.01053269,
      "auxiliary_loss_mlp": 0.01003398,
      "balance_loss_clip": 1.01615739,
      "balance_loss_mlp": 1.00221777,
      "epoch": 0.6103529128840257,
      "flos": 60644612551680.0,
      "grad_norm": 0.768844702105615,
      "language_loss": 0.57489496,
      "learning_rate": 1.3920501410320387e-06,
      "loss": 0.59546161,
      "num_input_tokens_seen": 109410690,
      "step": 5076,
      "time_per_iteration": 3.041224479675293
    },
    {
      "auxiliary_loss_clip": 0.0114142,
      "auxiliary_loss_mlp": 0.01026004,
      "balance_loss_clip": 1.045928,
      "balance_loss_mlp": 1.01832736,
      "epoch": 0.6104731557746649,
      "flos": 19021307806080.0,
      "grad_norm": 2.1129162305171776,
      "language_loss": 0.76131272,
      "learning_rate": 1.3913080764301333e-06,
      "loss": 0.78298688,
      "num_input_tokens_seen": 109427650,
      "step": 5077,
      "time_per_iteration": 2.508819818496704
    },
    {
      "auxiliary_loss_clip": 0.01122067,
      "auxiliary_loss_mlp": 0.01031384,
      "balance_loss_clip": 1.04278874,
      "balance_loss_mlp": 1.02432382,
      "epoch": 0.6105933986653039,
      "flos": 23366894083200.0,
      "grad_norm": 1.7692549118955077,
      "language_loss": 0.71272635,
      "learning_rate": 1.3905661041663085e-06,
      "loss": 0.73426092,
      "num_input_tokens_seen": 109448835,
      "step": 5078,
      "time_per_iteration": 3.3891961574554443
    },
    {
      "auxiliary_loss_clip": 0.01156254,
      "auxiliary_loss_mlp": 0.01030824,
      "balance_loss_clip": 1.04969811,
      "balance_loss_mlp": 1.02197862,
      "epoch": 0.610713641555943,
      "flos": 34637565006720.0,
      "grad_norm": 1.9947400898472356,
      "language_loss": 0.65335715,
      "learning_rate": 1.389824224353122e-06,
      "loss": 0.67522788,
      "num_input_tokens_seen": 109470425,
      "step": 5079,
      "time_per_iteration": 2.609548330307007
    },
    {
      "auxiliary_loss_clip": 0.01154684,
      "auxiliary_loss_mlp": 0.0102778,
      "balance_loss_clip": 1.04941726,
      "balance_loss_mlp": 1.02053213,
      "epoch": 0.610833884446582,
      "flos": 26646471504000.0,
      "grad_norm": 1.4812910735915281,
      "language_loss": 0.76813293,
      "learning_rate": 1.389082437103115e-06,
      "loss": 0.78995752,
      "num_input_tokens_seen": 109489695,
      "step": 5080,
      "time_per_iteration": 2.5400121212005615
    },
    {
      "auxiliary_loss_clip": 0.01125895,
      "auxiliary_loss_mlp": 0.01030136,
      "balance_loss_clip": 1.04383016,
      "balance_loss_mlp": 1.02219057,
      "epoch": 0.6109541273372212,
      "flos": 21215126868480.0,
      "grad_norm": 1.7562643721522815,
      "language_loss": 0.77691829,
      "learning_rate": 1.3883407425288172e-06,
      "loss": 0.7984786,
      "num_input_tokens_seen": 109510030,
      "step": 5081,
      "time_per_iteration": 3.431467056274414
    },
    {
      "auxiliary_loss_clip": 0.01138565,
      "auxiliary_loss_mlp": 0.01030433,
      "balance_loss_clip": 1.04603624,
      "balance_loss_mlp": 1.02266622,
      "epoch": 0.6110743702278603,
      "flos": 20084084438400.0,
      "grad_norm": 2.4662909038627836,
      "language_loss": 0.79948795,
      "learning_rate": 1.3875991407427417e-06,
      "loss": 0.8211779,
      "num_input_tokens_seen": 109528255,
      "step": 5082,
      "time_per_iteration": 3.2683560848236084
    },
    {
      "auxiliary_loss_clip": 0.01035737,
      "auxiliary_loss_mlp": 0.01000541,
      "balance_loss_clip": 1.01433074,
      "balance_loss_mlp": 0.99934906,
      "epoch": 0.6111946131184993,
      "flos": 68302957438080.0,
      "grad_norm": 0.7748458364660881,
      "language_loss": 0.58270597,
      "learning_rate": 1.38685763185739e-06,
      "loss": 0.60306877,
      "num_input_tokens_seen": 109581915,
      "step": 5083,
      "time_per_iteration": 3.173628568649292
    },
    {
      "auxiliary_loss_clip": 0.01167222,
      "auxiliary_loss_mlp": 0.01032243,
      "balance_loss_clip": 1.04969764,
      "balance_loss_mlp": 1.02467048,
      "epoch": 0.6113148560091385,
      "flos": 19937676602880.0,
      "grad_norm": 2.3920738268788853,
      "language_loss": 0.67444587,
      "learning_rate": 1.3861162159852476e-06,
      "loss": 0.69644052,
      "num_input_tokens_seen": 109600050,
      "step": 5084,
      "time_per_iteration": 2.4421184062957764
    },
    {
      "auxiliary_loss_clip": 0.01145765,
      "auxiliary_loss_mlp": 0.01028204,
      "balance_loss_clip": 1.0475322,
      "balance_loss_mlp": 1.02092934,
      "epoch": 0.6114350988997775,
      "flos": 23731854220800.0,
      "grad_norm": 1.7998744903246418,
      "language_loss": 0.79485965,
      "learning_rate": 1.3853748932387875e-06,
      "loss": 0.81659937,
      "num_input_tokens_seen": 109620690,
      "step": 5085,
      "time_per_iteration": 2.547948122024536
    },
    {
      "auxiliary_loss_clip": 0.01131295,
      "auxiliary_loss_mlp": 0.01027899,
      "balance_loss_clip": 1.04632354,
      "balance_loss_mlp": 1.02004313,
      "epoch": 0.6115553417904166,
      "flos": 24023700224640.0,
      "grad_norm": 2.328699132984908,
      "language_loss": 0.75204015,
      "learning_rate": 1.3846336637304671e-06,
      "loss": 0.77363205,
      "num_input_tokens_seen": 109638960,
      "step": 5086,
      "time_per_iteration": 2.5320703983306885
    },
    {
      "auxiliary_loss_clip": 0.01130793,
      "auxiliary_loss_mlp": 0.01024551,
      "balance_loss_clip": 1.04467821,
      "balance_loss_mlp": 1.01725221,
      "epoch": 0.6116755846810558,
      "flos": 23733542160000.0,
      "grad_norm": 2.187169061554951,
      "language_loss": 0.8331092,
      "learning_rate": 1.3838925275727316e-06,
      "loss": 0.85466266,
      "num_input_tokens_seen": 109659700,
      "step": 5087,
      "time_per_iteration": 2.540968418121338
    },
    {
      "auxiliary_loss_clip": 0.011714,
      "auxiliary_loss_mlp": 0.01025289,
      "balance_loss_clip": 1.05166662,
      "balance_loss_mlp": 1.01795816,
      "epoch": 0.6117958275716948,
      "flos": 18661626967680.0,
      "grad_norm": 1.7073873585641228,
      "language_loss": 0.78776324,
      "learning_rate": 1.3831514848780089e-06,
      "loss": 0.80973011,
      "num_input_tokens_seen": 109679275,
      "step": 5088,
      "time_per_iteration": 2.4484424591064453
    },
    {
      "auxiliary_loss_clip": 0.0114915,
      "auxiliary_loss_mlp": 0.01026421,
      "balance_loss_clip": 1.04802632,
      "balance_loss_mlp": 1.01931572,
      "epoch": 0.6119160704623339,
      "flos": 16471183783680.0,
      "grad_norm": 2.2447062906859205,
      "language_loss": 0.92023253,
      "learning_rate": 1.3824105357587152e-06,
      "loss": 0.94198823,
      "num_input_tokens_seen": 109696380,
      "step": 5089,
      "time_per_iteration": 2.446146011352539
    },
    {
      "auxiliary_loss_clip": 0.01138244,
      "auxiliary_loss_mlp": 0.01025289,
      "balance_loss_clip": 1.04691768,
      "balance_loss_mlp": 1.01852679,
      "epoch": 0.612036313352973,
      "flos": 23915465568000.0,
      "grad_norm": 1.4901028536063987,
      "language_loss": 0.82544166,
      "learning_rate": 1.381669680327253e-06,
      "loss": 0.84707701,
      "num_input_tokens_seen": 109718060,
      "step": 5090,
      "time_per_iteration": 2.5735507011413574
    },
    {
      "auxiliary_loss_clip": 0.01135288,
      "auxiliary_loss_mlp": 0.01027187,
      "balance_loss_clip": 1.0489527,
      "balance_loss_mlp": 1.01972747,
      "epoch": 0.6121565562436121,
      "flos": 26974766833920.0,
      "grad_norm": 1.9017451731346935,
      "language_loss": 0.70652163,
      "learning_rate": 1.380928918696008e-06,
      "loss": 0.72814631,
      "num_input_tokens_seen": 109736830,
      "step": 5091,
      "time_per_iteration": 2.5679984092712402
    },
    {
      "auxiliary_loss_clip": 0.0115316,
      "auxiliary_loss_mlp": 0.01025495,
      "balance_loss_clip": 1.04807949,
      "balance_loss_mlp": 1.0186466,
      "epoch": 0.6122767991342511,
      "flos": 15668867646720.0,
      "grad_norm": 2.4413743526894875,
      "language_loss": 0.71899438,
      "learning_rate": 1.3801882509773548e-06,
      "loss": 0.74078089,
      "num_input_tokens_seen": 109754690,
      "step": 5092,
      "time_per_iteration": 2.4641082286834717
    },
    {
      "auxiliary_loss_clip": 0.01148473,
      "auxiliary_loss_mlp": 0.01027659,
      "balance_loss_clip": 1.0458957,
      "balance_loss_mlp": 1.01994038,
      "epoch": 0.6123970420248903,
      "flos": 27964321591680.0,
      "grad_norm": 1.868708455105071,
      "language_loss": 0.81670088,
      "learning_rate": 1.3794476772836503e-06,
      "loss": 0.83846223,
      "num_input_tokens_seen": 109775790,
      "step": 5093,
      "time_per_iteration": 2.5375571250915527
    },
    {
      "auxiliary_loss_clip": 0.01119633,
      "auxiliary_loss_mlp": 0.01024019,
      "balance_loss_clip": 1.0457257,
      "balance_loss_mlp": 1.01601434,
      "epoch": 0.6125172849155294,
      "flos": 21468727866240.0,
      "grad_norm": 1.5319802667187998,
      "language_loss": 0.84635758,
      "learning_rate": 1.3787071977272402e-06,
      "loss": 0.86779404,
      "num_input_tokens_seen": 109795050,
      "step": 5094,
      "time_per_iteration": 2.5589678287506104
    },
    {
      "auxiliary_loss_clip": 0.0111084,
      "auxiliary_loss_mlp": 0.01028323,
      "balance_loss_clip": 1.04736161,
      "balance_loss_mlp": 1.02093482,
      "epoch": 0.6126375278061684,
      "flos": 16248321849600.0,
      "grad_norm": 3.1281301667762227,
      "language_loss": 0.72140563,
      "learning_rate": 1.3779668124204535e-06,
      "loss": 0.74279726,
      "num_input_tokens_seen": 109811465,
      "step": 5095,
      "time_per_iteration": 2.542752504348755
    },
    {
      "auxiliary_loss_clip": 0.01136961,
      "auxiliary_loss_mlp": 0.0102566,
      "balance_loss_clip": 1.04880309,
      "balance_loss_mlp": 1.01869869,
      "epoch": 0.6127577706968076,
      "flos": 20448865008000.0,
      "grad_norm": 1.5883424930327767,
      "language_loss": 0.80849802,
      "learning_rate": 1.3772265214756074e-06,
      "loss": 0.83012426,
      "num_input_tokens_seen": 109831225,
      "step": 5096,
      "time_per_iteration": 2.513594150543213
    },
    {
      "auxiliary_loss_clip": 0.01159231,
      "auxiliary_loss_mlp": 0.01029037,
      "balance_loss_clip": 1.04833245,
      "balance_loss_mlp": 1.0215683,
      "epoch": 0.6128780135874466,
      "flos": 18260397072000.0,
      "grad_norm": 1.70103814607756,
      "language_loss": 0.7483443,
      "learning_rate": 1.3764863250050025e-06,
      "loss": 0.77022696,
      "num_input_tokens_seen": 109849465,
      "step": 5097,
      "time_per_iteration": 2.4705862998962402
    },
    {
      "auxiliary_loss_clip": 0.01128483,
      "auxiliary_loss_mlp": 0.01022984,
      "balance_loss_clip": 1.04585886,
      "balance_loss_mlp": 1.01609039,
      "epoch": 0.6129982564780857,
      "flos": 24937088192640.0,
      "grad_norm": 1.7405721520147095,
      "language_loss": 0.80436003,
      "learning_rate": 1.3757462231209272e-06,
      "loss": 0.82587469,
      "num_input_tokens_seen": 109869770,
      "step": 5098,
      "time_per_iteration": 3.432140588760376
    },
    {
      "auxiliary_loss_clip": 0.01138014,
      "auxiliary_loss_mlp": 0.01038384,
      "balance_loss_clip": 1.04875433,
      "balance_loss_mlp": 1.03068316,
      "epoch": 0.6131184993687249,
      "flos": 22492038430080.0,
      "grad_norm": 1.8462632323286408,
      "language_loss": 0.88827646,
      "learning_rate": 1.3750062159356525e-06,
      "loss": 0.91004044,
      "num_input_tokens_seen": 109889120,
      "step": 5099,
      "time_per_iteration": 2.537116050720215
    },
    {
      "auxiliary_loss_clip": 0.01119972,
      "auxiliary_loss_mlp": 0.0102801,
      "balance_loss_clip": 1.0457232,
      "balance_loss_mlp": 1.02123916,
      "epoch": 0.6132387422593639,
      "flos": 15885839750400.0,
      "grad_norm": 1.894764994743442,
      "language_loss": 0.83217216,
      "learning_rate": 1.3742663035614382e-06,
      "loss": 0.853652,
      "num_input_tokens_seen": 109906490,
      "step": 5100,
      "time_per_iteration": 2.5497753620147705
    },
    {
      "auxiliary_loss_clip": 0.01171924,
      "auxiliary_loss_mlp": 0.01033005,
      "balance_loss_clip": 1.05189514,
      "balance_loss_mlp": 1.02551568,
      "epoch": 0.613358985150003,
      "flos": 25411539962880.0,
      "grad_norm": 1.6052755787774182,
      "language_loss": 0.79927951,
      "learning_rate": 1.3735264861105283e-06,
      "loss": 0.82132882,
      "num_input_tokens_seen": 109927130,
      "step": 5101,
      "time_per_iteration": 2.4985053539276123
    },
    {
      "auxiliary_loss_clip": 0.01128078,
      "auxiliary_loss_mlp": 0.01024911,
      "balance_loss_clip": 1.04623282,
      "balance_loss_mlp": 1.01772308,
      "epoch": 0.6134792280406421,
      "flos": 21361283308800.0,
      "grad_norm": 1.9466554206314726,
      "language_loss": 0.78397942,
      "learning_rate": 1.372786763695152e-06,
      "loss": 0.80550921,
      "num_input_tokens_seen": 109945890,
      "step": 5102,
      "time_per_iteration": 2.556319236755371
    },
    {
      "auxiliary_loss_clip": 0.01156306,
      "auxiliary_loss_mlp": 0.01032631,
      "balance_loss_clip": 1.04867911,
      "balance_loss_mlp": 1.02462018,
      "epoch": 0.6135994709312812,
      "flos": 21211248199680.0,
      "grad_norm": 2.085458628864434,
      "language_loss": 0.77113438,
      "learning_rate": 1.3720471364275257e-06,
      "loss": 0.79302371,
      "num_input_tokens_seen": 109965535,
      "step": 5103,
      "time_per_iteration": 2.491875648498535
    },
    {
      "auxiliary_loss_clip": 0.01121856,
      "auxiliary_loss_mlp": 0.00761488,
      "balance_loss_clip": 1.0432353,
      "balance_loss_mlp": 1.00013697,
      "epoch": 0.6137197138219203,
      "flos": 14794047907200.0,
      "grad_norm": 1.900040135435428,
      "language_loss": 0.78064907,
      "learning_rate": 1.3713076044198486e-06,
      "loss": 0.79948252,
      "num_input_tokens_seen": 109982345,
      "step": 5104,
      "time_per_iteration": 3.3065407276153564
    },
    {
      "auxiliary_loss_clip": 0.01137691,
      "auxiliary_loss_mlp": 0.01026121,
      "balance_loss_clip": 1.04720807,
      "balance_loss_mlp": 1.0183723,
      "epoch": 0.6138399567125594,
      "flos": 20084515401600.0,
      "grad_norm": 8.833371000956078,
      "language_loss": 0.81079447,
      "learning_rate": 1.3705681677843086e-06,
      "loss": 0.83243263,
      "num_input_tokens_seen": 110000940,
      "step": 5105,
      "time_per_iteration": 2.513887405395508
    },
    {
      "auxiliary_loss_clip": 0.0106434,
      "auxiliary_loss_mlp": 0.01001597,
      "balance_loss_clip": 1.01523805,
      "balance_loss_mlp": 1.00045896,
      "epoch": 0.6139601996031985,
      "flos": 60123838193280.0,
      "grad_norm": 0.7720923111961804,
      "language_loss": 0.60613787,
      "learning_rate": 1.3698288266330768e-06,
      "loss": 0.62679732,
      "num_input_tokens_seen": 110061565,
      "step": 5106,
      "time_per_iteration": 3.1550843715667725
    },
    {
      "auxiliary_loss_clip": 0.01139938,
      "auxiliary_loss_mlp": 0.0102285,
      "balance_loss_clip": 1.05245185,
      "balance_loss_mlp": 1.01556969,
      "epoch": 0.6140804424938375,
      "flos": 23586703361280.0,
      "grad_norm": 2.5320270139646457,
      "language_loss": 0.72414708,
      "learning_rate": 1.3690895810783113e-06,
      "loss": 0.74577492,
      "num_input_tokens_seen": 110080360,
      "step": 5107,
      "time_per_iteration": 3.3648622035980225
    },
    {
      "auxiliary_loss_clip": 0.01103276,
      "auxiliary_loss_mlp": 0.00761392,
      "balance_loss_clip": 1.04260683,
      "balance_loss_mlp": 1.00017059,
      "epoch": 0.6142006853844767,
      "flos": 21398199511680.0,
      "grad_norm": 2.2052196361125,
      "language_loss": 0.71117556,
      "learning_rate": 1.3683504312321543e-06,
      "loss": 0.72982216,
      "num_input_tokens_seen": 110100695,
      "step": 5108,
      "time_per_iteration": 3.35969614982605
    },
    {
      "auxiliary_loss_clip": 0.01158673,
      "auxiliary_loss_mlp": 0.01027079,
      "balance_loss_clip": 1.04862237,
      "balance_loss_mlp": 1.01978612,
      "epoch": 0.6143209282751158,
      "flos": 12057367622400.0,
      "grad_norm": 1.9012011811642793,
      "language_loss": 0.80142009,
      "learning_rate": 1.3676113772067355e-06,
      "loss": 0.82327765,
      "num_input_tokens_seen": 110117750,
      "step": 5109,
      "time_per_iteration": 2.4974019527435303
    },
    {
      "auxiliary_loss_clip": 0.01120972,
      "auxiliary_loss_mlp": 0.01024398,
      "balance_loss_clip": 1.04603148,
      "balance_loss_mlp": 1.01721001,
      "epoch": 0.6144411711657548,
      "flos": 25082274965760.0,
      "grad_norm": 2.184381966241592,
      "language_loss": 0.72748137,
      "learning_rate": 1.3668724191141671e-06,
      "loss": 0.7489351,
      "num_input_tokens_seen": 110137020,
      "step": 5110,
      "time_per_iteration": 2.618596315383911
    },
    {
      "auxiliary_loss_clip": 0.01123361,
      "auxiliary_loss_mlp": 0.01030558,
      "balance_loss_clip": 1.04983997,
      "balance_loss_mlp": 1.02372468,
      "epoch": 0.6145614140563939,
      "flos": 20114069316480.0,
      "grad_norm": 2.236350660298103,
      "language_loss": 0.66543901,
      "learning_rate": 1.3661335570665493e-06,
      "loss": 0.68697822,
      "num_input_tokens_seen": 110154930,
      "step": 5111,
      "time_per_iteration": 2.5453264713287354
    },
    {
      "auxiliary_loss_clip": 0.01144514,
      "auxiliary_loss_mlp": 0.01029042,
      "balance_loss_clip": 1.04988217,
      "balance_loss_mlp": 1.02177,
      "epoch": 0.614681656947033,
      "flos": 16800376953600.0,
      "grad_norm": 2.454641646748015,
      "language_loss": 0.69908071,
      "learning_rate": 1.3653947911759676e-06,
      "loss": 0.72081631,
      "num_input_tokens_seen": 110172480,
      "step": 5112,
      "time_per_iteration": 2.4968929290771484
    },
    {
      "auxiliary_loss_clip": 0.01108398,
      "auxiliary_loss_mlp": 0.01028799,
      "balance_loss_clip": 1.04445302,
      "balance_loss_mlp": 1.02101469,
      "epoch": 0.6148018998376721,
      "flos": 38801587011840.0,
      "grad_norm": 1.7818902040001832,
      "language_loss": 0.743451,
      "learning_rate": 1.3646561215544904e-06,
      "loss": 0.76482296,
      "num_input_tokens_seen": 110197120,
      "step": 5113,
      "time_per_iteration": 2.744920015335083
    },
    {
      "auxiliary_loss_clip": 0.01155308,
      "auxiliary_loss_mlp": 0.01027082,
      "balance_loss_clip": 1.05056739,
      "balance_loss_mlp": 1.01963186,
      "epoch": 0.6149221427283111,
      "flos": 23327032965120.0,
      "grad_norm": 1.973460872023696,
      "language_loss": 0.79144919,
      "learning_rate": 1.363917548314176e-06,
      "loss": 0.81327307,
      "num_input_tokens_seen": 110216385,
      "step": 5114,
      "time_per_iteration": 2.4940006732940674
    },
    {
      "auxiliary_loss_clip": 0.01160742,
      "auxiliary_loss_mlp": 0.01031521,
      "balance_loss_clip": 1.04994845,
      "balance_loss_mlp": 1.02404928,
      "epoch": 0.6150423856189503,
      "flos": 22379494141440.0,
      "grad_norm": 1.7971877698039849,
      "language_loss": 0.73122835,
      "learning_rate": 1.3631790715670626e-06,
      "loss": 0.75315094,
      "num_input_tokens_seen": 110234790,
      "step": 5115,
      "time_per_iteration": 2.499669075012207
    },
    {
      "auxiliary_loss_clip": 0.01079876,
      "auxiliary_loss_mlp": 0.01028371,
      "balance_loss_clip": 1.04505038,
      "balance_loss_mlp": 1.02199054,
      "epoch": 0.6151626285095894,
      "flos": 18692078722560.0,
      "grad_norm": 1.9748437391176514,
      "language_loss": 0.85479105,
      "learning_rate": 1.3624406914251783e-06,
      "loss": 0.87587357,
      "num_input_tokens_seen": 110251910,
      "step": 5116,
      "time_per_iteration": 2.7628390789031982
    },
    {
      "auxiliary_loss_clip": 0.01155609,
      "auxiliary_loss_mlp": 0.01029689,
      "balance_loss_clip": 1.04831481,
      "balance_loss_mlp": 1.02309346,
      "epoch": 0.6152828714002284,
      "flos": 15851688894720.0,
      "grad_norm": 1.9363700028114572,
      "language_loss": 0.87913376,
      "learning_rate": 1.3617024080005335e-06,
      "loss": 0.90098679,
      "num_input_tokens_seen": 110268810,
      "step": 5117,
      "time_per_iteration": 2.879237174987793
    },
    {
      "auxiliary_loss_clip": 0.01141634,
      "auxiliary_loss_mlp": 0.00761521,
      "balance_loss_clip": 1.04713154,
      "balance_loss_mlp": 1.00014758,
      "epoch": 0.6154031142908676,
      "flos": 24869792062080.0,
      "grad_norm": 2.3746447507404786,
      "language_loss": 0.74315381,
      "learning_rate": 1.3609642214051266e-06,
      "loss": 0.76218534,
      "num_input_tokens_seen": 110293035,
      "step": 5118,
      "time_per_iteration": 2.669893264770508
    },
    {
      "auxiliary_loss_clip": 0.0113348,
      "auxiliary_loss_mlp": 0.01028135,
      "balance_loss_clip": 1.04695034,
      "balance_loss_mlp": 1.02086043,
      "epoch": 0.6155233571815066,
      "flos": 19244744357760.0,
      "grad_norm": 1.7206600486248265,
      "language_loss": 0.66299653,
      "learning_rate": 1.3602261317509385e-06,
      "loss": 0.68461269,
      "num_input_tokens_seen": 110309695,
      "step": 5119,
      "time_per_iteration": 2.5378570556640625
    },
    {
      "auxiliary_loss_clip": 0.01156219,
      "auxiliary_loss_mlp": 0.01026658,
      "balance_loss_clip": 1.04955721,
      "balance_loss_mlp": 1.01865315,
      "epoch": 0.6156436000721457,
      "flos": 18770077105920.0,
      "grad_norm": 2.2830852463141955,
      "language_loss": 0.82961619,
      "learning_rate": 1.3594881391499387e-06,
      "loss": 0.85144508,
      "num_input_tokens_seen": 110328610,
      "step": 5120,
      "time_per_iteration": 2.4864237308502197
    },
    {
      "auxiliary_loss_clip": 0.01141937,
      "auxiliary_loss_mlp": 0.01032121,
      "balance_loss_clip": 1.04813349,
      "balance_loss_mlp": 1.0250814,
      "epoch": 0.6157638429627849,
      "flos": 18041198325120.0,
      "grad_norm": 1.6921306289473252,
      "language_loss": 0.79324675,
      "learning_rate": 1.3587502437140778e-06,
      "loss": 0.81498736,
      "num_input_tokens_seen": 110346775,
      "step": 5121,
      "time_per_iteration": 2.5101499557495117
    },
    {
      "auxiliary_loss_clip": 0.01143971,
      "auxiliary_loss_mlp": 0.01026471,
      "balance_loss_clip": 1.0463047,
      "balance_loss_mlp": 1.01895475,
      "epoch": 0.6158840858534239,
      "flos": 25556726736000.0,
      "grad_norm": 1.9647656160343652,
      "language_loss": 0.84895867,
      "learning_rate": 1.3580124455552952e-06,
      "loss": 0.87066311,
      "num_input_tokens_seen": 110366140,
      "step": 5122,
      "time_per_iteration": 2.56144118309021
    },
    {
      "auxiliary_loss_clip": 0.01156729,
      "auxiliary_loss_mlp": 0.00761023,
      "balance_loss_clip": 1.05109012,
      "balance_loss_mlp": 1.00017369,
      "epoch": 0.616004328744063,
      "flos": 24640788902400.0,
      "grad_norm": 1.8589675293419443,
      "language_loss": 0.87505305,
      "learning_rate": 1.3572747447855148e-06,
      "loss": 0.8942306,
      "num_input_tokens_seen": 110386550,
      "step": 5123,
      "time_per_iteration": 2.5181820392608643
    },
    {
      "auxiliary_loss_clip": 0.01171773,
      "auxiliary_loss_mlp": 0.01034021,
      "balance_loss_clip": 1.05178034,
      "balance_loss_mlp": 1.02617419,
      "epoch": 0.6161245716347021,
      "flos": 21689686379520.0,
      "grad_norm": 1.8578992538636085,
      "language_loss": 0.6940375,
      "learning_rate": 1.356537141516644e-06,
      "loss": 0.71609545,
      "num_input_tokens_seen": 110403970,
      "step": 5124,
      "time_per_iteration": 3.2836720943450928
    },
    {
      "auxiliary_loss_clip": 0.01156229,
      "auxiliary_loss_mlp": 0.01029344,
      "balance_loss_clip": 1.05226302,
      "balance_loss_mlp": 1.02216816,
      "epoch": 0.6162448145253412,
      "flos": 35189225061120.0,
      "grad_norm": 1.7464976920291766,
      "language_loss": 0.61609071,
      "learning_rate": 1.3557996358605775e-06,
      "loss": 0.63794643,
      "num_input_tokens_seen": 110423890,
      "step": 5125,
      "time_per_iteration": 2.604205369949341
    },
    {
      "auxiliary_loss_clip": 0.01156832,
      "auxiliary_loss_mlp": 0.01027107,
      "balance_loss_clip": 1.05020261,
      "balance_loss_mlp": 1.02052927,
      "epoch": 0.6163650574159802,
      "flos": 21615279356160.0,
      "grad_norm": 2.024759929836643,
      "language_loss": 0.70077312,
      "learning_rate": 1.3550622279291941e-06,
      "loss": 0.7226125,
      "num_input_tokens_seen": 110442035,
      "step": 5126,
      "time_per_iteration": 2.4953577518463135
    },
    {
      "auxiliary_loss_clip": 0.01102677,
      "auxiliary_loss_mlp": 0.01031381,
      "balance_loss_clip": 1.0411247,
      "balance_loss_mlp": 1.02416015,
      "epoch": 0.6164853003066194,
      "flos": 24572163968640.0,
      "grad_norm": 1.349276355731954,
      "language_loss": 0.8323983,
      "learning_rate": 1.354324917834358e-06,
      "loss": 0.8537389,
      "num_input_tokens_seen": 110463280,
      "step": 5127,
      "time_per_iteration": 2.6244421005249023
    },
    {
      "auxiliary_loss_clip": 0.01095351,
      "auxiliary_loss_mlp": 0.00760731,
      "balance_loss_clip": 1.04074001,
      "balance_loss_mlp": 1.00014448,
      "epoch": 0.6166055431972585,
      "flos": 21835986474240.0,
      "grad_norm": 1.6489113312843469,
      "language_loss": 0.76688659,
      "learning_rate": 1.353587705687918e-06,
      "loss": 0.78544736,
      "num_input_tokens_seen": 110481455,
      "step": 5128,
      "time_per_iteration": 2.639486074447632
    },
    {
      "auxiliary_loss_clip": 0.01149973,
      "auxiliary_loss_mlp": 0.01028236,
      "balance_loss_clip": 1.05194712,
      "balance_loss_mlp": 1.02047491,
      "epoch": 0.6167257860878975,
      "flos": 17785262943360.0,
      "grad_norm": 2.7266679249330115,
      "language_loss": 0.72912669,
      "learning_rate": 1.3528505916017096e-06,
      "loss": 0.75090879,
      "num_input_tokens_seen": 110499155,
      "step": 5129,
      "time_per_iteration": 2.4892218112945557
    },
    {
      "auxiliary_loss_clip": 0.0115537,
      "auxiliary_loss_mlp": 0.01030674,
      "balance_loss_clip": 1.04825783,
      "balance_loss_mlp": 1.02340794,
      "epoch": 0.6168460289785367,
      "flos": 23214811898880.0,
      "grad_norm": 2.2556958409390844,
      "language_loss": 0.88579953,
      "learning_rate": 1.3521135756875514e-06,
      "loss": 0.90765995,
      "num_input_tokens_seen": 110515470,
      "step": 5130,
      "time_per_iteration": 3.2697927951812744
    },
    {
      "auxiliary_loss_clip": 0.01091419,
      "auxiliary_loss_mlp": 0.01029873,
      "balance_loss_clip": 1.04314935,
      "balance_loss_mlp": 1.02347171,
      "epoch": 0.6169662718691757,
      "flos": 26213281482240.0,
      "grad_norm": 1.473702870846425,
      "language_loss": 0.8590433,
      "learning_rate": 1.3513766580572496e-06,
      "loss": 0.88025618,
      "num_input_tokens_seen": 110538290,
      "step": 5131,
      "time_per_iteration": 2.6853346824645996
    },
    {
      "auxiliary_loss_clip": 0.01155136,
      "auxiliary_loss_mlp": 0.01032689,
      "balance_loss_clip": 1.04974079,
      "balance_loss_mlp": 1.02535152,
      "epoch": 0.6170865147598148,
      "flos": 19026120228480.0,
      "grad_norm": 2.2318019701687914,
      "language_loss": 0.77394843,
      "learning_rate": 1.3506398388225924e-06,
      "loss": 0.79582667,
      "num_input_tokens_seen": 110555610,
      "step": 5132,
      "time_per_iteration": 2.4916577339172363
    },
    {
      "auxiliary_loss_clip": 0.0116945,
      "auxiliary_loss_mlp": 0.01029731,
      "balance_loss_clip": 1.05154431,
      "balance_loss_mlp": 1.02271843,
      "epoch": 0.617206757650454,
      "flos": 18260361158400.0,
      "grad_norm": 1.6473480401905296,
      "language_loss": 0.71852016,
      "learning_rate": 1.349903118095355e-06,
      "loss": 0.74051201,
      "num_input_tokens_seen": 110574745,
      "step": 5133,
      "time_per_iteration": 3.6419479846954346
    },
    {
      "auxiliary_loss_clip": 0.01160027,
      "auxiliary_loss_mlp": 0.01025821,
      "balance_loss_clip": 1.05050302,
      "balance_loss_mlp": 1.01897502,
      "epoch": 0.617327000541093,
      "flos": 18186959715840.0,
      "grad_norm": 1.704792285248535,
      "language_loss": 0.73468184,
      "learning_rate": 1.349166495987298e-06,
      "loss": 0.75654042,
      "num_input_tokens_seen": 110593310,
      "step": 5134,
      "time_per_iteration": 3.1654062271118164
    },
    {
      "auxiliary_loss_clip": 0.01039519,
      "auxiliary_loss_mlp": 0.01004577,
      "balance_loss_clip": 1.01569462,
      "balance_loss_mlp": 1.00334346,
      "epoch": 0.6174472434317321,
      "flos": 61833796122240.0,
      "grad_norm": 0.8190713262003482,
      "language_loss": 0.60926819,
      "learning_rate": 1.348429972610166e-06,
      "loss": 0.62970912,
      "num_input_tokens_seen": 110657615,
      "step": 5135,
      "time_per_iteration": 3.1971275806427
    },
    {
      "auxiliary_loss_clip": 0.01017331,
      "auxiliary_loss_mlp": 0.01005008,
      "balance_loss_clip": 1.01587629,
      "balance_loss_mlp": 1.00360155,
      "epoch": 0.6175674863223712,
      "flos": 71230970494080.0,
      "grad_norm": 1.1309420185634191,
      "language_loss": 0.57801795,
      "learning_rate": 1.3476935480756897e-06,
      "loss": 0.59824133,
      "num_input_tokens_seen": 110714365,
      "step": 5136,
      "time_per_iteration": 3.046581268310547
    },
    {
      "auxiliary_loss_clip": 0.01119105,
      "auxiliary_loss_mlp": 0.0102817,
      "balance_loss_clip": 1.04420412,
      "balance_loss_mlp": 1.02076983,
      "epoch": 0.6176877292130103,
      "flos": 21835447770240.0,
      "grad_norm": 2.159644096005312,
      "language_loss": 0.75553846,
      "learning_rate": 1.346957222495583e-06,
      "loss": 0.77701128,
      "num_input_tokens_seen": 110732160,
      "step": 5137,
      "time_per_iteration": 2.6107821464538574
    },
    {
      "auxiliary_loss_clip": 0.01146568,
      "auxiliary_loss_mlp": 0.00761105,
      "balance_loss_clip": 1.05201781,
      "balance_loss_mlp": 1.00016606,
      "epoch": 0.6178079721036493,
      "flos": 17741738638080.0,
      "grad_norm": 2.487076116597684,
      "language_loss": 0.71079648,
      "learning_rate": 1.3462209959815466e-06,
      "loss": 0.72987324,
      "num_input_tokens_seen": 110746900,
      "step": 5138,
      "time_per_iteration": 2.4936232566833496
    },
    {
      "auxiliary_loss_clip": 0.0114354,
      "auxiliary_loss_mlp": 0.01025093,
      "balance_loss_clip": 1.04907668,
      "balance_loss_mlp": 1.01835763,
      "epoch": 0.6179282149942885,
      "flos": 22633131052800.0,
      "grad_norm": 2.0658169266479582,
      "language_loss": 0.7439509,
      "learning_rate": 1.345484868645265e-06,
      "loss": 0.76563722,
      "num_input_tokens_seen": 110765710,
      "step": 5139,
      "time_per_iteration": 2.5597472190856934
    },
    {
      "auxiliary_loss_clip": 0.01132761,
      "auxiliary_loss_mlp": 0.01027726,
      "balance_loss_clip": 1.0477761,
      "balance_loss_mlp": 1.02085352,
      "epoch": 0.6180484578849276,
      "flos": 22310330503680.0,
      "grad_norm": 2.7320806023250777,
      "language_loss": 0.78470039,
      "learning_rate": 1.3447488405984088e-06,
      "loss": 0.80630523,
      "num_input_tokens_seen": 110783970,
      "step": 5140,
      "time_per_iteration": 2.601616621017456
    },
    {
      "auxiliary_loss_clip": 0.0114236,
      "auxiliary_loss_mlp": 0.01030361,
      "balance_loss_clip": 1.05124903,
      "balance_loss_mlp": 1.02309775,
      "epoch": 0.6181687007755666,
      "flos": 35225458905600.0,
      "grad_norm": 2.1072495627136583,
      "language_loss": 0.69734251,
      "learning_rate": 1.3440129119526322e-06,
      "loss": 0.71906966,
      "num_input_tokens_seen": 110806395,
      "step": 5141,
      "time_per_iteration": 2.6859753131866455
    },
    {
      "auxiliary_loss_clip": 0.010656,
      "auxiliary_loss_mlp": 0.01002826,
      "balance_loss_clip": 1.01699114,
      "balance_loss_mlp": 1.00170505,
      "epoch": 0.6182889436662057,
      "flos": 61547370094080.0,
      "grad_norm": 0.7979084947118404,
      "language_loss": 0.51171196,
      "learning_rate": 1.3432770828195762e-06,
      "loss": 0.5323962,
      "num_input_tokens_seen": 110867380,
      "step": 5142,
      "time_per_iteration": 3.256830930709839
    },
    {
      "auxiliary_loss_clip": 0.01117193,
      "auxiliary_loss_mlp": 0.01033751,
      "balance_loss_clip": 1.04363632,
      "balance_loss_mlp": 1.02646184,
      "epoch": 0.6184091865568448,
      "flos": 19609991804160.0,
      "grad_norm": 2.224131915559719,
      "language_loss": 0.70293677,
      "learning_rate": 1.3425413533108635e-06,
      "loss": 0.72444624,
      "num_input_tokens_seen": 110885980,
      "step": 5143,
      "time_per_iteration": 2.5785207748413086
    },
    {
      "auxiliary_loss_clip": 0.01117583,
      "auxiliary_loss_mlp": 0.01033547,
      "balance_loss_clip": 1.04955065,
      "balance_loss_mlp": 1.02634668,
      "epoch": 0.6185294294474839,
      "flos": 23586882929280.0,
      "grad_norm": 3.0625458382975044,
      "language_loss": 0.70477664,
      "learning_rate": 1.341805723538105e-06,
      "loss": 0.72628796,
      "num_input_tokens_seen": 110906085,
      "step": 5144,
      "time_per_iteration": 2.6237895488739014
    },
    {
      "auxiliary_loss_clip": 0.01146921,
      "auxiliary_loss_mlp": 0.0103161,
      "balance_loss_clip": 1.04944217,
      "balance_loss_mlp": 1.02434146,
      "epoch": 0.618649672338123,
      "flos": 26762032535040.0,
      "grad_norm": 1.9601492941775835,
      "language_loss": 0.77373695,
      "learning_rate": 1.3410701936128948e-06,
      "loss": 0.79552227,
      "num_input_tokens_seen": 110928865,
      "step": 5145,
      "time_per_iteration": 2.703033924102783
    },
    {
      "auxiliary_loss_clip": 0.01156372,
      "auxiliary_loss_mlp": 0.01032008,
      "balance_loss_clip": 1.05088019,
      "balance_loss_mlp": 1.02471232,
      "epoch": 0.6187699152287621,
      "flos": 14456630522880.0,
      "grad_norm": 2.480518609282498,
      "language_loss": 0.84703076,
      "learning_rate": 1.340334763646812e-06,
      "loss": 0.86891448,
      "num_input_tokens_seen": 110943000,
      "step": 5146,
      "time_per_iteration": 2.509705066680908
    },
    {
      "auxiliary_loss_clip": 0.01173382,
      "auxiliary_loss_mlp": 0.0103409,
      "balance_loss_clip": 1.05239439,
      "balance_loss_mlp": 1.0265913,
      "epoch": 0.6188901581194012,
      "flos": 20084766796800.0,
      "grad_norm": 1.8103273862304827,
      "language_loss": 0.74371338,
      "learning_rate": 1.3395994337514218e-06,
      "loss": 0.76578808,
      "num_input_tokens_seen": 110963170,
      "step": 5147,
      "time_per_iteration": 2.517326831817627
    },
    {
      "auxiliary_loss_clip": 0.01146696,
      "auxiliary_loss_mlp": 0.01029153,
      "balance_loss_clip": 1.04609561,
      "balance_loss_mlp": 1.02197051,
      "epoch": 0.6190104010100402,
      "flos": 25700728360320.0,
      "grad_norm": 1.5218938877917312,
      "language_loss": 0.78868079,
      "learning_rate": 1.3388642040382725e-06,
      "loss": 0.81043923,
      "num_input_tokens_seen": 110983595,
      "step": 5148,
      "time_per_iteration": 2.561779499053955
    },
    {
      "auxiliary_loss_clip": 0.01130573,
      "auxiliary_loss_mlp": 0.01029597,
      "balance_loss_clip": 1.04388106,
      "balance_loss_mlp": 1.02221751,
      "epoch": 0.6191306439006794,
      "flos": 30442372974720.0,
      "grad_norm": 1.640214104135997,
      "language_loss": 0.84181845,
      "learning_rate": 1.3381290746188975e-06,
      "loss": 0.86342019,
      "num_input_tokens_seen": 111002965,
      "step": 5149,
      "time_per_iteration": 3.484020709991455
    },
    {
      "auxiliary_loss_clip": 0.01160113,
      "auxiliary_loss_mlp": 0.01028642,
      "balance_loss_clip": 1.05448604,
      "balance_loss_mlp": 1.02158153,
      "epoch": 0.6192508867913185,
      "flos": 26685793918080.0,
      "grad_norm": 1.7873049273108141,
      "language_loss": 0.67142868,
      "learning_rate": 1.3373940456048152e-06,
      "loss": 0.69331622,
      "num_input_tokens_seen": 111022990,
      "step": 5150,
      "time_per_iteration": 2.567626714706421
    },
    {
      "auxiliary_loss_clip": 0.01168376,
      "auxiliary_loss_mlp": 0.01028632,
      "balance_loss_clip": 1.05033398,
      "balance_loss_mlp": 1.02149451,
      "epoch": 0.6193711296819575,
      "flos": 36722036090880.0,
      "grad_norm": 1.6964579370371657,
      "language_loss": 0.59114981,
      "learning_rate": 1.3366591171075299e-06,
      "loss": 0.6131199,
      "num_input_tokens_seen": 111046495,
      "step": 5151,
      "time_per_iteration": 2.6399786472320557
    },
    {
      "auxiliary_loss_clip": 0.01140954,
      "auxiliary_loss_mlp": 0.01028153,
      "balance_loss_clip": 1.0498271,
      "balance_loss_mlp": 1.02180552,
      "epoch": 0.6194913725725967,
      "flos": 25192556697600.0,
      "grad_norm": 1.9983548170306384,
      "language_loss": 0.91059065,
      "learning_rate": 1.335924289238529e-06,
      "loss": 0.93228179,
      "num_input_tokens_seen": 111065705,
      "step": 5152,
      "time_per_iteration": 2.5936784744262695
    },
    {
      "auxiliary_loss_clip": 0.01156047,
      "auxiliary_loss_mlp": 0.00760572,
      "balance_loss_clip": 1.05357289,
      "balance_loss_mlp": 1.00015712,
      "epoch": 0.6196116154632357,
      "flos": 21178821196800.0,
      "grad_norm": 2.3335301913165183,
      "language_loss": 0.76665592,
      "learning_rate": 1.3351895621092859e-06,
      "loss": 0.78582215,
      "num_input_tokens_seen": 111086050,
      "step": 5153,
      "time_per_iteration": 2.544772148132324
    },
    {
      "auxiliary_loss_clip": 0.01057259,
      "auxiliary_loss_mlp": 0.01025575,
      "balance_loss_clip": 1.03254128,
      "balance_loss_mlp": 1.01883411,
      "epoch": 0.6197318583538748,
      "flos": 16253744803200.0,
      "grad_norm": 2.2905575702297014,
      "language_loss": 0.76629651,
      "learning_rate": 1.3344549358312567e-06,
      "loss": 0.78712487,
      "num_input_tokens_seen": 111104450,
      "step": 5154,
      "time_per_iteration": 2.8105413913726807
    },
    {
      "auxiliary_loss_clip": 0.01161015,
      "auxiliary_loss_mlp": 0.01026976,
      "balance_loss_clip": 1.05375409,
      "balance_loss_mlp": 1.01987123,
      "epoch": 0.619852101244514,
      "flos": 24425612478720.0,
      "grad_norm": 1.8226751688676548,
      "language_loss": 0.78392243,
      "learning_rate": 1.3337204105158852e-06,
      "loss": 0.80580235,
      "num_input_tokens_seen": 111123320,
      "step": 5155,
      "time_per_iteration": 2.7605700492858887
    },
    {
      "auxiliary_loss_clip": 0.01112349,
      "auxiliary_loss_mlp": 0.01031491,
      "balance_loss_clip": 1.03800929,
      "balance_loss_mlp": 1.02471375,
      "epoch": 0.619972344135153,
      "flos": 16727298733440.0,
      "grad_norm": 1.9816947116489763,
      "language_loss": 0.73077148,
      "learning_rate": 1.332985986274597e-06,
      "loss": 0.7522099,
      "num_input_tokens_seen": 111140950,
      "step": 5156,
      "time_per_iteration": 3.3562145233154297
    },
    {
      "auxiliary_loss_clip": 0.01093804,
      "auxiliary_loss_mlp": 0.00760622,
      "balance_loss_clip": 1.04363656,
      "balance_loss_mlp": 1.00017738,
      "epoch": 0.6200925870257921,
      "flos": 12495190498560.0,
      "grad_norm": 2.052623051624734,
      "language_loss": 0.75271773,
      "learning_rate": 1.3322516632188047e-06,
      "loss": 0.77126199,
      "num_input_tokens_seen": 111157845,
      "step": 5157,
      "time_per_iteration": 2.6294710636138916
    },
    {
      "auxiliary_loss_clip": 0.01125089,
      "auxiliary_loss_mlp": 0.01030069,
      "balance_loss_clip": 1.04587913,
      "balance_loss_mlp": 1.02250206,
      "epoch": 0.6202128299164312,
      "flos": 26539350168960.0,
      "grad_norm": 1.8071717079403955,
      "language_loss": 0.67194307,
      "learning_rate": 1.3315174414599045e-06,
      "loss": 0.69349468,
      "num_input_tokens_seen": 111179165,
      "step": 5158,
      "time_per_iteration": 2.630873441696167
    },
    {
      "auxiliary_loss_clip": 0.01149249,
      "auxiliary_loss_mlp": 0.01028262,
      "balance_loss_clip": 1.04702973,
      "balance_loss_mlp": 1.02090061,
      "epoch": 0.6203330728070703,
      "flos": 18770508069120.0,
      "grad_norm": 1.720232384646922,
      "language_loss": 0.75248718,
      "learning_rate": 1.3307833211092768e-06,
      "loss": 0.77426225,
      "num_input_tokens_seen": 111197830,
      "step": 5159,
      "time_per_iteration": 3.3772566318511963
    },
    {
      "auxiliary_loss_clip": 0.01169966,
      "auxiliary_loss_mlp": 0.0103653,
      "balance_loss_clip": 1.05132747,
      "balance_loss_mlp": 1.02909112,
      "epoch": 0.6204533156977093,
      "flos": 20629782835200.0,
      "grad_norm": 1.5449880578574986,
      "language_loss": 0.75166082,
      "learning_rate": 1.3300493022782873e-06,
      "loss": 0.77372575,
      "num_input_tokens_seen": 111218400,
      "step": 5160,
      "time_per_iteration": 3.2620089054107666
    },
    {
      "auxiliary_loss_clip": 0.01101538,
      "auxiliary_loss_mlp": 0.00760727,
      "balance_loss_clip": 1.04208565,
      "balance_loss_mlp": 1.00012589,
      "epoch": 0.6205735585883485,
      "flos": 17348050598400.0,
      "grad_norm": 1.7828310269634837,
      "language_loss": 0.72114384,
      "learning_rate": 1.3293153850782855e-06,
      "loss": 0.73976648,
      "num_input_tokens_seen": 111236720,
      "step": 5161,
      "time_per_iteration": 2.5783932209014893
    },
    {
      "auxiliary_loss_clip": 0.01116894,
      "auxiliary_loss_mlp": 0.01033176,
      "balance_loss_clip": 1.04381883,
      "balance_loss_mlp": 1.02476037,
      "epoch": 0.6206938014789876,
      "flos": 22965017742720.0,
      "grad_norm": 1.7214100184002883,
      "language_loss": 0.71327335,
      "learning_rate": 1.3285815696206069e-06,
      "loss": 0.73477405,
      "num_input_tokens_seen": 111258265,
      "step": 5162,
      "time_per_iteration": 2.5998733043670654
    },
    {
      "auxiliary_loss_clip": 0.01125252,
      "auxiliary_loss_mlp": 0.010341,
      "balance_loss_clip": 1.04244089,
      "balance_loss_mlp": 1.02634585,
      "epoch": 0.6208140443696266,
      "flos": 23983192661760.0,
      "grad_norm": 1.7170823976815268,
      "language_loss": 0.76853412,
      "learning_rate": 1.32784785601657e-06,
      "loss": 0.79012764,
      "num_input_tokens_seen": 111277675,
      "step": 5163,
      "time_per_iteration": 2.607658624649048
    },
    {
      "auxiliary_loss_clip": 0.0114339,
      "auxiliary_loss_mlp": 0.01025723,
      "balance_loss_clip": 1.04639435,
      "balance_loss_mlp": 1.01841879,
      "epoch": 0.6209342872602658,
      "flos": 35077291303680.0,
      "grad_norm": 1.7215905297211473,
      "language_loss": 0.74007201,
      "learning_rate": 1.3271142443774798e-06,
      "loss": 0.76176322,
      "num_input_tokens_seen": 111299910,
      "step": 5164,
      "time_per_iteration": 2.665806293487549
    },
    {
      "auxiliary_loss_clip": 0.01138201,
      "auxiliary_loss_mlp": 0.01028395,
      "balance_loss_clip": 1.04683566,
      "balance_loss_mlp": 1.02146018,
      "epoch": 0.6210545301509048,
      "flos": 26979327861120.0,
      "grad_norm": 1.6689304339663622,
      "language_loss": 0.8140825,
      "learning_rate": 1.3263807348146228e-06,
      "loss": 0.83574843,
      "num_input_tokens_seen": 111319765,
      "step": 5165,
      "time_per_iteration": 2.6081652641296387
    },
    {
      "auxiliary_loss_clip": 0.01137896,
      "auxiliary_loss_mlp": 0.01029873,
      "balance_loss_clip": 1.04421425,
      "balance_loss_mlp": 1.02155793,
      "epoch": 0.6211747730415439,
      "flos": 33618240852480.0,
      "grad_norm": 1.8873975851623666,
      "language_loss": 0.73288655,
      "learning_rate": 1.3256473274392733e-06,
      "loss": 0.75456429,
      "num_input_tokens_seen": 111341110,
      "step": 5166,
      "time_per_iteration": 2.6589717864990234
    },
    {
      "auxiliary_loss_clip": 0.01168509,
      "auxiliary_loss_mlp": 0.01034472,
      "balance_loss_clip": 1.05080104,
      "balance_loss_mlp": 1.02684855,
      "epoch": 0.6212950159321831,
      "flos": 34167099646080.0,
      "grad_norm": 2.283986737012843,
      "language_loss": 0.70028925,
      "learning_rate": 1.3249140223626873e-06,
      "loss": 0.72231913,
      "num_input_tokens_seen": 111362730,
      "step": 5167,
      "time_per_iteration": 2.5960729122161865
    },
    {
      "auxiliary_loss_clip": 0.01153583,
      "auxiliary_loss_mlp": 0.01026763,
      "balance_loss_clip": 1.0499289,
      "balance_loss_mlp": 1.01989996,
      "epoch": 0.6214152588228221,
      "flos": 27965758135680.0,
      "grad_norm": 1.9727413954173814,
      "language_loss": 0.75221986,
      "learning_rate": 1.3241808196961077e-06,
      "loss": 0.77402329,
      "num_input_tokens_seen": 111383855,
      "step": 5168,
      "time_per_iteration": 2.570199489593506
    },
    {
      "auxiliary_loss_clip": 0.0112971,
      "auxiliary_loss_mlp": 0.0102787,
      "balance_loss_clip": 1.04542994,
      "balance_loss_mlp": 1.02105117,
      "epoch": 0.6215355017134612,
      "flos": 20230204965120.0,
      "grad_norm": 1.6366827693567902,
      "language_loss": 0.7063421,
      "learning_rate": 1.3234477195507608e-06,
      "loss": 0.72791779,
      "num_input_tokens_seen": 111402685,
      "step": 5169,
      "time_per_iteration": 2.5499320030212402
    },
    {
      "auxiliary_loss_clip": 0.01127864,
      "auxiliary_loss_mlp": 0.01029059,
      "balance_loss_clip": 1.04820776,
      "balance_loss_mlp": 1.0218972,
      "epoch": 0.6216557446041003,
      "flos": 41428129219200.0,
      "grad_norm": 2.716323751079925,
      "language_loss": 0.62144917,
      "learning_rate": 1.322714722037857e-06,
      "loss": 0.64301836,
      "num_input_tokens_seen": 111424130,
      "step": 5170,
      "time_per_iteration": 2.7369654178619385
    },
    {
      "auxiliary_loss_clip": 0.01134497,
      "auxiliary_loss_mlp": 0.01029969,
      "balance_loss_clip": 1.04579306,
      "balance_loss_mlp": 1.02208281,
      "epoch": 0.6217759874947394,
      "flos": 27928770105600.0,
      "grad_norm": 1.951617636745396,
      "language_loss": 0.77565312,
      "learning_rate": 1.321981827268591e-06,
      "loss": 0.79729784,
      "num_input_tokens_seen": 111444785,
      "step": 5171,
      "time_per_iteration": 2.6371102333068848
    },
    {
      "auxiliary_loss_clip": 0.01143694,
      "auxiliary_loss_mlp": 0.01033428,
      "balance_loss_clip": 1.04692936,
      "balance_loss_mlp": 1.02657628,
      "epoch": 0.6218962303853784,
      "flos": 21765673601280.0,
      "grad_norm": 1.8334906426195423,
      "language_loss": 0.81656498,
      "learning_rate": 1.3212490353541426e-06,
      "loss": 0.83833623,
      "num_input_tokens_seen": 111467045,
      "step": 5172,
      "time_per_iteration": 2.5682530403137207
    },
    {
      "auxiliary_loss_clip": 0.01168685,
      "auxiliary_loss_mlp": 0.01026438,
      "balance_loss_clip": 1.04919744,
      "balance_loss_mlp": 1.01868677,
      "epoch": 0.6220164732760175,
      "flos": 21246260981760.0,
      "grad_norm": 1.8543169350127724,
      "language_loss": 0.80459416,
      "learning_rate": 1.3205163464056762e-06,
      "loss": 0.82654536,
      "num_input_tokens_seen": 111483650,
      "step": 5173,
      "time_per_iteration": 2.470243453979492
    },
    {
      "auxiliary_loss_clip": 0.01151909,
      "auxiliary_loss_mlp": 0.01029583,
      "balance_loss_clip": 1.04816699,
      "balance_loss_mlp": 1.0228653,
      "epoch": 0.6221367161666567,
      "flos": 26136360506880.0,
      "grad_norm": 1.7124061624158653,
      "language_loss": 0.72822261,
      "learning_rate": 1.319783760534339e-06,
      "loss": 0.75003749,
      "num_input_tokens_seen": 111502895,
      "step": 5174,
      "time_per_iteration": 2.537566900253296
    },
    {
      "auxiliary_loss_clip": 0.01156203,
      "auxiliary_loss_mlp": 0.01031271,
      "balance_loss_clip": 1.05073893,
      "balance_loss_mlp": 1.02398741,
      "epoch": 0.6222569590572957,
      "flos": 16284196558080.0,
      "grad_norm": 2.0431924889410684,
      "language_loss": 0.75563133,
      "learning_rate": 1.319051277851266e-06,
      "loss": 0.77750605,
      "num_input_tokens_seen": 111519180,
      "step": 5175,
      "time_per_iteration": 3.6371822357177734
    },
    {
      "auxiliary_loss_clip": 0.01157383,
      "auxiliary_loss_mlp": 0.01023611,
      "balance_loss_clip": 1.05157399,
      "balance_loss_mlp": 1.01700425,
      "epoch": 0.6223772019479348,
      "flos": 18223840005120.0,
      "grad_norm": 2.0157851450972806,
      "language_loss": 0.83789229,
      "learning_rate": 1.3183188984675716e-06,
      "loss": 0.85970223,
      "num_input_tokens_seen": 111537545,
      "step": 5176,
      "time_per_iteration": 2.494339942932129
    },
    {
      "auxiliary_loss_clip": 0.01140615,
      "auxiliary_loss_mlp": 0.01022117,
      "balance_loss_clip": 1.04869926,
      "balance_loss_mlp": 1.01505971,
      "epoch": 0.6224974448385739,
      "flos": 27489797994240.0,
      "grad_norm": 2.3001817730458742,
      "language_loss": 0.71657372,
      "learning_rate": 1.3175866224943586e-06,
      "loss": 0.73820108,
      "num_input_tokens_seen": 111556265,
      "step": 5177,
      "time_per_iteration": 2.5685248374938965
    },
    {
      "auxiliary_loss_clip": 0.01146636,
      "auxiliary_loss_mlp": 0.0103147,
      "balance_loss_clip": 1.04965425,
      "balance_loss_mlp": 1.02412367,
      "epoch": 0.622617687729213,
      "flos": 19791951125760.0,
      "grad_norm": 2.040748052949397,
      "language_loss": 0.73570776,
      "learning_rate": 1.316854450042712e-06,
      "loss": 0.75748885,
      "num_input_tokens_seen": 111574205,
      "step": 5178,
      "time_per_iteration": 2.5504019260406494
    },
    {
      "auxiliary_loss_clip": 0.01160558,
      "auxiliary_loss_mlp": 0.01028748,
      "balance_loss_clip": 1.0512526,
      "balance_loss_mlp": 1.02120233,
      "epoch": 0.622737930619852,
      "flos": 23038886062080.0,
      "grad_norm": 1.832917914034419,
      "language_loss": 0.74279761,
      "learning_rate": 1.3161223812237024e-06,
      "loss": 0.76469064,
      "num_input_tokens_seen": 111593560,
      "step": 5179,
      "time_per_iteration": 2.512796640396118
    },
    {
      "auxiliary_loss_clip": 0.01167862,
      "auxiliary_loss_mlp": 0.01026672,
      "balance_loss_clip": 1.0491606,
      "balance_loss_mlp": 1.01954043,
      "epoch": 0.6228581735104912,
      "flos": 12634271959680.0,
      "grad_norm": 2.2559578846950523,
      "language_loss": 0.85758203,
      "learning_rate": 1.3153904161483842e-06,
      "loss": 0.87952733,
      "num_input_tokens_seen": 111608860,
      "step": 5180,
      "time_per_iteration": 2.4368677139282227
    },
    {
      "auxiliary_loss_clip": 0.01124765,
      "auxiliary_loss_mlp": 0.01026854,
      "balance_loss_clip": 1.04492068,
      "balance_loss_mlp": 1.01929021,
      "epoch": 0.6229784164011303,
      "flos": 23802813538560.0,
      "grad_norm": 2.0160247309044403,
      "language_loss": 0.85559416,
      "learning_rate": 1.3146585549277953e-06,
      "loss": 0.8771103,
      "num_input_tokens_seen": 111627500,
      "step": 5181,
      "time_per_iteration": 2.6080188751220703
    },
    {
      "auxiliary_loss_clip": 0.01147027,
      "auxiliary_loss_mlp": 0.0102924,
      "balance_loss_clip": 1.04961109,
      "balance_loss_mlp": 1.02239776,
      "epoch": 0.6230986592917693,
      "flos": 22414219614720.0,
      "grad_norm": 3.0077358829789462,
      "language_loss": 0.78319687,
      "learning_rate": 1.3139267976729591e-06,
      "loss": 0.80495954,
      "num_input_tokens_seen": 111647690,
      "step": 5182,
      "time_per_iteration": 3.321894645690918
    },
    {
      "auxiliary_loss_clip": 0.01159637,
      "auxiliary_loss_mlp": 0.01028486,
      "balance_loss_clip": 1.05029225,
      "balance_loss_mlp": 1.02112436,
      "epoch": 0.6232189021824085,
      "flos": 34528217028480.0,
      "grad_norm": 1.7583238406418198,
      "language_loss": 0.71962136,
      "learning_rate": 1.3131951444948815e-06,
      "loss": 0.74150258,
      "num_input_tokens_seen": 111667090,
      "step": 5183,
      "time_per_iteration": 2.6119823455810547
    },
    {
      "auxiliary_loss_clip": 0.01148265,
      "auxiliary_loss_mlp": 0.01024249,
      "balance_loss_clip": 1.05118251,
      "balance_loss_mlp": 1.0170188,
      "epoch": 0.6233391450730476,
      "flos": 22237000888320.0,
      "grad_norm": 1.9355787623256409,
      "language_loss": 0.75934672,
      "learning_rate": 1.3124635955045546e-06,
      "loss": 0.78107178,
      "num_input_tokens_seen": 111686905,
      "step": 5184,
      "time_per_iteration": 2.5423145294189453
    },
    {
      "auxiliary_loss_clip": 0.01099455,
      "auxiliary_loss_mlp": 0.00760478,
      "balance_loss_clip": 1.04134107,
      "balance_loss_mlp": 1.00017524,
      "epoch": 0.6234593879636866,
      "flos": 20332693445760.0,
      "grad_norm": 1.768472984692508,
      "language_loss": 0.84171474,
      "learning_rate": 1.3117321508129537e-06,
      "loss": 0.86031413,
      "num_input_tokens_seen": 111704985,
      "step": 5185,
      "time_per_iteration": 3.4371814727783203
    },
    {
      "auxiliary_loss_clip": 0.01146794,
      "auxiliary_loss_mlp": 0.01029664,
      "balance_loss_clip": 1.05001211,
      "balance_loss_mlp": 1.02262521,
      "epoch": 0.6235796308543258,
      "flos": 20664903358080.0,
      "grad_norm": 1.52837673526448,
      "language_loss": 0.76622415,
      "learning_rate": 1.3110008105310388e-06,
      "loss": 0.78798872,
      "num_input_tokens_seen": 111724805,
      "step": 5186,
      "time_per_iteration": 2.5734033584594727
    },
    {
      "auxiliary_loss_clip": 0.01168696,
      "auxiliary_loss_mlp": 0.01024289,
      "balance_loss_clip": 1.04785001,
      "balance_loss_mlp": 1.01701713,
      "epoch": 0.6236998737449648,
      "flos": 26618641441920.0,
      "grad_norm": 1.600723080134172,
      "language_loss": 0.77884406,
      "learning_rate": 1.3102695747697526e-06,
      "loss": 0.80077398,
      "num_input_tokens_seen": 111747675,
      "step": 5187,
      "time_per_iteration": 3.2823331356048584
    },
    {
      "auxiliary_loss_clip": 0.01102006,
      "auxiliary_loss_mlp": 0.01035022,
      "balance_loss_clip": 1.04736614,
      "balance_loss_mlp": 1.02757716,
      "epoch": 0.6238201166356039,
      "flos": 12674599954560.0,
      "grad_norm": 2.3129180404277667,
      "language_loss": 0.90380132,
      "learning_rate": 1.3095384436400237e-06,
      "loss": 0.92517161,
      "num_input_tokens_seen": 111759205,
      "step": 5188,
      "time_per_iteration": 2.560361385345459
    },
    {
      "auxiliary_loss_clip": 0.01146146,
      "auxiliary_loss_mlp": 0.01028129,
      "balance_loss_clip": 1.04902518,
      "balance_loss_mlp": 1.02074957,
      "epoch": 0.623940359526243,
      "flos": 10452160730880.0,
      "grad_norm": 1.9810178738965096,
      "language_loss": 0.8193481,
      "learning_rate": 1.3088074172527633e-06,
      "loss": 0.8410908,
      "num_input_tokens_seen": 111776335,
      "step": 5189,
      "time_per_iteration": 2.5189342498779297
    },
    {
      "auxiliary_loss_clip": 0.01143915,
      "auxiliary_loss_mlp": 0.01034559,
      "balance_loss_clip": 1.04769528,
      "balance_loss_mlp": 1.02675676,
      "epoch": 0.6240606024168821,
      "flos": 29059525226880.0,
      "grad_norm": 2.20769104268534,
      "language_loss": 0.71468085,
      "learning_rate": 1.3080764957188684e-06,
      "loss": 0.73646557,
      "num_input_tokens_seen": 111796580,
      "step": 5190,
      "time_per_iteration": 2.6017136573791504
    },
    {
      "auxiliary_loss_clip": 0.01112981,
      "auxiliary_loss_mlp": 0.01027448,
      "balance_loss_clip": 1.04456329,
      "balance_loss_mlp": 1.02017629,
      "epoch": 0.6241808453075212,
      "flos": 22018089450240.0,
      "grad_norm": 1.7542403417941204,
      "language_loss": 0.70396137,
      "learning_rate": 1.3073456791492192e-06,
      "loss": 0.72536564,
      "num_input_tokens_seen": 111816290,
      "step": 5191,
      "time_per_iteration": 2.618889808654785
    },
    {
      "auxiliary_loss_clip": 0.01142517,
      "auxiliary_loss_mlp": 0.01028263,
      "balance_loss_clip": 1.04556227,
      "balance_loss_mlp": 1.02089298,
      "epoch": 0.6243010881981603,
      "flos": 21138708683520.0,
      "grad_norm": 1.8197945253148466,
      "language_loss": 0.78072929,
      "learning_rate": 1.3066149676546801e-06,
      "loss": 0.80243707,
      "num_input_tokens_seen": 111834470,
      "step": 5192,
      "time_per_iteration": 2.5318827629089355
    },
    {
      "auxiliary_loss_clip": 0.011379,
      "auxiliary_loss_mlp": 0.0102957,
      "balance_loss_clip": 1.04870832,
      "balance_loss_mlp": 1.02213132,
      "epoch": 0.6244213310887994,
      "flos": 22344948236160.0,
      "grad_norm": 1.7766372821926621,
      "language_loss": 0.66128749,
      "learning_rate": 1.3058843613460985e-06,
      "loss": 0.68296218,
      "num_input_tokens_seen": 111852410,
      "step": 5193,
      "time_per_iteration": 2.5914816856384277
    },
    {
      "auxiliary_loss_clip": 0.01134029,
      "auxiliary_loss_mlp": 0.01028407,
      "balance_loss_clip": 1.04646051,
      "balance_loss_mlp": 1.02093291,
      "epoch": 0.6245415739794384,
      "flos": 15231978524160.0,
      "grad_norm": 2.0117209516271513,
      "language_loss": 0.74408537,
      "learning_rate": 1.3051538603343075e-06,
      "loss": 0.76570976,
      "num_input_tokens_seen": 111870340,
      "step": 5194,
      "time_per_iteration": 2.568589925765991
    },
    {
      "auxiliary_loss_clip": 0.01157628,
      "auxiliary_loss_mlp": 0.010325,
      "balance_loss_clip": 1.05144441,
      "balance_loss_mlp": 1.02542841,
      "epoch": 0.6246618168700776,
      "flos": 18879891960960.0,
      "grad_norm": 1.794311297939756,
      "language_loss": 0.67598224,
      "learning_rate": 1.3044234647301235e-06,
      "loss": 0.69788355,
      "num_input_tokens_seen": 111888365,
      "step": 5195,
      "time_per_iteration": 2.496995210647583
    },
    {
      "auxiliary_loss_clip": 0.01153044,
      "auxiliary_loss_mlp": 0.01026313,
      "balance_loss_clip": 1.04899132,
      "balance_loss_mlp": 1.01918125,
      "epoch": 0.6247820597607167,
      "flos": 14319201087360.0,
      "grad_norm": 1.7610892013042936,
      "language_loss": 0.72327089,
      "learning_rate": 1.303693174644347e-06,
      "loss": 0.7450645,
      "num_input_tokens_seen": 111905840,
      "step": 5196,
      "time_per_iteration": 2.4909324645996094
    },
    {
      "auxiliary_loss_clip": 0.01136989,
      "auxiliary_loss_mlp": 0.01028997,
      "balance_loss_clip": 1.04617727,
      "balance_loss_mlp": 1.02153397,
      "epoch": 0.6249023026513557,
      "flos": 22637979388800.0,
      "grad_norm": 1.9314742743347986,
      "language_loss": 0.80449784,
      "learning_rate": 1.3029629901877625e-06,
      "loss": 0.82615769,
      "num_input_tokens_seen": 111925215,
      "step": 5197,
      "time_per_iteration": 2.5404744148254395
    },
    {
      "auxiliary_loss_clip": 0.01159302,
      "auxiliary_loss_mlp": 0.01023337,
      "balance_loss_clip": 1.05096984,
      "balance_loss_mlp": 1.01540923,
      "epoch": 0.6250225455419949,
      "flos": 20266690204800.0,
      "grad_norm": 2.7731438562953623,
      "language_loss": 0.77385223,
      "learning_rate": 1.3022329114711376e-06,
      "loss": 0.79567862,
      "num_input_tokens_seen": 111943925,
      "step": 5198,
      "time_per_iteration": 2.4914488792419434
    },
    {
      "auxiliary_loss_clip": 0.01136987,
      "auxiliary_loss_mlp": 0.01027011,
      "balance_loss_clip": 1.04812264,
      "balance_loss_mlp": 1.01947665,
      "epoch": 0.6251427884326339,
      "flos": 23437853400960.0,
      "grad_norm": 1.7919818908305056,
      "language_loss": 0.6962198,
      "learning_rate": 1.3015029386052256e-06,
      "loss": 0.71785975,
      "num_input_tokens_seen": 111964095,
      "step": 5199,
      "time_per_iteration": 2.5591471195220947
    },
    {
      "auxiliary_loss_clip": 0.011369,
      "auxiliary_loss_mlp": 0.01031399,
      "balance_loss_clip": 1.04892445,
      "balance_loss_mlp": 1.0240171,
      "epoch": 0.625263031323273,
      "flos": 31723055464320.0,
      "grad_norm": 1.8315489891565093,
      "language_loss": 0.72909236,
      "learning_rate": 1.3007730717007622e-06,
      "loss": 0.75077528,
      "num_input_tokens_seen": 111984910,
      "step": 5200,
      "time_per_iteration": 2.6530284881591797
    },
    {
      "auxiliary_loss_clip": 0.01172937,
      "auxiliary_loss_mlp": 0.01028709,
      "balance_loss_clip": 1.05205774,
      "balance_loss_mlp": 1.02067447,
      "epoch": 0.6253832742139122,
      "flos": 24134341092480.0,
      "grad_norm": 1.8346682050444139,
      "language_loss": 0.75605935,
      "learning_rate": 1.3000433108684676e-06,
      "loss": 0.77807581,
      "num_input_tokens_seen": 112005410,
      "step": 5201,
      "time_per_iteration": 0.6659543514251709
    },
    {
      "auxiliary_loss_clip": 0.01154229,
      "auxiliary_loss_mlp": 0.01025495,
      "balance_loss_clip": 1.05041695,
      "balance_loss_mlp": 1.01802015,
      "epoch": 0.6255035171045512,
      "flos": 27668812400640.0,
      "grad_norm": 2.541733654732183,
      "language_loss": 0.80216455,
      "learning_rate": 1.2993136562190467e-06,
      "loss": 0.82396173,
      "num_input_tokens_seen": 112024530,
      "step": 5202,
      "time_per_iteration": 2.5568699836730957
    },
    {
      "auxiliary_loss_clip": 0.01147457,
      "auxiliary_loss_mlp": 0.01025065,
      "balance_loss_clip": 1.04928041,
      "balance_loss_mlp": 1.01796317,
      "epoch": 0.6256237599951903,
      "flos": 20227798753920.0,
      "grad_norm": 1.5321611409742733,
      "language_loss": 0.70474744,
      "learning_rate": 1.2985841078631871e-06,
      "loss": 0.72647274,
      "num_input_tokens_seen": 112043850,
      "step": 5203,
      "time_per_iteration": 2.5611369609832764
    },
    {
      "auxiliary_loss_clip": 0.01097131,
      "auxiliary_loss_mlp": 0.01032775,
      "balance_loss_clip": 1.03868425,
      "balance_loss_mlp": 1.02524996,
      "epoch": 0.6257440028858293,
      "flos": 24170574936960.0,
      "grad_norm": 1.7145014743278557,
      "language_loss": 0.78358847,
      "learning_rate": 1.2978546659115608e-06,
      "loss": 0.80488747,
      "num_input_tokens_seen": 112061930,
      "step": 5204,
      "time_per_iteration": 2.655010223388672
    },
    {
      "auxiliary_loss_clip": 0.01146859,
      "auxiliary_loss_mlp": 0.01030734,
      "balance_loss_clip": 1.04952753,
      "balance_loss_mlp": 1.0237726,
      "epoch": 0.6258642457764685,
      "flos": 15851940289920.0,
      "grad_norm": 1.8766782093286296,
      "language_loss": 0.85402501,
      "learning_rate": 1.2971253304748228e-06,
      "loss": 0.87580097,
      "num_input_tokens_seen": 112079645,
      "step": 5205,
      "time_per_iteration": 2.534308671951294
    },
    {
      "auxiliary_loss_clip": 0.01159732,
      "auxiliary_loss_mlp": 0.01032658,
      "balance_loss_clip": 1.05185747,
      "balance_loss_mlp": 1.02463484,
      "epoch": 0.6259844886671075,
      "flos": 11911354836480.0,
      "grad_norm": 1.699075465959052,
      "language_loss": 0.74669278,
      "learning_rate": 1.296396101663614e-06,
      "loss": 0.76861668,
      "num_input_tokens_seen": 112096205,
      "step": 5206,
      "time_per_iteration": 2.4956533908843994
    },
    {
      "auxiliary_loss_clip": 0.0116016,
      "auxiliary_loss_mlp": 0.01027557,
      "balance_loss_clip": 1.05104709,
      "balance_loss_mlp": 1.01968634,
      "epoch": 0.6261047315577466,
      "flos": 15887958652800.0,
      "grad_norm": 2.2929426295305912,
      "language_loss": 0.84167844,
      "learning_rate": 1.2956669795885565e-06,
      "loss": 0.86355561,
      "num_input_tokens_seen": 112112835,
      "step": 5207,
      "time_per_iteration": 2.5008697509765625
    },
    {
      "auxiliary_loss_clip": 0.0112475,
      "auxiliary_loss_mlp": 0.01038334,
      "balance_loss_clip": 1.04937518,
      "balance_loss_mlp": 1.03019166,
      "epoch": 0.6262249744483858,
      "flos": 31248926916480.0,
      "grad_norm": 1.9414601215441698,
      "language_loss": 0.68169481,
      "learning_rate": 1.294937964360259e-06,
      "loss": 0.70332569,
      "num_input_tokens_seen": 112133105,
      "step": 5208,
      "time_per_iteration": 3.4493167400360107
    },
    {
      "auxiliary_loss_clip": 0.01145109,
      "auxiliary_loss_mlp": 0.01027528,
      "balance_loss_clip": 1.04747438,
      "balance_loss_mlp": 1.01966906,
      "epoch": 0.6263452173390248,
      "flos": 27198598435200.0,
      "grad_norm": 2.2348863702348964,
      "language_loss": 0.71896529,
      "learning_rate": 1.2942090560893108e-06,
      "loss": 0.74069166,
      "num_input_tokens_seen": 112152510,
      "step": 5209,
      "time_per_iteration": 2.589372396469116
    },
    {
      "auxiliary_loss_clip": 0.01169763,
      "auxiliary_loss_mlp": 0.01030486,
      "balance_loss_clip": 1.05144584,
      "balance_loss_mlp": 1.02317834,
      "epoch": 0.6264654602296639,
      "flos": 37342069683840.0,
      "grad_norm": 1.8133664996820804,
      "language_loss": 0.60170698,
      "learning_rate": 1.2934802548862882e-06,
      "loss": 0.6237095,
      "num_input_tokens_seen": 112175295,
      "step": 5210,
      "time_per_iteration": 2.6099376678466797
    },
    {
      "auxiliary_loss_clip": 0.01139698,
      "auxiliary_loss_mlp": 0.01025256,
      "balance_loss_clip": 1.04686618,
      "balance_loss_mlp": 1.01835346,
      "epoch": 0.626585703120303,
      "flos": 14756952136320.0,
      "grad_norm": 1.929557069517171,
      "language_loss": 0.82623327,
      "learning_rate": 1.292751560861749e-06,
      "loss": 0.84788275,
      "num_input_tokens_seen": 112190200,
      "step": 5211,
      "time_per_iteration": 3.336229085922241
    },
    {
      "auxiliary_loss_clip": 0.01174293,
      "auxiliary_loss_mlp": 0.0102262,
      "balance_loss_clip": 1.05301905,
      "balance_loss_mlp": 1.01513422,
      "epoch": 0.6267059460109421,
      "flos": 22347318533760.0,
      "grad_norm": 1.6852147687894672,
      "language_loss": 0.7974298,
      "learning_rate": 1.2920229741262354e-06,
      "loss": 0.819399,
      "num_input_tokens_seen": 112208205,
      "step": 5212,
      "time_per_iteration": 2.499915838241577
    },
    {
      "auxiliary_loss_clip": 0.01146073,
      "auxiliary_loss_mlp": 0.01024091,
      "balance_loss_clip": 1.0495106,
      "balance_loss_mlp": 1.01677203,
      "epoch": 0.6268261889015811,
      "flos": 17748813617280.0,
      "grad_norm": 2.1766925770676577,
      "language_loss": 0.75174648,
      "learning_rate": 1.2912944947902739e-06,
      "loss": 0.77344811,
      "num_input_tokens_seen": 112224690,
      "step": 5213,
      "time_per_iteration": 3.2579081058502197
    },
    {
      "auxiliary_loss_clip": 0.01148193,
      "auxiliary_loss_mlp": 0.01030456,
      "balance_loss_clip": 1.04871202,
      "balance_loss_mlp": 1.02298164,
      "epoch": 0.6269464317922203,
      "flos": 32846484211200.0,
      "grad_norm": 2.3897229614298356,
      "language_loss": 0.71500623,
      "learning_rate": 1.2905661229643742e-06,
      "loss": 0.73679268,
      "num_input_tokens_seen": 112244450,
      "step": 5214,
      "time_per_iteration": 2.6466479301452637
    },
    {
      "auxiliary_loss_clip": 0.01171117,
      "auxiliary_loss_mlp": 0.01027445,
      "balance_loss_clip": 1.05068719,
      "balance_loss_mlp": 1.0199405,
      "epoch": 0.6270666746828594,
      "flos": 17929192740480.0,
      "grad_norm": 2.146285455772101,
      "language_loss": 0.84430504,
      "learning_rate": 1.2898378587590299e-06,
      "loss": 0.86629069,
      "num_input_tokens_seen": 112261050,
      "step": 5215,
      "time_per_iteration": 2.451789140701294
    },
    {
      "auxiliary_loss_clip": 0.01155099,
      "auxiliary_loss_mlp": 0.01034074,
      "balance_loss_clip": 1.05147719,
      "balance_loss_mlp": 1.02666497,
      "epoch": 0.6271869175734984,
      "flos": 17457326749440.0,
      "grad_norm": 1.702420332328967,
      "language_loss": 0.87213141,
      "learning_rate": 1.2891097022847173e-06,
      "loss": 0.89402312,
      "num_input_tokens_seen": 112278395,
      "step": 5216,
      "time_per_iteration": 2.472630023956299
    },
    {
      "auxiliary_loss_clip": 0.01146782,
      "auxiliary_loss_mlp": 0.01032374,
      "balance_loss_clip": 1.05018306,
      "balance_loss_mlp": 1.02408314,
      "epoch": 0.6273071604641376,
      "flos": 26868615166080.0,
      "grad_norm": 1.8197510911623902,
      "language_loss": 0.66592324,
      "learning_rate": 1.2883816536518978e-06,
      "loss": 0.68771487,
      "num_input_tokens_seen": 112299535,
      "step": 5217,
      "time_per_iteration": 2.599257469177246
    },
    {
      "auxiliary_loss_clip": 0.01154425,
      "auxiliary_loss_mlp": 0.0102718,
      "balance_loss_clip": 1.04974794,
      "balance_loss_mlp": 1.01989055,
      "epoch": 0.6274274033547766,
      "flos": 26062384446720.0,
      "grad_norm": 1.8153244458861182,
      "language_loss": 0.81823868,
      "learning_rate": 1.2876537129710155e-06,
      "loss": 0.84005469,
      "num_input_tokens_seen": 112317265,
      "step": 5218,
      "time_per_iteration": 2.5419886112213135
    },
    {
      "auxiliary_loss_clip": 0.01137331,
      "auxiliary_loss_mlp": 0.01028481,
      "balance_loss_clip": 1.0506041,
      "balance_loss_mlp": 1.02048206,
      "epoch": 0.6275476462454157,
      "flos": 20266259241600.0,
      "grad_norm": 1.9794411490862094,
      "language_loss": 0.75545585,
      "learning_rate": 1.286925880352499e-06,
      "loss": 0.77711391,
      "num_input_tokens_seen": 112336125,
      "step": 5219,
      "time_per_iteration": 2.527164936065674
    },
    {
      "auxiliary_loss_clip": 0.0113687,
      "auxiliary_loss_mlp": 0.01027478,
      "balance_loss_clip": 1.0460633,
      "balance_loss_mlp": 1.01998818,
      "epoch": 0.6276678891360549,
      "flos": 26320402817280.0,
      "grad_norm": 1.8570717461055009,
      "language_loss": 0.7139672,
      "learning_rate": 1.2861981559067592e-06,
      "loss": 0.7356106,
      "num_input_tokens_seen": 112356730,
      "step": 5220,
      "time_per_iteration": 2.5783236026763916
    },
    {
      "auxiliary_loss_clip": 0.01107449,
      "auxiliary_loss_mlp": 0.01024124,
      "balance_loss_clip": 1.04382324,
      "balance_loss_mlp": 1.01718581,
      "epoch": 0.6277881320266939,
      "flos": 13912512324480.0,
      "grad_norm": 1.9140501614911964,
      "language_loss": 0.80344212,
      "learning_rate": 1.2854705397441917e-06,
      "loss": 0.82475781,
      "num_input_tokens_seen": 112372270,
      "step": 5221,
      "time_per_iteration": 2.5935115814208984
    },
    {
      "auxiliary_loss_clip": 0.01126249,
      "auxiliary_loss_mlp": 0.0102702,
      "balance_loss_clip": 1.04588342,
      "balance_loss_mlp": 1.02028453,
      "epoch": 0.627908374917333,
      "flos": 27048922462080.0,
      "grad_norm": 2.0061590324379353,
      "language_loss": 0.77362049,
      "learning_rate": 1.2847430319751747e-06,
      "loss": 0.79515314,
      "num_input_tokens_seen": 112390365,
      "step": 5222,
      "time_per_iteration": 2.6301941871643066
    },
    {
      "auxiliary_loss_clip": 0.01151667,
      "auxiliary_loss_mlp": 0.0103216,
      "balance_loss_clip": 1.05072439,
      "balance_loss_mlp": 1.0248909,
      "epoch": 0.6280286178079721,
      "flos": 23769201386880.0,
      "grad_norm": 2.2119271624858143,
      "language_loss": 0.67226642,
      "learning_rate": 1.2840156327100712e-06,
      "loss": 0.69410467,
      "num_input_tokens_seen": 112407490,
      "step": 5223,
      "time_per_iteration": 2.5097033977508545
    },
    {
      "auxiliary_loss_clip": 0.01171648,
      "auxiliary_loss_mlp": 0.01025604,
      "balance_loss_clip": 1.05217803,
      "balance_loss_mlp": 1.01820135,
      "epoch": 0.6281488606986112,
      "flos": 26359150613760.0,
      "grad_norm": 1.7317833074892703,
      "language_loss": 0.72275043,
      "learning_rate": 1.2832883420592272e-06,
      "loss": 0.74472296,
      "num_input_tokens_seen": 112426385,
      "step": 5224,
      "time_per_iteration": 2.5106117725372314
    },
    {
      "auxiliary_loss_clip": 0.01138828,
      "auxiliary_loss_mlp": 0.01027795,
      "balance_loss_clip": 1.04916835,
      "balance_loss_mlp": 1.02081525,
      "epoch": 0.6282691035892503,
      "flos": 36137194848000.0,
      "grad_norm": 2.2476171281782773,
      "language_loss": 0.64646047,
      "learning_rate": 1.282561160132972e-06,
      "loss": 0.6681267,
      "num_input_tokens_seen": 112446905,
      "step": 5225,
      "time_per_iteration": 2.661320686340332
    },
    {
      "auxiliary_loss_clip": 0.01147553,
      "auxiliary_loss_mlp": 0.01032194,
      "balance_loss_clip": 1.04704201,
      "balance_loss_mlp": 1.02475512,
      "epoch": 0.6283893464798894,
      "flos": 26537231266560.0,
      "grad_norm": 1.5761578062372898,
      "language_loss": 0.80815202,
      "learning_rate": 1.2818340870416186e-06,
      "loss": 0.8299495,
      "num_input_tokens_seen": 112468040,
      "step": 5226,
      "time_per_iteration": 2.6052603721618652
    },
    {
      "auxiliary_loss_clip": 0.01134962,
      "auxiliary_loss_mlp": 0.01028382,
      "balance_loss_clip": 1.04452109,
      "balance_loss_mlp": 1.0204246,
      "epoch": 0.6285095893705285,
      "flos": 22237216369920.0,
      "grad_norm": 1.8138038511275894,
      "language_loss": 0.75804353,
      "learning_rate": 1.2811071228954626e-06,
      "loss": 0.77967697,
      "num_input_tokens_seen": 112486675,
      "step": 5227,
      "time_per_iteration": 3.3870127201080322
    },
    {
      "auxiliary_loss_clip": 0.01140014,
      "auxiliary_loss_mlp": 0.01027383,
      "balance_loss_clip": 1.04765677,
      "balance_loss_mlp": 1.02028358,
      "epoch": 0.6286298322611675,
      "flos": 26542259170560.0,
      "grad_norm": 1.9712488372984,
      "language_loss": 0.80846465,
      "learning_rate": 1.2803802678047846e-06,
      "loss": 0.83013856,
      "num_input_tokens_seen": 112506825,
      "step": 5228,
      "time_per_iteration": 2.5906550884246826
    },
    {
      "auxiliary_loss_clip": 0.01148906,
      "auxiliary_loss_mlp": 0.01032897,
      "balance_loss_clip": 1.05155087,
      "balance_loss_mlp": 1.0247308,
      "epoch": 0.6287500751518067,
      "flos": 21795227516160.0,
      "grad_norm": 1.6809937527976335,
      "language_loss": 0.73997855,
      "learning_rate": 1.279653521879848e-06,
      "loss": 0.76179653,
      "num_input_tokens_seen": 112526890,
      "step": 5229,
      "time_per_iteration": 2.541074752807617
    },
    {
      "auxiliary_loss_clip": 0.01079756,
      "auxiliary_loss_mlp": 0.01033265,
      "balance_loss_clip": 1.04111624,
      "balance_loss_mlp": 1.02635968,
      "epoch": 0.6288703180424458,
      "flos": 20009605587840.0,
      "grad_norm": 2.0193330013215007,
      "language_loss": 0.83691359,
      "learning_rate": 1.2789268852308997e-06,
      "loss": 0.85804379,
      "num_input_tokens_seen": 112542100,
      "step": 5230,
      "time_per_iteration": 2.633125066757202
    },
    {
      "auxiliary_loss_clip": 0.01148882,
      "auxiliary_loss_mlp": 0.01029244,
      "balance_loss_clip": 1.04881716,
      "balance_loss_mlp": 1.02206719,
      "epoch": 0.6289905609330848,
      "flos": 22124923476480.0,
      "grad_norm": 1.85697216489786,
      "language_loss": 0.70357013,
      "learning_rate": 1.2782003579681688e-06,
      "loss": 0.72535145,
      "num_input_tokens_seen": 112561630,
      "step": 5231,
      "time_per_iteration": 2.5269412994384766
    },
    {
      "auxiliary_loss_clip": 0.01175377,
      "auxiliary_loss_mlp": 0.01033884,
      "balance_loss_clip": 1.05518019,
      "balance_loss_mlp": 1.02606678,
      "epoch": 0.629110803823724,
      "flos": 25518481729920.0,
      "grad_norm": 1.578646478280308,
      "language_loss": 0.74172223,
      "learning_rate": 1.2774739402018701e-06,
      "loss": 0.76381481,
      "num_input_tokens_seen": 112582465,
      "step": 5232,
      "time_per_iteration": 2.5098023414611816
    },
    {
      "auxiliary_loss_clip": 0.0115308,
      "auxiliary_loss_mlp": 0.01029405,
      "balance_loss_clip": 1.04943681,
      "balance_loss_mlp": 1.02154374,
      "epoch": 0.629231046714363,
      "flos": 20886616056960.0,
      "grad_norm": 1.598995911542712,
      "language_loss": 0.7286483,
      "learning_rate": 1.2767476320422002e-06,
      "loss": 0.75047314,
      "num_input_tokens_seen": 112602390,
      "step": 5233,
      "time_per_iteration": 2.5157248973846436
    },
    {
      "auxiliary_loss_clip": 0.01037167,
      "auxiliary_loss_mlp": 0.01002988,
      "balance_loss_clip": 1.01615548,
      "balance_loss_mlp": 1.00164688,
      "epoch": 0.6293512896050021,
      "flos": 65050027908480.0,
      "grad_norm": 0.6795422908223762,
      "language_loss": 0.57200038,
      "learning_rate": 1.2760214335993392e-06,
      "loss": 0.59240198,
      "num_input_tokens_seen": 112669035,
      "step": 5234,
      "time_per_iteration": 4.00806188583374
    },
    {
      "auxiliary_loss_clip": 0.01148257,
      "auxiliary_loss_mlp": 0.01026214,
      "balance_loss_clip": 1.04674411,
      "balance_loss_mlp": 1.01939559,
      "epoch": 0.6294715324956413,
      "flos": 34677857088000.0,
      "grad_norm": 2.1005456680586327,
      "language_loss": 0.58687347,
      "learning_rate": 1.2752953449834514e-06,
      "loss": 0.60861814,
      "num_input_tokens_seen": 112691485,
      "step": 5235,
      "time_per_iteration": 2.6186487674713135
    },
    {
      "auxiliary_loss_clip": 0.01168195,
      "auxiliary_loss_mlp": 0.01026622,
      "balance_loss_clip": 1.05009174,
      "balance_loss_mlp": 1.01955247,
      "epoch": 0.6295917753862803,
      "flos": 22784207656320.0,
      "grad_norm": 1.5590057609326045,
      "language_loss": 0.80316848,
      "learning_rate": 1.2745693663046836e-06,
      "loss": 0.82511663,
      "num_input_tokens_seen": 112710555,
      "step": 5236,
      "time_per_iteration": 2.472851514816284
    },
    {
      "auxiliary_loss_clip": 0.01153379,
      "auxiliary_loss_mlp": 0.01023058,
      "balance_loss_clip": 1.05023289,
      "balance_loss_mlp": 1.01629281,
      "epoch": 0.6297120182769194,
      "flos": 20850454039680.0,
      "grad_norm": 1.7347657083228893,
      "language_loss": 0.80440217,
      "learning_rate": 1.2738434976731662e-06,
      "loss": 0.82616651,
      "num_input_tokens_seen": 112728740,
      "step": 5237,
      "time_per_iteration": 3.3013107776641846
    },
    {
      "auxiliary_loss_clip": 0.01143186,
      "auxiliary_loss_mlp": 0.01027937,
      "balance_loss_clip": 1.05051565,
      "balance_loss_mlp": 1.02054608,
      "epoch": 0.6298322611675584,
      "flos": 19497662997120.0,
      "grad_norm": 1.6899578468461065,
      "language_loss": 0.75417429,
      "learning_rate": 1.2731177391990125e-06,
      "loss": 0.77588552,
      "num_input_tokens_seen": 112748665,
      "step": 5238,
      "time_per_iteration": 3.2653064727783203
    },
    {
      "auxiliary_loss_clip": 0.01142885,
      "auxiliary_loss_mlp": 0.01026974,
      "balance_loss_clip": 1.04751539,
      "balance_loss_mlp": 1.01967192,
      "epoch": 0.6299525040581976,
      "flos": 12604466649600.0,
      "grad_norm": 3.9175031458442633,
      "language_loss": 0.81791008,
      "learning_rate": 1.2723920909923203e-06,
      "loss": 0.83960867,
      "num_input_tokens_seen": 112764410,
      "step": 5239,
      "time_per_iteration": 2.504490613937378
    },
    {
      "auxiliary_loss_clip": 0.01066408,
      "auxiliary_loss_mlp": 0.01002761,
      "balance_loss_clip": 1.01783037,
      "balance_loss_mlp": 1.00162268,
      "epoch": 0.6300727469488366,
      "flos": 57725685636480.0,
      "grad_norm": 0.847045568964251,
      "language_loss": 0.60465908,
      "learning_rate": 1.2716665531631688e-06,
      "loss": 0.62535071,
      "num_input_tokens_seen": 112818695,
      "step": 5240,
      "time_per_iteration": 3.0114898681640625
    },
    {
      "auxiliary_loss_clip": 0.01158583,
      "auxiliary_loss_mlp": 0.01024111,
      "balance_loss_clip": 1.04941559,
      "balance_loss_mlp": 1.0167532,
      "epoch": 0.6301929898394757,
      "flos": 22527302607360.0,
      "grad_norm": 1.6242927907048788,
      "language_loss": 0.77171874,
      "learning_rate": 1.270941125821623e-06,
      "loss": 0.79354572,
      "num_input_tokens_seen": 112839120,
      "step": 5241,
      "time_per_iteration": 2.5108468532562256
    },
    {
      "auxiliary_loss_clip": 0.01151103,
      "auxiliary_loss_mlp": 0.01027164,
      "balance_loss_clip": 1.04716671,
      "balance_loss_mlp": 1.01977921,
      "epoch": 0.6303132327301149,
      "flos": 28293550675200.0,
      "grad_norm": 1.5762895922832727,
      "language_loss": 0.75326675,
      "learning_rate": 1.2702158090777278e-06,
      "loss": 0.77504945,
      "num_input_tokens_seen": 112860210,
      "step": 5242,
      "time_per_iteration": 2.5658206939697266
    },
    {
      "auxiliary_loss_clip": 0.01124019,
      "auxiliary_loss_mlp": 0.01032182,
      "balance_loss_clip": 1.04518557,
      "balance_loss_mlp": 1.02541709,
      "epoch": 0.6304334756207539,
      "flos": 25264521596160.0,
      "grad_norm": 1.841698726774463,
      "language_loss": 0.74236083,
      "learning_rate": 1.2694906030415148e-06,
      "loss": 0.76392281,
      "num_input_tokens_seen": 112877955,
      "step": 5243,
      "time_per_iteration": 2.6280767917633057
    },
    {
      "auxiliary_loss_clip": 0.01152035,
      "auxiliary_loss_mlp": 0.01028711,
      "balance_loss_clip": 1.049317,
      "balance_loss_mlp": 1.02142739,
      "epoch": 0.630553718511393,
      "flos": 18033548728320.0,
      "grad_norm": 2.3355892446748956,
      "language_loss": 0.82109106,
      "learning_rate": 1.2687655078229958e-06,
      "loss": 0.84289849,
      "num_input_tokens_seen": 112892285,
      "step": 5244,
      "time_per_iteration": 2.483689785003662
    },
    {
      "auxiliary_loss_clip": 0.01141443,
      "auxiliary_loss_mlp": 0.01032853,
      "balance_loss_clip": 1.04911995,
      "balance_loss_mlp": 1.02563179,
      "epoch": 0.6306739614020321,
      "flos": 27304103658240.0,
      "grad_norm": 2.4028599496150136,
      "language_loss": 0.69385219,
      "learning_rate": 1.2680405235321678e-06,
      "loss": 0.71559513,
      "num_input_tokens_seen": 112913620,
      "step": 5245,
      "time_per_iteration": 2.6019532680511475
    },
    {
      "auxiliary_loss_clip": 0.01146026,
      "auxiliary_loss_mlp": 0.00761629,
      "balance_loss_clip": 1.05246842,
      "balance_loss_mlp": 1.00022018,
      "epoch": 0.6307942042926712,
      "flos": 15341434243200.0,
      "grad_norm": 2.1548742754596018,
      "language_loss": 0.7840693,
      "learning_rate": 1.267315650279011e-06,
      "loss": 0.80314589,
      "num_input_tokens_seen": 112932090,
      "step": 5246,
      "time_per_iteration": 2.5243239402770996
    },
    {
      "auxiliary_loss_clip": 0.0112408,
      "auxiliary_loss_mlp": 0.01025901,
      "balance_loss_clip": 1.04981184,
      "balance_loss_mlp": 1.01900232,
      "epoch": 0.6309144471833102,
      "flos": 19606400444160.0,
      "grad_norm": 1.7956486715892954,
      "language_loss": 0.74300712,
      "learning_rate": 1.2665908881734874e-06,
      "loss": 0.76450694,
      "num_input_tokens_seen": 112950925,
      "step": 5247,
      "time_per_iteration": 2.56534743309021
    },
    {
      "auxiliary_loss_clip": 0.01155315,
      "auxiliary_loss_mlp": 0.01027075,
      "balance_loss_clip": 1.04957557,
      "balance_loss_mlp": 1.01992297,
      "epoch": 0.6310346900739494,
      "flos": 17493345112320.0,
      "grad_norm": 2.205427431304428,
      "language_loss": 0.85000455,
      "learning_rate": 1.2658662373255432e-06,
      "loss": 0.87182844,
      "num_input_tokens_seen": 112969315,
      "step": 5248,
      "time_per_iteration": 2.4878129959106445
    },
    {
      "auxiliary_loss_clip": 0.01042392,
      "auxiliary_loss_mlp": 0.01000422,
      "balance_loss_clip": 1.01508808,
      "balance_loss_mlp": 0.99920636,
      "epoch": 0.6311549329645885,
      "flos": 55070164131840.0,
      "grad_norm": 0.7088834820248641,
      "language_loss": 0.5232054,
      "learning_rate": 1.2651416978451063e-06,
      "loss": 0.54363358,
      "num_input_tokens_seen": 113034700,
      "step": 5249,
      "time_per_iteration": 3.1999776363372803
    },
    {
      "auxiliary_loss_clip": 0.01175047,
      "auxiliary_loss_mlp": 0.01025881,
      "balance_loss_clip": 1.05469072,
      "balance_loss_mlp": 1.01813221,
      "epoch": 0.6312751758552275,
      "flos": 41902545075840.0,
      "grad_norm": 1.7954949156765625,
      "language_loss": 0.64946747,
      "learning_rate": 1.2644172698420903e-06,
      "loss": 0.67147672,
      "num_input_tokens_seen": 113056805,
      "step": 5250,
      "time_per_iteration": 2.6572580337524414
    },
    {
      "auxiliary_loss_clip": 0.01128867,
      "auxiliary_loss_mlp": 0.01025845,
      "balance_loss_clip": 1.04681277,
      "balance_loss_mlp": 1.01833153,
      "epoch": 0.6313954187458667,
      "flos": 19646800266240.0,
      "grad_norm": 1.8926707312391093,
      "language_loss": 0.84562075,
      "learning_rate": 1.2636929534263892e-06,
      "loss": 0.86716783,
      "num_input_tokens_seen": 113075790,
      "step": 5251,
      "time_per_iteration": 2.5637598037719727
    },
    {
      "auxiliary_loss_clip": 0.01127453,
      "auxiliary_loss_mlp": 0.01028883,
      "balance_loss_clip": 1.04222584,
      "balance_loss_mlp": 1.02166438,
      "epoch": 0.6315156616365057,
      "flos": 22894273906560.0,
      "grad_norm": 1.6215485082009613,
      "language_loss": 0.77812272,
      "learning_rate": 1.2629687487078821e-06,
      "loss": 0.79968607,
      "num_input_tokens_seen": 113094600,
      "step": 5252,
      "time_per_iteration": 3.3799195289611816
    },
    {
      "auxiliary_loss_clip": 0.01157519,
      "auxiliary_loss_mlp": 0.01023315,
      "balance_loss_clip": 1.04858518,
      "balance_loss_mlp": 1.0161562,
      "epoch": 0.6316359045271448,
      "flos": 23726251699200.0,
      "grad_norm": 1.9671342806008167,
      "language_loss": 0.7660585,
      "learning_rate": 1.2622446557964293e-06,
      "loss": 0.78786683,
      "num_input_tokens_seen": 113112605,
      "step": 5253,
      "time_per_iteration": 2.509108781814575
    },
    {
      "auxiliary_loss_clip": 0.01141001,
      "auxiliary_loss_mlp": 0.01025798,
      "balance_loss_clip": 1.04520106,
      "balance_loss_mlp": 1.01812065,
      "epoch": 0.631756147417784,
      "flos": 33108417164160.0,
      "grad_norm": 1.733373803407461,
      "language_loss": 0.71466649,
      "learning_rate": 1.261520674801876e-06,
      "loss": 0.73633444,
      "num_input_tokens_seen": 113133200,
      "step": 5254,
      "time_per_iteration": 2.6446657180786133
    },
    {
      "auxiliary_loss_clip": 0.01140635,
      "auxiliary_loss_mlp": 0.01028362,
      "balance_loss_clip": 1.05056453,
      "balance_loss_mlp": 1.0202204,
      "epoch": 0.631876390308423,
      "flos": 31248424126080.0,
      "grad_norm": 2.465347172163128,
      "language_loss": 0.72640914,
      "learning_rate": 1.2607968058340488e-06,
      "loss": 0.74809909,
      "num_input_tokens_seen": 113152895,
      "step": 5255,
      "time_per_iteration": 2.6063601970672607
    },
    {
      "auxiliary_loss_clip": 0.01137228,
      "auxiliary_loss_mlp": 0.01024234,
      "balance_loss_clip": 1.045964,
      "balance_loss_mlp": 1.01728106,
      "epoch": 0.6319966331990621,
      "flos": 24681152810880.0,
      "grad_norm": 1.6619899516137349,
      "language_loss": 0.72924769,
      "learning_rate": 1.2600730490027583e-06,
      "loss": 0.75086236,
      "num_input_tokens_seen": 113173135,
      "step": 5256,
      "time_per_iteration": 2.586622714996338
    },
    {
      "auxiliary_loss_clip": 0.01129008,
      "auxiliary_loss_mlp": 0.01029954,
      "balance_loss_clip": 1.04696083,
      "balance_loss_mlp": 1.02269399,
      "epoch": 0.6321168760897012,
      "flos": 17491764913920.0,
      "grad_norm": 1.6945186845314215,
      "language_loss": 0.80235016,
      "learning_rate": 1.2593494044177984e-06,
      "loss": 0.8239398,
      "num_input_tokens_seen": 113191440,
      "step": 5257,
      "time_per_iteration": 2.553191661834717
    },
    {
      "auxiliary_loss_clip": 0.01172076,
      "auxiliary_loss_mlp": 0.01027334,
      "balance_loss_clip": 1.04980528,
      "balance_loss_mlp": 1.01907921,
      "epoch": 0.6322371189803403,
      "flos": 18295373940480.0,
      "grad_norm": 2.7746532501027885,
      "language_loss": 0.80743104,
      "learning_rate": 1.2586258721889448e-06,
      "loss": 0.82942522,
      "num_input_tokens_seen": 113208790,
      "step": 5258,
      "time_per_iteration": 2.438913345336914
    },
    {
      "auxiliary_loss_clip": 0.01105792,
      "auxiliary_loss_mlp": 0.01029974,
      "balance_loss_clip": 1.04590118,
      "balance_loss_mlp": 1.02233911,
      "epoch": 0.6323573618709794,
      "flos": 20157270399360.0,
      "grad_norm": 2.2843716298573917,
      "language_loss": 0.81504565,
      "learning_rate": 1.2579024524259573e-06,
      "loss": 0.83640337,
      "num_input_tokens_seen": 113225050,
      "step": 5259,
      "time_per_iteration": 2.570824384689331
    },
    {
      "auxiliary_loss_clip": 0.01135414,
      "auxiliary_loss_mlp": 0.01023085,
      "balance_loss_clip": 1.04289341,
      "balance_loss_mlp": 1.01602745,
      "epoch": 0.6324776047616185,
      "flos": 20042391726720.0,
      "grad_norm": 1.8595788246039904,
      "language_loss": 0.91569132,
      "learning_rate": 1.2571791452385768e-06,
      "loss": 0.93727636,
      "num_input_tokens_seen": 113242315,
      "step": 5260,
      "time_per_iteration": 3.3257579803466797
    },
    {
      "auxiliary_loss_clip": 0.01141395,
      "auxiliary_loss_mlp": 0.01025675,
      "balance_loss_clip": 1.04862928,
      "balance_loss_mlp": 1.01919639,
      "epoch": 0.6325978476522576,
      "flos": 30848235724800.0,
      "grad_norm": 1.8265499148477788,
      "language_loss": 0.7705543,
      "learning_rate": 1.2564559507365301e-06,
      "loss": 0.79222506,
      "num_input_tokens_seen": 113264720,
      "step": 5261,
      "time_per_iteration": 2.6189589500427246
    },
    {
      "auxiliary_loss_clip": 0.01143137,
      "auxiliary_loss_mlp": 0.01023816,
      "balance_loss_clip": 1.05031931,
      "balance_loss_mlp": 1.01613259,
      "epoch": 0.6327180905428966,
      "flos": 24535104111360.0,
      "grad_norm": 2.2016424927665823,
      "language_loss": 0.79216838,
      "learning_rate": 1.2557328690295244e-06,
      "loss": 0.81383789,
      "num_input_tokens_seen": 113282910,
      "step": 5262,
      "time_per_iteration": 2.5651912689208984
    },
    {
      "auxiliary_loss_clip": 0.01130835,
      "auxiliary_loss_mlp": 0.01024337,
      "balance_loss_clip": 1.04833591,
      "balance_loss_mlp": 1.01692843,
      "epoch": 0.6328383334335358,
      "flos": 21575274583680.0,
      "grad_norm": 1.5665022343446005,
      "language_loss": 0.76015413,
      "learning_rate": 1.255009900227251e-06,
      "loss": 0.7817058,
      "num_input_tokens_seen": 113301935,
      "step": 5263,
      "time_per_iteration": 3.4105286598205566
    },
    {
      "auxiliary_loss_clip": 0.01166635,
      "auxiliary_loss_mlp": 0.01024807,
      "balance_loss_clip": 1.05072403,
      "balance_loss_mlp": 1.018152,
      "epoch": 0.6329585763241748,
      "flos": 22929861306240.0,
      "grad_norm": 1.8962725714554596,
      "language_loss": 0.79651684,
      "learning_rate": 1.254287044439383e-06,
      "loss": 0.81843126,
      "num_input_tokens_seen": 113321540,
      "step": 5264,
      "time_per_iteration": 3.202737331390381
    },
    {
      "auxiliary_loss_clip": 0.01062243,
      "auxiliary_loss_mlp": 0.01004344,
      "balance_loss_clip": 1.01417518,
      "balance_loss_mlp": 1.00324178,
      "epoch": 0.6330788192148139,
      "flos": 70936897847040.0,
      "grad_norm": 0.780415206594517,
      "language_loss": 0.54503775,
      "learning_rate": 1.2535643017755776e-06,
      "loss": 0.56570363,
      "num_input_tokens_seen": 113383730,
      "step": 5265,
      "time_per_iteration": 3.1741766929626465
    },
    {
      "auxiliary_loss_clip": 0.01125211,
      "auxiliary_loss_mlp": 0.01026569,
      "balance_loss_clip": 1.04392219,
      "balance_loss_mlp": 1.01933002,
      "epoch": 0.6331990621054531,
      "flos": 21244501215360.0,
      "grad_norm": 2.553520083964701,
      "language_loss": 0.72059226,
      "learning_rate": 1.2528416723454737e-06,
      "loss": 0.74211007,
      "num_input_tokens_seen": 113400400,
      "step": 5266,
      "time_per_iteration": 2.5623018741607666
    },
    {
      "auxiliary_loss_clip": 0.01172493,
      "auxiliary_loss_mlp": 0.01028641,
      "balance_loss_clip": 1.05536699,
      "balance_loss_mlp": 1.02171516,
      "epoch": 0.6333193049960921,
      "flos": 34459412526720.0,
      "grad_norm": 1.4038826377435485,
      "language_loss": 0.71164888,
      "learning_rate": 1.2521191562586945e-06,
      "loss": 0.73366022,
      "num_input_tokens_seen": 113424050,
      "step": 5267,
      "time_per_iteration": 2.5977985858917236
    },
    {
      "auxiliary_loss_clip": 0.01168323,
      "auxiliary_loss_mlp": 0.00760945,
      "balance_loss_clip": 1.05116308,
      "balance_loss_mlp": 1.00019634,
      "epoch": 0.6334395478867312,
      "flos": 18329883932160.0,
      "grad_norm": 1.9642862195475197,
      "language_loss": 0.76787412,
      "learning_rate": 1.2513967536248445e-06,
      "loss": 0.78716683,
      "num_input_tokens_seen": 113440370,
      "step": 5268,
      "time_per_iteration": 2.4554340839385986
    },
    {
      "auxiliary_loss_clip": 0.01153462,
      "auxiliary_loss_mlp": 0.01033884,
      "balance_loss_clip": 1.0515604,
      "balance_loss_mlp": 1.02688956,
      "epoch": 0.6335597907773702,
      "flos": 23623152687360.0,
      "grad_norm": 1.725292107880279,
      "language_loss": 0.81158435,
      "learning_rate": 1.2506744645535117e-06,
      "loss": 0.83345783,
      "num_input_tokens_seen": 113460800,
      "step": 5269,
      "time_per_iteration": 2.518219232559204
    },
    {
      "auxiliary_loss_clip": 0.01131916,
      "auxiliary_loss_mlp": 0.01028623,
      "balance_loss_clip": 1.04156303,
      "balance_loss_mlp": 1.02186418,
      "epoch": 0.6336800336680094,
      "flos": 22710913954560.0,
      "grad_norm": 1.877429165637924,
      "language_loss": 0.60157049,
      "learning_rate": 1.249952289154267e-06,
      "loss": 0.62317592,
      "num_input_tokens_seen": 113480840,
      "step": 5270,
      "time_per_iteration": 2.549922466278076
    },
    {
      "auxiliary_loss_clip": 0.01086025,
      "auxiliary_loss_mlp": 0.01023256,
      "balance_loss_clip": 1.04093385,
      "balance_loss_mlp": 1.01630068,
      "epoch": 0.6338002765586485,
      "flos": 23622757637760.0,
      "grad_norm": 1.615336110551679,
      "language_loss": 0.7660858,
      "learning_rate": 1.2492302275366635e-06,
      "loss": 0.78717858,
      "num_input_tokens_seen": 113500515,
      "step": 5271,
      "time_per_iteration": 2.6407384872436523
    },
    {
      "auxiliary_loss_clip": 0.01148978,
      "auxiliary_loss_mlp": 0.01024192,
      "balance_loss_clip": 1.04731846,
      "balance_loss_mlp": 1.01576948,
      "epoch": 0.6339205194492875,
      "flos": 26505450708480.0,
      "grad_norm": 2.6687680457954506,
      "language_loss": 0.65455782,
      "learning_rate": 1.2485082798102377e-06,
      "loss": 0.67628944,
      "num_input_tokens_seen": 113520930,
      "step": 5272,
      "time_per_iteration": 2.5411806106567383
    },
    {
      "auxiliary_loss_clip": 0.01132202,
      "auxiliary_loss_mlp": 0.0102847,
      "balance_loss_clip": 1.04443288,
      "balance_loss_mlp": 1.02078128,
      "epoch": 0.6340407623399267,
      "flos": 18544306170240.0,
      "grad_norm": 2.048976169153386,
      "language_loss": 0.68498552,
      "learning_rate": 1.2477864460845084e-06,
      "loss": 0.7065922,
      "num_input_tokens_seen": 113537330,
      "step": 5273,
      "time_per_iteration": 2.5432846546173096
    },
    {
      "auxiliary_loss_clip": 0.01141896,
      "auxiliary_loss_mlp": 0.01029799,
      "balance_loss_clip": 1.04804111,
      "balance_loss_mlp": 1.02136743,
      "epoch": 0.6341610052305657,
      "flos": 17712579772800.0,
      "grad_norm": 2.88532119228345,
      "language_loss": 0.73068696,
      "learning_rate": 1.2470647264689776e-06,
      "loss": 0.75240386,
      "num_input_tokens_seen": 113555810,
      "step": 5274,
      "time_per_iteration": 2.520463466644287
    },
    {
      "auxiliary_loss_clip": 0.01101042,
      "auxiliary_loss_mlp": 0.01024554,
      "balance_loss_clip": 1.03979683,
      "balance_loss_mlp": 1.01730597,
      "epoch": 0.6342812481212048,
      "flos": 23587026583680.0,
      "grad_norm": 2.002164949840915,
      "language_loss": 0.70989579,
      "learning_rate": 1.2463431210731282e-06,
      "loss": 0.73115176,
      "num_input_tokens_seen": 113575395,
      "step": 5275,
      "time_per_iteration": 2.670114278793335
    },
    {
      "auxiliary_loss_clip": 0.01117937,
      "auxiliary_loss_mlp": 0.01026362,
      "balance_loss_clip": 1.04212427,
      "balance_loss_mlp": 1.01874161,
      "epoch": 0.634401491011844,
      "flos": 17821927751040.0,
      "grad_norm": 2.5214791031613615,
      "language_loss": 0.76576108,
      "learning_rate": 1.2456216300064289e-06,
      "loss": 0.78720409,
      "num_input_tokens_seen": 113592945,
      "step": 5276,
      "time_per_iteration": 2.595416307449341
    },
    {
      "auxiliary_loss_clip": 0.01134302,
      "auxiliary_loss_mlp": 0.01035185,
      "balance_loss_clip": 1.04536629,
      "balance_loss_mlp": 1.02692342,
      "epoch": 0.634521733902483,
      "flos": 21358158825600.0,
      "grad_norm": 1.737726130442711,
      "language_loss": 0.78408641,
      "learning_rate": 1.244900253378328e-06,
      "loss": 0.80578125,
      "num_input_tokens_seen": 113613000,
      "step": 5277,
      "time_per_iteration": 2.5708508491516113
    },
    {
      "auxiliary_loss_clip": 0.01073846,
      "auxiliary_loss_mlp": 0.01022183,
      "balance_loss_clip": 1.04371357,
      "balance_loss_mlp": 1.01538253,
      "epoch": 0.6346419767931221,
      "flos": 16545052103040.0,
      "grad_norm": 2.014718374703409,
      "language_loss": 0.69116527,
      "learning_rate": 1.2441789912982583e-06,
      "loss": 0.71212554,
      "num_input_tokens_seen": 113630085,
      "step": 5278,
      "time_per_iteration": 2.83880877494812
    },
    {
      "auxiliary_loss_clip": 0.01159478,
      "auxiliary_loss_mlp": 0.01026713,
      "balance_loss_clip": 1.05175853,
      "balance_loss_mlp": 1.01833844,
      "epoch": 0.6347622196837612,
      "flos": 24350989973760.0,
      "grad_norm": 2.408596342511094,
      "language_loss": 0.64826864,
      "learning_rate": 1.2434578438756346e-06,
      "loss": 0.67013049,
      "num_input_tokens_seen": 113650515,
      "step": 5279,
      "time_per_iteration": 3.3460628986358643
    },
    {
      "auxiliary_loss_clip": 0.01156665,
      "auxiliary_loss_mlp": 0.01022568,
      "balance_loss_clip": 1.04742002,
      "balance_loss_mlp": 1.01510525,
      "epoch": 0.6348824625744003,
      "flos": 64523178195840.0,
      "grad_norm": 1.926782960058554,
      "language_loss": 0.78127623,
      "learning_rate": 1.242736811219855e-06,
      "loss": 0.80306864,
      "num_input_tokens_seen": 113676475,
      "step": 5280,
      "time_per_iteration": 2.9064671993255615
    },
    {
      "auxiliary_loss_clip": 0.01149578,
      "auxiliary_loss_mlp": 0.010283,
      "balance_loss_clip": 1.04745221,
      "balance_loss_mlp": 1.02135015,
      "epoch": 0.6350027054650393,
      "flos": 28622133313920.0,
      "grad_norm": 1.6583067417777395,
      "language_loss": 0.82038379,
      "learning_rate": 1.2420158934402988e-06,
      "loss": 0.84216261,
      "num_input_tokens_seen": 113697090,
      "step": 5281,
      "time_per_iteration": 2.551565647125244
    },
    {
      "auxiliary_loss_clip": 0.01113948,
      "auxiliary_loss_mlp": 0.0102774,
      "balance_loss_clip": 1.0424459,
      "balance_loss_mlp": 1.01965201,
      "epoch": 0.6351229483556785,
      "flos": 23002544476800.0,
      "grad_norm": 1.8271357936703216,
      "language_loss": 0.84529698,
      "learning_rate": 1.2412950906463286e-06,
      "loss": 0.86671382,
      "num_input_tokens_seen": 113714395,
      "step": 5282,
      "time_per_iteration": 2.5827860832214355
    },
    {
      "auxiliary_loss_clip": 0.01116136,
      "auxiliary_loss_mlp": 0.01023427,
      "balance_loss_clip": 1.04631901,
      "balance_loss_mlp": 1.01660252,
      "epoch": 0.6352431912463176,
      "flos": 21939300967680.0,
      "grad_norm": 2.193792713707502,
      "language_loss": 0.90084457,
      "learning_rate": 1.2405744029472902e-06,
      "loss": 0.9222402,
      "num_input_tokens_seen": 113733880,
      "step": 5283,
      "time_per_iteration": 2.6503188610076904
    },
    {
      "auxiliary_loss_clip": 0.01138498,
      "auxiliary_loss_mlp": 0.01027846,
      "balance_loss_clip": 1.04782104,
      "balance_loss_mlp": 1.02052081,
      "epoch": 0.6353634341369566,
      "flos": 13735257684480.0,
      "grad_norm": 1.9441473537636285,
      "language_loss": 0.76425076,
      "learning_rate": 1.2398538304525108e-06,
      "loss": 0.78591418,
      "num_input_tokens_seen": 113752505,
      "step": 5284,
      "time_per_iteration": 2.5487475395202637
    },
    {
      "auxiliary_loss_clip": 0.01123924,
      "auxiliary_loss_mlp": 0.01032807,
      "balance_loss_clip": 1.04745722,
      "balance_loss_mlp": 1.02463543,
      "epoch": 0.6354836770275958,
      "flos": 19316170552320.0,
      "grad_norm": 2.2782106040304244,
      "language_loss": 0.75025249,
      "learning_rate": 1.2391333732713016e-06,
      "loss": 0.77181977,
      "num_input_tokens_seen": 113770310,
      "step": 5285,
      "time_per_iteration": 2.596153497695923
    },
    {
      "auxiliary_loss_clip": 0.01124427,
      "auxiliary_loss_mlp": 0.0102494,
      "balance_loss_clip": 1.04335284,
      "balance_loss_mlp": 1.01728082,
      "epoch": 0.6356039199182348,
      "flos": 21613375935360.0,
      "grad_norm": 2.1060014544051398,
      "language_loss": 0.79210305,
      "learning_rate": 1.2384130315129543e-06,
      "loss": 0.81359673,
      "num_input_tokens_seen": 113788635,
      "step": 5286,
      "time_per_iteration": 3.386382818222046
    },
    {
      "auxiliary_loss_clip": 0.01061647,
      "auxiliary_loss_mlp": 0.01029721,
      "balance_loss_clip": 1.03833008,
      "balance_loss_mlp": 1.02205873,
      "epoch": 0.6357241628088739,
      "flos": 18111978074880.0,
      "grad_norm": 1.991844715565865,
      "language_loss": 0.733805,
      "learning_rate": 1.2376928052867447e-06,
      "loss": 0.75471866,
      "num_input_tokens_seen": 113807755,
      "step": 5287,
      "time_per_iteration": 3.1269989013671875
    },
    {
      "auxiliary_loss_clip": 0.01145728,
      "auxiliary_loss_mlp": 0.01024759,
      "balance_loss_clip": 1.05085278,
      "balance_loss_mlp": 1.01790464,
      "epoch": 0.6358444056995131,
      "flos": 24935256599040.0,
      "grad_norm": 1.8368698816193973,
      "language_loss": 0.77488351,
      "learning_rate": 1.2369726947019299e-06,
      "loss": 0.79658836,
      "num_input_tokens_seen": 113828230,
      "step": 5288,
      "time_per_iteration": 2.7839736938476562
    },
    {
      "auxiliary_loss_clip": 0.0115172,
      "auxiliary_loss_mlp": 0.01025416,
      "balance_loss_clip": 1.04659235,
      "balance_loss_mlp": 1.01890087,
      "epoch": 0.6359646485901521,
      "flos": 23293348986240.0,
      "grad_norm": 2.1746476698967188,
      "language_loss": 0.67292738,
      "learning_rate": 1.2362526998677511e-06,
      "loss": 0.69469869,
      "num_input_tokens_seen": 113844595,
      "step": 5289,
      "time_per_iteration": 4.854312181472778
    },
    {
      "auxiliary_loss_clip": 0.01146013,
      "auxiliary_loss_mlp": 0.01028402,
      "balance_loss_clip": 1.04804385,
      "balance_loss_mlp": 1.02185392,
      "epoch": 0.6360848914807912,
      "flos": 20887442069760.0,
      "grad_norm": 1.6461488326199933,
      "language_loss": 0.84336793,
      "learning_rate": 1.2355328208934301e-06,
      "loss": 0.86511207,
      "num_input_tokens_seen": 113863470,
      "step": 5290,
      "time_per_iteration": 2.6122219562530518
    },
    {
      "auxiliary_loss_clip": 0.01155072,
      "auxiliary_loss_mlp": 0.00760837,
      "balance_loss_clip": 1.04703951,
      "balance_loss_mlp": 1.00023282,
      "epoch": 0.6362051343714303,
      "flos": 18479775386880.0,
      "grad_norm": 1.539212755255796,
      "language_loss": 0.72212434,
      "learning_rate": 1.2348130578881728e-06,
      "loss": 0.74128348,
      "num_input_tokens_seen": 113881690,
      "step": 5291,
      "time_per_iteration": 2.4878292083740234
    },
    {
      "auxiliary_loss_clip": 0.01172405,
      "auxiliary_loss_mlp": 0.01030203,
      "balance_loss_clip": 1.05200052,
      "balance_loss_mlp": 1.0223645,
      "epoch": 0.6363253772620694,
      "flos": 24389594115840.0,
      "grad_norm": 2.207314816324262,
      "language_loss": 0.76037192,
      "learning_rate": 1.2340934109611664e-06,
      "loss": 0.78239799,
      "num_input_tokens_seen": 113902450,
      "step": 5292,
      "time_per_iteration": 2.5352420806884766
    },
    {
      "auxiliary_loss_clip": 0.01144658,
      "auxiliary_loss_mlp": 0.01029554,
      "balance_loss_clip": 1.04801011,
      "balance_loss_mlp": 1.02126884,
      "epoch": 0.6364456201527084,
      "flos": 25958243940480.0,
      "grad_norm": 2.1454029544831412,
      "language_loss": 0.68477327,
      "learning_rate": 1.2333738802215798e-06,
      "loss": 0.70651543,
      "num_input_tokens_seen": 113922670,
      "step": 5293,
      "time_per_iteration": 2.5666635036468506
    },
    {
      "auxiliary_loss_clip": 0.01108014,
      "auxiliary_loss_mlp": 0.01035796,
      "balance_loss_clip": 1.04162312,
      "balance_loss_mlp": 1.02862537,
      "epoch": 0.6365658630433476,
      "flos": 20740711011840.0,
      "grad_norm": 1.9158997556390531,
      "language_loss": 0.81001699,
      "learning_rate": 1.2326544657785668e-06,
      "loss": 0.83145511,
      "num_input_tokens_seen": 113942360,
      "step": 5294,
      "time_per_iteration": 2.620126485824585
    },
    {
      "auxiliary_loss_clip": 0.01116684,
      "auxiliary_loss_mlp": 0.01028442,
      "balance_loss_clip": 1.04347181,
      "balance_loss_mlp": 1.02122366,
      "epoch": 0.6366861059339867,
      "flos": 21434146047360.0,
      "grad_norm": 2.194107931801718,
      "language_loss": 0.74390751,
      "learning_rate": 1.2319351677412608e-06,
      "loss": 0.76535869,
      "num_input_tokens_seen": 113959405,
      "step": 5295,
      "time_per_iteration": 2.549358367919922
    },
    {
      "auxiliary_loss_clip": 0.01134545,
      "auxiliary_loss_mlp": 0.01029238,
      "balance_loss_clip": 1.04663062,
      "balance_loss_mlp": 1.0219574,
      "epoch": 0.6368063488246257,
      "flos": 22267093507200.0,
      "grad_norm": 1.703136582340067,
      "language_loss": 0.73985112,
      "learning_rate": 1.2312159862187796e-06,
      "loss": 0.76148903,
      "num_input_tokens_seen": 113977815,
      "step": 5296,
      "time_per_iteration": 2.5587401390075684
    },
    {
      "auxiliary_loss_clip": 0.01172419,
      "auxiliary_loss_mlp": 0.01025827,
      "balance_loss_clip": 1.05201221,
      "balance_loss_mlp": 1.01770329,
      "epoch": 0.6369265917152649,
      "flos": 22420719976320.0,
      "grad_norm": 1.499902184555326,
      "language_loss": 0.75962514,
      "learning_rate": 1.2304969213202217e-06,
      "loss": 0.78160763,
      "num_input_tokens_seen": 113999075,
      "step": 5297,
      "time_per_iteration": 2.486853837966919
    },
    {
      "auxiliary_loss_clip": 0.01138533,
      "auxiliary_loss_mlp": 0.01033229,
      "balance_loss_clip": 1.04767323,
      "balance_loss_mlp": 1.02580833,
      "epoch": 0.6370468346059039,
      "flos": 24718176754560.0,
      "grad_norm": 2.2964027136958034,
      "language_loss": 0.79360092,
      "learning_rate": 1.2297779731546692e-06,
      "loss": 0.81531847,
      "num_input_tokens_seen": 114018170,
      "step": 5298,
      "time_per_iteration": 2.5833945274353027
    },
    {
      "auxiliary_loss_clip": 0.01143609,
      "auxiliary_loss_mlp": 0.01023112,
      "balance_loss_clip": 1.05226398,
      "balance_loss_mlp": 1.01621306,
      "epoch": 0.637167077496543,
      "flos": 25296589463040.0,
      "grad_norm": 1.8573831777631826,
      "language_loss": 0.78000706,
      "learning_rate": 1.2290591418311853e-06,
      "loss": 0.80167425,
      "num_input_tokens_seen": 114035565,
      "step": 5299,
      "time_per_iteration": 2.536646842956543
    },
    {
      "auxiliary_loss_clip": 0.0115389,
      "auxiliary_loss_mlp": 0.01033407,
      "balance_loss_clip": 1.04873061,
      "balance_loss_mlp": 1.02631664,
      "epoch": 0.637287320387182,
      "flos": 27671110871040.0,
      "grad_norm": 1.4926993010194123,
      "language_loss": 0.72275853,
      "learning_rate": 1.2283404274588172e-06,
      "loss": 0.74463153,
      "num_input_tokens_seen": 114054510,
      "step": 5300,
      "time_per_iteration": 2.5669572353363037
    },
    {
      "auxiliary_loss_clip": 0.00993237,
      "auxiliary_loss_mlp": 0.01001944,
      "balance_loss_clip": 1.01319146,
      "balance_loss_mlp": 1.00088906,
      "epoch": 0.6374075632778212,
      "flos": 63173406873600.0,
      "grad_norm": 0.7433207957837333,
      "language_loss": 0.527785,
      "learning_rate": 1.227621830146592e-06,
      "loss": 0.54773682,
      "num_input_tokens_seen": 114109875,
      "step": 5301,
      "time_per_iteration": 3.236919641494751
    },
    {
      "auxiliary_loss_clip": 0.01129018,
      "auxiliary_loss_mlp": 0.0102771,
      "balance_loss_clip": 1.04689705,
      "balance_loss_mlp": 1.02025306,
      "epoch": 0.6375278061684603,
      "flos": 25558127366400.0,
      "grad_norm": 1.8092996947499271,
      "language_loss": 0.79031444,
      "learning_rate": 1.2269033500035217e-06,
      "loss": 0.81188172,
      "num_input_tokens_seen": 114130010,
      "step": 5302,
      "time_per_iteration": 2.996455192565918
    },
    {
      "auxiliary_loss_clip": 0.0112503,
      "auxiliary_loss_mlp": 0.01026729,
      "balance_loss_clip": 1.04778111,
      "balance_loss_mlp": 1.01959991,
      "epoch": 0.6376480490590993,
      "flos": 25666362023040.0,
      "grad_norm": 1.8663737679600674,
      "language_loss": 0.74120653,
      "learning_rate": 1.2261849871385988e-06,
      "loss": 0.7627241,
      "num_input_tokens_seen": 114151115,
      "step": 5303,
      "time_per_iteration": 2.5912888050079346
    },
    {
      "auxiliary_loss_clip": 0.01172063,
      "auxiliary_loss_mlp": 0.01027325,
      "balance_loss_clip": 1.05189908,
      "balance_loss_mlp": 1.01987422,
      "epoch": 0.6377682919497385,
      "flos": 31537684350720.0,
      "grad_norm": 2.185380739485516,
      "language_loss": 0.62498164,
      "learning_rate": 1.2254667416607972e-06,
      "loss": 0.64697552,
      "num_input_tokens_seen": 114172715,
      "step": 5304,
      "time_per_iteration": 3.8812804222106934
    },
    {
      "auxiliary_loss_clip": 0.01156201,
      "auxiliary_loss_mlp": 0.01028128,
      "balance_loss_clip": 1.05010748,
      "balance_loss_mlp": 1.02041495,
      "epoch": 0.6378885348403776,
      "flos": 23039209284480.0,
      "grad_norm": 1.8849406424888104,
      "language_loss": 0.82934165,
      "learning_rate": 1.2247486136790756e-06,
      "loss": 0.85118496,
      "num_input_tokens_seen": 114192195,
      "step": 5305,
      "time_per_iteration": 2.552899122238159
    },
    {
      "auxiliary_loss_clip": 0.01157715,
      "auxiliary_loss_mlp": 0.01031763,
      "balance_loss_clip": 1.05034399,
      "balance_loss_mlp": 1.0243305,
      "epoch": 0.6380087777310166,
      "flos": 18697070712960.0,
      "grad_norm": 2.292850998365218,
      "language_loss": 0.80222344,
      "learning_rate": 1.2240306033023726e-06,
      "loss": 0.82411826,
      "num_input_tokens_seen": 114210020,
      "step": 5306,
      "time_per_iteration": 2.489532232284546
    },
    {
      "auxiliary_loss_clip": 0.01128256,
      "auxiliary_loss_mlp": 0.0102778,
      "balance_loss_clip": 1.04167151,
      "balance_loss_mlp": 1.02080059,
      "epoch": 0.6381290206216558,
      "flos": 23331558078720.0,
      "grad_norm": 2.343393505043547,
      "language_loss": 0.71983123,
      "learning_rate": 1.223312710639611e-06,
      "loss": 0.7413916,
      "num_input_tokens_seen": 114228740,
      "step": 5307,
      "time_per_iteration": 2.566988468170166
    },
    {
      "auxiliary_loss_clip": 0.01140645,
      "auxiliary_loss_mlp": 0.01025856,
      "balance_loss_clip": 1.04869556,
      "balance_loss_mlp": 1.01803839,
      "epoch": 0.6382492635122948,
      "flos": 18880466578560.0,
      "grad_norm": 2.060896434759809,
      "language_loss": 0.87113333,
      "learning_rate": 1.2225949357996928e-06,
      "loss": 0.8927983,
      "num_input_tokens_seen": 114246865,
      "step": 5308,
      "time_per_iteration": 2.531541585922241
    },
    {
      "auxiliary_loss_clip": 0.01152582,
      "auxiliary_loss_mlp": 0.01032973,
      "balance_loss_clip": 1.0505985,
      "balance_loss_mlp": 1.0258379,
      "epoch": 0.6383695064029339,
      "flos": 27819134818560.0,
      "grad_norm": 1.744613598380152,
      "language_loss": 0.80231178,
      "learning_rate": 1.221877278891505e-06,
      "loss": 0.82416725,
      "num_input_tokens_seen": 114266120,
      "step": 5309,
      "time_per_iteration": 2.5497078895568848
    },
    {
      "auxiliary_loss_clip": 0.01158983,
      "auxiliary_loss_mlp": 0.01030384,
      "balance_loss_clip": 1.05230653,
      "balance_loss_mlp": 1.0223434,
      "epoch": 0.638489749293573,
      "flos": 26395635853440.0,
      "grad_norm": 1.9288654505998788,
      "language_loss": 0.71635884,
      "learning_rate": 1.221159740023915e-06,
      "loss": 0.73825252,
      "num_input_tokens_seen": 114285950,
      "step": 5310,
      "time_per_iteration": 2.5483405590057373
    },
    {
      "auxiliary_loss_clip": 0.01136718,
      "auxiliary_loss_mlp": 0.00761116,
      "balance_loss_clip": 1.04830408,
      "balance_loss_mlp": 1.00022054,
      "epoch": 0.6386099921842121,
      "flos": 23988328306560.0,
      "grad_norm": 1.881137350257827,
      "language_loss": 0.72880101,
      "learning_rate": 1.2204423193057735e-06,
      "loss": 0.74777937,
      "num_input_tokens_seen": 114304780,
      "step": 5311,
      "time_per_iteration": 2.607191324234009
    },
    {
      "auxiliary_loss_clip": 0.01041303,
      "auxiliary_loss_mlp": 0.01001608,
      "balance_loss_clip": 1.01267374,
      "balance_loss_mlp": 1.00055909,
      "epoch": 0.6387302350748512,
      "flos": 71731169337600.0,
      "grad_norm": 0.8584354543625721,
      "language_loss": 0.63398916,
      "learning_rate": 1.2197250168459122e-06,
      "loss": 0.65441823,
      "num_input_tokens_seen": 114361180,
      "step": 5312,
      "time_per_iteration": 3.903045892715454
    },
    {
      "auxiliary_loss_clip": 0.01158273,
      "auxiliary_loss_mlp": 0.01029101,
      "balance_loss_clip": 1.04963994,
      "balance_loss_mlp": 1.02182364,
      "epoch": 0.6388504779654903,
      "flos": 14535778141440.0,
      "grad_norm": 1.7338731332971495,
      "language_loss": 0.7436738,
      "learning_rate": 1.2190078327531454e-06,
      "loss": 0.76554757,
      "num_input_tokens_seen": 114377425,
      "step": 5313,
      "time_per_iteration": 2.478257894515991
    },
    {
      "auxiliary_loss_clip": 0.01154777,
      "auxiliary_loss_mlp": 0.01030104,
      "balance_loss_clip": 1.04784405,
      "balance_loss_mlp": 1.02317774,
      "epoch": 0.6389707208561294,
      "flos": 22346133384960.0,
      "grad_norm": 1.493586880893391,
      "language_loss": 0.7272464,
      "learning_rate": 1.2182907671362697e-06,
      "loss": 0.7490952,
      "num_input_tokens_seen": 114398120,
      "step": 5314,
      "time_per_iteration": 2.530974864959717
    },
    {
      "auxiliary_loss_clip": 0.01157423,
      "auxiliary_loss_mlp": 0.01029146,
      "balance_loss_clip": 1.05184221,
      "balance_loss_mlp": 1.02188015,
      "epoch": 0.6390909637467684,
      "flos": 19426883247360.0,
      "grad_norm": 1.8606054502117242,
      "language_loss": 0.78452051,
      "learning_rate": 1.2175738201040626e-06,
      "loss": 0.80638623,
      "num_input_tokens_seen": 114415160,
      "step": 5315,
      "time_per_iteration": 4.0692524909973145
    },
    {
      "auxiliary_loss_clip": 0.01156175,
      "auxiliary_loss_mlp": 0.01027996,
      "balance_loss_clip": 1.05044675,
      "balance_loss_mlp": 1.02051544,
      "epoch": 0.6392112066374076,
      "flos": 24090852700800.0,
      "grad_norm": 1.6015531886200642,
      "language_loss": 0.78554583,
      "learning_rate": 1.2168569917652855e-06,
      "loss": 0.80738747,
      "num_input_tokens_seen": 114435015,
      "step": 5316,
      "time_per_iteration": 2.5467865467071533
    },
    {
      "auxiliary_loss_clip": 0.01154443,
      "auxiliary_loss_mlp": 0.01025294,
      "balance_loss_clip": 1.04961967,
      "balance_loss_mlp": 1.01815343,
      "epoch": 0.6393314495280467,
      "flos": 26795141896320.0,
      "grad_norm": 1.6070525992818563,
      "language_loss": 0.63701081,
      "learning_rate": 1.2161402822286797e-06,
      "loss": 0.65880811,
      "num_input_tokens_seen": 114455700,
      "step": 5317,
      "time_per_iteration": 2.5565969944000244
    },
    {
      "auxiliary_loss_clip": 0.01126601,
      "auxiliary_loss_mlp": 0.01027483,
      "balance_loss_clip": 1.04617918,
      "balance_loss_mlp": 1.01987159,
      "epoch": 0.6394516924186857,
      "flos": 20260692633600.0,
      "grad_norm": 1.9517276936326482,
      "language_loss": 0.7865243,
      "learning_rate": 1.2154236916029703e-06,
      "loss": 0.80806518,
      "num_input_tokens_seen": 114473675,
      "step": 5318,
      "time_per_iteration": 2.5691888332366943
    },
    {
      "auxiliary_loss_clip": 0.01111736,
      "auxiliary_loss_mlp": 0.01029389,
      "balance_loss_clip": 1.0401715,
      "balance_loss_mlp": 1.02212667,
      "epoch": 0.6395719353093249,
      "flos": 18368847210240.0,
      "grad_norm": 2.096708933658815,
      "language_loss": 0.73253292,
      "learning_rate": 1.2147072199968627e-06,
      "loss": 0.75394416,
      "num_input_tokens_seen": 114492310,
      "step": 5319,
      "time_per_iteration": 2.578903913497925
    },
    {
      "auxiliary_loss_clip": 0.01154714,
      "auxiliary_loss_mlp": 0.01024417,
      "balance_loss_clip": 1.04996204,
      "balance_loss_mlp": 1.01769412,
      "epoch": 0.6396921781999639,
      "flos": 17566315591680.0,
      "grad_norm": 1.8064551682845527,
      "language_loss": 0.7100569,
      "learning_rate": 1.2139908675190454e-06,
      "loss": 0.73184824,
      "num_input_tokens_seen": 114511520,
      "step": 5320,
      "time_per_iteration": 2.5008301734924316
    },
    {
      "auxiliary_loss_clip": 0.01090968,
      "auxiliary_loss_mlp": 0.01028118,
      "balance_loss_clip": 1.03939795,
      "balance_loss_mlp": 1.02067947,
      "epoch": 0.639812421090603,
      "flos": 21251252972160.0,
      "grad_norm": 1.8653591746839606,
      "language_loss": 0.74788666,
      "learning_rate": 1.2132746342781883e-06,
      "loss": 0.76907748,
      "num_input_tokens_seen": 114532680,
      "step": 5321,
      "time_per_iteration": 2.666193962097168
    },
    {
      "auxiliary_loss_clip": 0.01173198,
      "auxiliary_loss_mlp": 0.0103672,
      "balance_loss_clip": 1.05298245,
      "balance_loss_mlp": 1.02909017,
      "epoch": 0.6399326639812422,
      "flos": 11180967684480.0,
      "grad_norm": 2.2567669373505512,
      "language_loss": 0.79549533,
      "learning_rate": 1.2125585203829442e-06,
      "loss": 0.81759453,
      "num_input_tokens_seen": 114548320,
      "step": 5322,
      "time_per_iteration": 2.4794082641601562
    },
    {
      "auxiliary_loss_clip": 0.01115329,
      "auxiliary_loss_mlp": 0.01025434,
      "balance_loss_clip": 1.04640651,
      "balance_loss_mlp": 1.01800728,
      "epoch": 0.6400529068718812,
      "flos": 23911048195200.0,
      "grad_norm": 1.9052960527467953,
      "language_loss": 0.73815221,
      "learning_rate": 1.211842525941946e-06,
      "loss": 0.75955981,
      "num_input_tokens_seen": 114568115,
      "step": 5323,
      "time_per_iteration": 2.5815987586975098
    },
    {
      "auxiliary_loss_clip": 0.01109879,
      "auxiliary_loss_mlp": 0.01027824,
      "balance_loss_clip": 1.04437089,
      "balance_loss_mlp": 1.02047217,
      "epoch": 0.6401731497625203,
      "flos": 44018724890880.0,
      "grad_norm": 1.7496701544823012,
      "language_loss": 0.78863549,
      "learning_rate": 1.2111266510638105e-06,
      "loss": 0.81001252,
      "num_input_tokens_seen": 114591040,
      "step": 5324,
      "time_per_iteration": 2.809509515762329
    },
    {
      "auxiliary_loss_clip": 0.01090346,
      "auxiliary_loss_mlp": 0.01025716,
      "balance_loss_clip": 1.04070365,
      "balance_loss_mlp": 1.01793206,
      "epoch": 0.6402933926531594,
      "flos": 20662209838080.0,
      "grad_norm": 1.6622761084676105,
      "language_loss": 0.8002643,
      "learning_rate": 1.2104108958571346e-06,
      "loss": 0.82142496,
      "num_input_tokens_seen": 114609310,
      "step": 5325,
      "time_per_iteration": 2.6218109130859375
    },
    {
      "auxiliary_loss_clip": 0.01155688,
      "auxiliary_loss_mlp": 0.01025703,
      "balance_loss_clip": 1.05072904,
      "balance_loss_mlp": 1.01886344,
      "epoch": 0.6404136355437985,
      "flos": 24863327614080.0,
      "grad_norm": 1.4960894967485026,
      "language_loss": 0.75767696,
      "learning_rate": 1.2096952604304975e-06,
      "loss": 0.77949083,
      "num_input_tokens_seen": 114629740,
      "step": 5326,
      "time_per_iteration": 2.5579049587249756
    },
    {
      "auxiliary_loss_clip": 0.01155046,
      "auxiliary_loss_mlp": 0.0102907,
      "balance_loss_clip": 1.0473597,
      "balance_loss_mlp": 1.02133942,
      "epoch": 0.6405338784344375,
      "flos": 40479548901120.0,
      "grad_norm": 2.2826015944220206,
      "language_loss": 0.70150238,
      "learning_rate": 1.2089797448924616e-06,
      "loss": 0.72334361,
      "num_input_tokens_seen": 114653615,
      "step": 5327,
      "time_per_iteration": 2.654397487640381
    },
    {
      "auxiliary_loss_clip": 0.01116342,
      "auxiliary_loss_mlp": 0.01026772,
      "balance_loss_clip": 1.0433836,
      "balance_loss_mlp": 1.01914811,
      "epoch": 0.6406541213250767,
      "flos": 20886041439360.0,
      "grad_norm": 2.646278135927872,
      "language_loss": 0.65968931,
      "learning_rate": 1.2082643493515692e-06,
      "loss": 0.68112046,
      "num_input_tokens_seen": 114671935,
      "step": 5328,
      "time_per_iteration": 2.603133201599121
    },
    {
      "auxiliary_loss_clip": 0.01157849,
      "auxiliary_loss_mlp": 0.01027113,
      "balance_loss_clip": 1.05230606,
      "balance_loss_mlp": 1.01984131,
      "epoch": 0.6407743642157158,
      "flos": 23295970679040.0,
      "grad_norm": 2.6633337354509594,
      "language_loss": 0.81659806,
      "learning_rate": 1.207549073916346e-06,
      "loss": 0.83844763,
      "num_input_tokens_seen": 114692870,
      "step": 5329,
      "time_per_iteration": 2.5119776725769043
    },
    {
      "auxiliary_loss_clip": 0.01134814,
      "auxiliary_loss_mlp": 0.01028181,
      "balance_loss_clip": 1.04895616,
      "balance_loss_mlp": 1.02123451,
      "epoch": 0.6408946071063548,
      "flos": 15012636122880.0,
      "grad_norm": 2.026905653638198,
      "language_loss": 0.77309245,
      "learning_rate": 1.2068339186952976e-06,
      "loss": 0.79472244,
      "num_input_tokens_seen": 114710410,
      "step": 5330,
      "time_per_iteration": 3.326878547668457
    },
    {
      "auxiliary_loss_clip": 0.01160348,
      "auxiliary_loss_mlp": 0.01028487,
      "balance_loss_clip": 1.05142212,
      "balance_loss_mlp": 1.02049971,
      "epoch": 0.6410148499969939,
      "flos": 22528595496960.0,
      "grad_norm": 1.7485904327909214,
      "language_loss": 0.7268889,
      "learning_rate": 1.2061188837969136e-06,
      "loss": 0.74877715,
      "num_input_tokens_seen": 114730020,
      "step": 5331,
      "time_per_iteration": 2.4940669536590576
    },
    {
      "auxiliary_loss_clip": 0.01121398,
      "auxiliary_loss_mlp": 0.01030013,
      "balance_loss_clip": 1.04219723,
      "balance_loss_mlp": 1.02206802,
      "epoch": 0.641135092887633,
      "flos": 12422004537600.0,
      "grad_norm": 3.0693600279795263,
      "language_loss": 0.84271312,
      "learning_rate": 1.2054039693296631e-06,
      "loss": 0.86422724,
      "num_input_tokens_seen": 114748015,
      "step": 5332,
      "time_per_iteration": 2.5751919746398926
    },
    {
      "auxiliary_loss_clip": 0.01120564,
      "auxiliary_loss_mlp": 0.01029037,
      "balance_loss_clip": 1.04354858,
      "balance_loss_mlp": 1.02185714,
      "epoch": 0.6412553357782721,
      "flos": 22127329687680.0,
      "grad_norm": 1.694619781669432,
      "language_loss": 0.8184185,
      "learning_rate": 1.2046891754019992e-06,
      "loss": 0.83991456,
      "num_input_tokens_seen": 114768625,
      "step": 5333,
      "time_per_iteration": 2.5821332931518555
    },
    {
      "auxiliary_loss_clip": 0.01160501,
      "auxiliary_loss_mlp": 0.01027716,
      "balance_loss_clip": 1.05245495,
      "balance_loss_mlp": 1.02010703,
      "epoch": 0.6413755786689112,
      "flos": 15888605097600.0,
      "grad_norm": 2.0720602602752347,
      "language_loss": 0.82326102,
      "learning_rate": 1.2039745021223548e-06,
      "loss": 0.8451432,
      "num_input_tokens_seen": 114786045,
      "step": 5334,
      "time_per_iteration": 2.4994943141937256
    },
    {
      "auxiliary_loss_clip": 0.01019736,
      "auxiliary_loss_mlp": 0.01002474,
      "balance_loss_clip": 1.01732373,
      "balance_loss_mlp": 1.00120413,
      "epoch": 0.6414958215595503,
      "flos": 68039159955840.0,
      "grad_norm": 0.7952841622587897,
      "language_loss": 0.5706771,
      "learning_rate": 1.2032599495991456e-06,
      "loss": 0.59089917,
      "num_input_tokens_seen": 114850785,
      "step": 5335,
      "time_per_iteration": 3.274405002593994
    },
    {
      "auxiliary_loss_clip": 0.01160656,
      "auxiliary_loss_mlp": 0.01027123,
      "balance_loss_clip": 1.05311525,
      "balance_loss_mlp": 1.0192138,
      "epoch": 0.6416160644501894,
      "flos": 44091300320640.0,
      "grad_norm": 1.7944699092836465,
      "language_loss": 0.69842386,
      "learning_rate": 1.2025455179407685e-06,
      "loss": 0.72030163,
      "num_input_tokens_seen": 114871945,
      "step": 5336,
      "time_per_iteration": 2.732163667678833
    },
    {
      "auxiliary_loss_clip": 0.01152765,
      "auxiliary_loss_mlp": 0.00761228,
      "balance_loss_clip": 1.04901671,
      "balance_loss_mlp": 1.0002048,
      "epoch": 0.6417363073408284,
      "flos": 20959837931520.0,
      "grad_norm": 1.776341706315445,
      "language_loss": 0.73793161,
      "learning_rate": 1.2018312072556022e-06,
      "loss": 0.75707155,
      "num_input_tokens_seen": 114890445,
      "step": 5337,
      "time_per_iteration": 2.5288796424865723
    },
    {
      "auxiliary_loss_clip": 0.01169635,
      "auxiliary_loss_mlp": 0.00761228,
      "balance_loss_clip": 1.05173528,
      "balance_loss_mlp": 1.00021267,
      "epoch": 0.6418565502314676,
      "flos": 22455122227200.0,
      "grad_norm": 2.3045715371385223,
      "language_loss": 0.74091148,
      "learning_rate": 1.2011170176520077e-06,
      "loss": 0.76022011,
      "num_input_tokens_seen": 114911360,
      "step": 5338,
      "time_per_iteration": 3.2635276317596436
    },
    {
      "auxiliary_loss_clip": 0.01086764,
      "auxiliary_loss_mlp": 0.01033096,
      "balance_loss_clip": 1.04205465,
      "balance_loss_mlp": 1.02581263,
      "epoch": 0.6419767931221066,
      "flos": 25045502417280.0,
      "grad_norm": 1.4722780255137728,
      "language_loss": 0.81080997,
      "learning_rate": 1.2004029492383256e-06,
      "loss": 0.83200854,
      "num_input_tokens_seen": 114932700,
      "step": 5339,
      "time_per_iteration": 2.6622889041900635
    },
    {
      "auxiliary_loss_clip": 0.01156906,
      "auxiliary_loss_mlp": 0.01026789,
      "balance_loss_clip": 1.05168462,
      "balance_loss_mlp": 1.01901627,
      "epoch": 0.6420970360127457,
      "flos": 19463691709440.0,
      "grad_norm": 1.845981060939152,
      "language_loss": 0.7378515,
      "learning_rate": 1.1996890021228814e-06,
      "loss": 0.75968844,
      "num_input_tokens_seen": 114949475,
      "step": 5340,
      "time_per_iteration": 3.3060555458068848
    },
    {
      "auxiliary_loss_clip": 0.01137858,
      "auxiliary_loss_mlp": 0.01025032,
      "balance_loss_clip": 1.04565334,
      "balance_loss_mlp": 1.01773071,
      "epoch": 0.6422172789033849,
      "flos": 40406147458560.0,
      "grad_norm": 1.4828407652835038,
      "language_loss": 0.69936657,
      "learning_rate": 1.1989751764139785e-06,
      "loss": 0.72099543,
      "num_input_tokens_seen": 114973125,
      "step": 5341,
      "time_per_iteration": 2.7215371131896973
    },
    {
      "auxiliary_loss_clip": 0.0111086,
      "auxiliary_loss_mlp": 0.01020855,
      "balance_loss_clip": 1.04008222,
      "balance_loss_mlp": 1.01370251,
      "epoch": 0.6423375217940239,
      "flos": 27672870637440.0,
      "grad_norm": 1.7145142539739229,
      "language_loss": 0.8339873,
      "learning_rate": 1.1982614722199044e-06,
      "loss": 0.85530442,
      "num_input_tokens_seen": 114994300,
      "step": 5342,
      "time_per_iteration": 3.3978445529937744
    },
    {
      "auxiliary_loss_clip": 0.01145292,
      "auxiliary_loss_mlp": 0.01024699,
      "balance_loss_clip": 1.04703283,
      "balance_loss_mlp": 1.01783252,
      "epoch": 0.642457764684663,
      "flos": 18369242259840.0,
      "grad_norm": 2.0713618399692355,
      "language_loss": 0.77812189,
      "learning_rate": 1.1975478896489276e-06,
      "loss": 0.79982173,
      "num_input_tokens_seen": 115012135,
      "step": 5343,
      "time_per_iteration": 2.5136735439300537
    },
    {
      "auxiliary_loss_clip": 0.01169001,
      "auxiliary_loss_mlp": 0.01024328,
      "balance_loss_clip": 1.05041587,
      "balance_loss_mlp": 1.0172919,
      "epoch": 0.6425780075753021,
      "flos": 19750509809280.0,
      "grad_norm": 1.8741224024659697,
      "language_loss": 0.76482832,
      "learning_rate": 1.1968344288092981e-06,
      "loss": 0.78676164,
      "num_input_tokens_seen": 115028715,
      "step": 5344,
      "time_per_iteration": 2.476367473602295
    },
    {
      "auxiliary_loss_clip": 0.01157202,
      "auxiliary_loss_mlp": 0.00761397,
      "balance_loss_clip": 1.05154371,
      "balance_loss_mlp": 1.00023222,
      "epoch": 0.6426982504659412,
      "flos": 20558536208640.0,
      "grad_norm": 1.7183875266634723,
      "language_loss": 0.64680231,
      "learning_rate": 1.1961210898092468e-06,
      "loss": 0.66598833,
      "num_input_tokens_seen": 115047665,
      "step": 5345,
      "time_per_iteration": 2.5369062423706055
    },
    {
      "auxiliary_loss_clip": 0.01152069,
      "auxiliary_loss_mlp": 0.01029582,
      "balance_loss_clip": 1.05145597,
      "balance_loss_mlp": 1.02212226,
      "epoch": 0.6428184933565803,
      "flos": 17851984456320.0,
      "grad_norm": 2.335823856246759,
      "language_loss": 0.79095185,
      "learning_rate": 1.1954078727569874e-06,
      "loss": 0.81276834,
      "num_input_tokens_seen": 115064965,
      "step": 5346,
      "time_per_iteration": 2.5327959060668945
    },
    {
      "auxiliary_loss_clip": 0.01133006,
      "auxiliary_loss_mlp": 0.00760958,
      "balance_loss_clip": 1.04581511,
      "balance_loss_mlp": 1.00024271,
      "epoch": 0.6429387362472194,
      "flos": 22456953820800.0,
      "grad_norm": 1.7657001314341272,
      "language_loss": 0.77960992,
      "learning_rate": 1.1946947777607141e-06,
      "loss": 0.79854953,
      "num_input_tokens_seen": 115086100,
      "step": 5347,
      "time_per_iteration": 2.5912585258483887
    },
    {
      "auxiliary_loss_clip": 0.0110736,
      "auxiliary_loss_mlp": 0.0102867,
      "balance_loss_clip": 1.04359448,
      "balance_loss_mlp": 1.02148438,
      "epoch": 0.6430589791378585,
      "flos": 24752579005440.0,
      "grad_norm": 1.8490055534702332,
      "language_loss": 0.79921544,
      "learning_rate": 1.1939818049286024e-06,
      "loss": 0.82057571,
      "num_input_tokens_seen": 115104260,
      "step": 5348,
      "time_per_iteration": 2.6298482418060303
    },
    {
      "auxiliary_loss_clip": 0.0109005,
      "auxiliary_loss_mlp": 0.01032197,
      "balance_loss_clip": 1.04339373,
      "balance_loss_mlp": 1.02480912,
      "epoch": 0.6431792220284975,
      "flos": 24901249397760.0,
      "grad_norm": 1.515994914793429,
      "language_loss": 0.75695705,
      "learning_rate": 1.1932689543688101e-06,
      "loss": 0.77817953,
      "num_input_tokens_seen": 115125365,
      "step": 5349,
      "time_per_iteration": 2.649275779724121
    },
    {
      "auxiliary_loss_clip": 0.01141227,
      "auxiliary_loss_mlp": 0.01026526,
      "balance_loss_clip": 1.04952312,
      "balance_loss_mlp": 1.01851511,
      "epoch": 0.6432994649191367,
      "flos": 21032305620480.0,
      "grad_norm": 1.7424819569520504,
      "language_loss": 0.72868478,
      "learning_rate": 1.1925562261894756e-06,
      "loss": 0.75036228,
      "num_input_tokens_seen": 115144445,
      "step": 5350,
      "time_per_iteration": 2.5542449951171875
    },
    {
      "auxiliary_loss_clip": 0.01137557,
      "auxiliary_loss_mlp": 0.01031016,
      "balance_loss_clip": 1.04685187,
      "balance_loss_mlp": 1.02422118,
      "epoch": 0.6434197078097758,
      "flos": 30884433655680.0,
      "grad_norm": 1.8190494357330105,
      "language_loss": 0.77534831,
      "learning_rate": 1.1918436204987207e-06,
      "loss": 0.79703408,
      "num_input_tokens_seen": 115166305,
      "step": 5351,
      "time_per_iteration": 2.6096158027648926
    },
    {
      "auxiliary_loss_clip": 0.01150767,
      "auxiliary_loss_mlp": 0.01025251,
      "balance_loss_clip": 1.05028403,
      "balance_loss_mlp": 1.01829529,
      "epoch": 0.6435399507004148,
      "flos": 15012492468480.0,
      "grad_norm": 2.2447958918494764,
      "language_loss": 0.81698859,
      "learning_rate": 1.191131137404645e-06,
      "loss": 0.83874869,
      "num_input_tokens_seen": 115183045,
      "step": 5352,
      "time_per_iteration": 2.489264965057373
    },
    {
      "auxiliary_loss_clip": 0.01117981,
      "auxiliary_loss_mlp": 0.01033293,
      "balance_loss_clip": 1.04570544,
      "balance_loss_mlp": 1.02562809,
      "epoch": 0.643660193591054,
      "flos": 19901981462400.0,
      "grad_norm": 2.017726852567579,
      "language_loss": 0.76927447,
      "learning_rate": 1.190418777015333e-06,
      "loss": 0.79078728,
      "num_input_tokens_seen": 115201955,
      "step": 5353,
      "time_per_iteration": 2.5625014305114746
    },
    {
      "auxiliary_loss_clip": 0.0114126,
      "auxiliary_loss_mlp": 0.01029248,
      "balance_loss_clip": 1.04857409,
      "balance_loss_mlp": 1.02254891,
      "epoch": 0.643780436481693,
      "flos": 24133622820480.0,
      "grad_norm": 1.3675177501409728,
      "language_loss": 0.73383647,
      "learning_rate": 1.1897065394388487e-06,
      "loss": 0.75554156,
      "num_input_tokens_seen": 115222395,
      "step": 5354,
      "time_per_iteration": 2.565554141998291
    },
    {
      "auxiliary_loss_clip": 0.01149425,
      "auxiliary_loss_mlp": 0.01041789,
      "balance_loss_clip": 1.0548408,
      "balance_loss_mlp": 1.03396893,
      "epoch": 0.6439006793723321,
      "flos": 23148808657920.0,
      "grad_norm": 1.5874281578699188,
      "language_loss": 0.76489425,
      "learning_rate": 1.1889944247832385e-06,
      "loss": 0.78680634,
      "num_input_tokens_seen": 115242635,
      "step": 5355,
      "time_per_iteration": 2.564833164215088
    },
    {
      "auxiliary_loss_clip": 0.01157221,
      "auxiliary_loss_mlp": 0.01025205,
      "balance_loss_clip": 1.04738903,
      "balance_loss_mlp": 1.01777887,
      "epoch": 0.6440209222629713,
      "flos": 23617909301760.0,
      "grad_norm": 1.7941243661368356,
      "language_loss": 0.70818055,
      "learning_rate": 1.1882824331565283e-06,
      "loss": 0.73000479,
      "num_input_tokens_seen": 115262095,
      "step": 5356,
      "time_per_iteration": 3.39786696434021
    },
    {
      "auxiliary_loss_clip": 0.01121581,
      "auxiliary_loss_mlp": 0.01028567,
      "balance_loss_clip": 1.04295182,
      "balance_loss_mlp": 1.02131927,
      "epoch": 0.6441411651536103,
      "flos": 16544872535040.0,
      "grad_norm": 2.0687142856011964,
      "language_loss": 0.89636707,
      "learning_rate": 1.1875705646667287e-06,
      "loss": 0.91786861,
      "num_input_tokens_seen": 115279985,
      "step": 5357,
      "time_per_iteration": 2.563932418823242
    },
    {
      "auxiliary_loss_clip": 0.01151178,
      "auxiliary_loss_mlp": 0.01027769,
      "balance_loss_clip": 1.04577005,
      "balance_loss_mlp": 1.02047598,
      "epoch": 0.6442614080442494,
      "flos": 25410965345280.0,
      "grad_norm": 2.047804123698494,
      "language_loss": 0.75786555,
      "learning_rate": 1.1868588194218282e-06,
      "loss": 0.77965498,
      "num_input_tokens_seen": 115300365,
      "step": 5358,
      "time_per_iteration": 2.5637528896331787
    },
    {
      "auxiliary_loss_clip": 0.01149667,
      "auxiliary_loss_mlp": 0.01032074,
      "balance_loss_clip": 1.04955506,
      "balance_loss_mlp": 1.0250349,
      "epoch": 0.6443816509348885,
      "flos": 28294017552000.0,
      "grad_norm": 1.5878208440473687,
      "language_loss": 0.74121487,
      "learning_rate": 1.1861471975297979e-06,
      "loss": 0.76303232,
      "num_input_tokens_seen": 115322060,
      "step": 5359,
      "time_per_iteration": 2.6340930461883545
    },
    {
      "auxiliary_loss_clip": 0.01125735,
      "auxiliary_loss_mlp": 0.01027455,
      "balance_loss_clip": 1.04814744,
      "balance_loss_mlp": 1.01983142,
      "epoch": 0.6445018938255276,
      "flos": 36690075964800.0,
      "grad_norm": 1.565260689621832,
      "language_loss": 0.70743787,
      "learning_rate": 1.185435699098591e-06,
      "loss": 0.72896975,
      "num_input_tokens_seen": 115348255,
      "step": 5360,
      "time_per_iteration": 2.7744717597961426
    },
    {
      "auxiliary_loss_clip": 0.01144947,
      "auxiliary_loss_mlp": 0.01033021,
      "balance_loss_clip": 1.04786217,
      "balance_loss_mlp": 1.02510548,
      "epoch": 0.6446221367161666,
      "flos": 14501411804160.0,
      "grad_norm": 2.266588614968,
      "language_loss": 0.78247392,
      "learning_rate": 1.1847243242361403e-06,
      "loss": 0.80425358,
      "num_input_tokens_seen": 115366845,
      "step": 5361,
      "time_per_iteration": 2.5077455043792725
    },
    {
      "auxiliary_loss_clip": 0.01140416,
      "auxiliary_loss_mlp": 0.01024999,
      "balance_loss_clip": 1.04631257,
      "balance_loss_mlp": 1.0175159,
      "epoch": 0.6447423796068057,
      "flos": 24609367480320.0,
      "grad_norm": 1.8848428232194765,
      "language_loss": 0.77935326,
      "learning_rate": 1.1840130730503624e-06,
      "loss": 0.80100745,
      "num_input_tokens_seen": 115388125,
      "step": 5362,
      "time_per_iteration": 2.636784791946411
    },
    {
      "auxiliary_loss_clip": 0.01172448,
      "auxiliary_loss_mlp": 0.01025049,
      "balance_loss_clip": 1.05288577,
      "balance_loss_mlp": 1.01820993,
      "epoch": 0.6448626224974449,
      "flos": 25047298097280.0,
      "grad_norm": 2.1675308976534486,
      "language_loss": 0.75112426,
      "learning_rate": 1.1833019456491518e-06,
      "loss": 0.7730993,
      "num_input_tokens_seen": 115409655,
      "step": 5363,
      "time_per_iteration": 2.4969983100891113
    },
    {
      "auxiliary_loss_clip": 0.01155973,
      "auxiliary_loss_mlp": 0.01028224,
      "balance_loss_clip": 1.0505898,
      "balance_loss_mlp": 1.02075005,
      "epoch": 0.6449828653880839,
      "flos": 22530355263360.0,
      "grad_norm": 2.6162378884423942,
      "language_loss": 0.78477418,
      "learning_rate": 1.1825909421403871e-06,
      "loss": 0.80661619,
      "num_input_tokens_seen": 115428750,
      "step": 5364,
      "time_per_iteration": 3.3288347721099854
    },
    {
      "auxiliary_loss_clip": 0.01156662,
      "auxiliary_loss_mlp": 0.01032105,
      "balance_loss_clip": 1.04946232,
      "balance_loss_mlp": 1.02486312,
      "epoch": 0.645103108278723,
      "flos": 25695736369920.0,
      "grad_norm": 1.6964544267802024,
      "language_loss": 0.76390147,
      "learning_rate": 1.181880062631926e-06,
      "loss": 0.78578913,
      "num_input_tokens_seen": 115448085,
      "step": 5365,
      "time_per_iteration": 2.5154478549957275
    },
    {
      "auxiliary_loss_clip": 0.01134966,
      "auxiliary_loss_mlp": 0.01029327,
      "balance_loss_clip": 1.04704237,
      "balance_loss_mlp": 1.02197433,
      "epoch": 0.6452233511693621,
      "flos": 27450331925760.0,
      "grad_norm": 1.9628322367629005,
      "language_loss": 0.84476554,
      "learning_rate": 1.1811693072316093e-06,
      "loss": 0.86640847,
      "num_input_tokens_seen": 115465765,
      "step": 5366,
      "time_per_iteration": 3.3790407180786133
    },
    {
      "auxiliary_loss_clip": 0.01169515,
      "auxiliary_loss_mlp": 0.00761505,
      "balance_loss_clip": 1.05032754,
      "balance_loss_mlp": 1.00021553,
      "epoch": 0.6453435940600012,
      "flos": 19208618254080.0,
      "grad_norm": 2.4848752931977396,
      "language_loss": 0.84082818,
      "learning_rate": 1.1804586760472574e-06,
      "loss": 0.86013836,
      "num_input_tokens_seen": 115482230,
      "step": 5367,
      "time_per_iteration": 3.2257301807403564
    },
    {
      "auxiliary_loss_clip": 0.01124829,
      "auxiliary_loss_mlp": 0.01024236,
      "balance_loss_clip": 1.04410255,
      "balance_loss_mlp": 1.01684546,
      "epoch": 0.6454638369506402,
      "flos": 25737680476800.0,
      "grad_norm": 2.0650440432238377,
      "language_loss": 0.80448234,
      "learning_rate": 1.1797481691866736e-06,
      "loss": 0.82597297,
      "num_input_tokens_seen": 115499455,
      "step": 5368,
      "time_per_iteration": 2.5998034477233887
    },
    {
      "auxiliary_loss_clip": 0.01130752,
      "auxiliary_loss_mlp": 0.01029606,
      "balance_loss_clip": 1.04577196,
      "balance_loss_mlp": 1.02198291,
      "epoch": 0.6455840798412794,
      "flos": 20989176364800.0,
      "grad_norm": 1.9576451787498792,
      "language_loss": 0.83128327,
      "learning_rate": 1.1790377867576393e-06,
      "loss": 0.8528868,
      "num_input_tokens_seen": 115517205,
      "step": 5369,
      "time_per_iteration": 2.5251669883728027
    },
    {
      "auxiliary_loss_clip": 0.01144125,
      "auxiliary_loss_mlp": 0.01034692,
      "balance_loss_clip": 1.04604614,
      "balance_loss_mlp": 1.02761149,
      "epoch": 0.6457043227319185,
      "flos": 26067556005120.0,
      "grad_norm": 1.8728559089025174,
      "language_loss": 0.76564741,
      "learning_rate": 1.1783275288679203e-06,
      "loss": 0.78743559,
      "num_input_tokens_seen": 115534370,
      "step": 5370,
      "time_per_iteration": 2.580638885498047
    },
    {
      "auxiliary_loss_clip": 0.010534,
      "auxiliary_loss_mlp": 0.01002279,
      "balance_loss_clip": 1.01363826,
      "balance_loss_mlp": 1.00117671,
      "epoch": 0.6458245656225575,
      "flos": 60370831088640.0,
      "grad_norm": 0.840903151825484,
      "language_loss": 0.57098281,
      "learning_rate": 1.177617395625262e-06,
      "loss": 0.59153962,
      "num_input_tokens_seen": 115592345,
      "step": 5371,
      "time_per_iteration": 3.059608221054077
    },
    {
      "auxiliary_loss_clip": 0.01156268,
      "auxiliary_loss_mlp": 0.01026136,
      "balance_loss_clip": 1.04993773,
      "balance_loss_mlp": 1.01873875,
      "epoch": 0.6459448085131967,
      "flos": 23076771932160.0,
      "grad_norm": 1.7923630342951629,
      "language_loss": 0.75284189,
      "learning_rate": 1.1769073871373908e-06,
      "loss": 0.77466595,
      "num_input_tokens_seen": 115612550,
      "step": 5372,
      "time_per_iteration": 2.5448827743530273
    },
    {
      "auxiliary_loss_clip": 0.01122155,
      "auxiliary_loss_mlp": 0.01025524,
      "balance_loss_clip": 1.0416826,
      "balance_loss_mlp": 1.01822567,
      "epoch": 0.6460650514038357,
      "flos": 22598190097920.0,
      "grad_norm": 2.4945505635053427,
      "language_loss": 0.83891279,
      "learning_rate": 1.176197503512015e-06,
      "loss": 0.86038959,
      "num_input_tokens_seen": 115632265,
      "step": 5373,
      "time_per_iteration": 2.5889172554016113
    },
    {
      "auxiliary_loss_clip": 0.01138557,
      "auxiliary_loss_mlp": 0.01029055,
      "balance_loss_clip": 1.04721987,
      "balance_loss_mlp": 1.0223701,
      "epoch": 0.6461852942944748,
      "flos": 20266726118400.0,
      "grad_norm": 2.83986386428047,
      "language_loss": 0.82229161,
      "learning_rate": 1.1754877448568223e-06,
      "loss": 0.84396774,
      "num_input_tokens_seen": 115651720,
      "step": 5374,
      "time_per_iteration": 2.549473524093628
    },
    {
      "auxiliary_loss_clip": 0.01142066,
      "auxiliary_loss_mlp": 0.01023371,
      "balance_loss_clip": 1.04834247,
      "balance_loss_mlp": 1.01625133,
      "epoch": 0.646305537185114,
      "flos": 23367109564800.0,
      "grad_norm": 1.88125963598905,
      "language_loss": 0.90054625,
      "learning_rate": 1.1747781112794837e-06,
      "loss": 0.92220062,
      "num_input_tokens_seen": 115668215,
      "step": 5375,
      "time_per_iteration": 2.5463366508483887
    },
    {
      "auxiliary_loss_clip": 0.01123557,
      "auxiliary_loss_mlp": 0.0103128,
      "balance_loss_clip": 1.04590499,
      "balance_loss_mlp": 1.02451825,
      "epoch": 0.646425780075753,
      "flos": 24277480790400.0,
      "grad_norm": 1.7577051076550172,
      "language_loss": 0.83150899,
      "learning_rate": 1.1740686028876487e-06,
      "loss": 0.85305738,
      "num_input_tokens_seen": 115687080,
      "step": 5376,
      "time_per_iteration": 2.6011385917663574
    },
    {
      "auxiliary_loss_clip": 0.01153916,
      "auxiliary_loss_mlp": 0.01025795,
      "balance_loss_clip": 1.05021322,
      "balance_loss_mlp": 1.01888382,
      "epoch": 0.6465460229663921,
      "flos": 20813968800000.0,
      "grad_norm": 2.04859452399454,
      "language_loss": 0.74640608,
      "learning_rate": 1.1733592197889507e-06,
      "loss": 0.76820314,
      "num_input_tokens_seen": 115703990,
      "step": 5377,
      "time_per_iteration": 2.48874568939209
    },
    {
      "auxiliary_loss_clip": 0.01150347,
      "auxiliary_loss_mlp": 0.01025677,
      "balance_loss_clip": 1.05021811,
      "balance_loss_mlp": 1.0192188,
      "epoch": 0.6466662658570312,
      "flos": 22853299466880.0,
      "grad_norm": 1.944513847269682,
      "language_loss": 0.7237671,
      "learning_rate": 1.1726499620910014e-06,
      "loss": 0.74552733,
      "num_input_tokens_seen": 115724270,
      "step": 5378,
      "time_per_iteration": 2.5738139152526855
    },
    {
      "auxiliary_loss_clip": 0.0115401,
      "auxiliary_loss_mlp": 0.01028624,
      "balance_loss_clip": 1.04959941,
      "balance_loss_mlp": 1.0209707,
      "epoch": 0.6467865087476703,
      "flos": 15304553953920.0,
      "grad_norm": 2.0835119195803498,
      "language_loss": 0.7804997,
      "learning_rate": 1.1719408299013955e-06,
      "loss": 0.80232602,
      "num_input_tokens_seen": 115742995,
      "step": 5379,
      "time_per_iteration": 2.4694271087646484
    },
    {
      "auxiliary_loss_clip": 0.011681,
      "auxiliary_loss_mlp": 0.01034719,
      "balance_loss_clip": 1.05111098,
      "balance_loss_mlp": 1.02725005,
      "epoch": 0.6469067516383094,
      "flos": 19573650218880.0,
      "grad_norm": 2.748265973259102,
      "language_loss": 0.75764418,
      "learning_rate": 1.1712318233277067e-06,
      "loss": 0.77967232,
      "num_input_tokens_seen": 115762015,
      "step": 5380,
      "time_per_iteration": 2.4712114334106445
    },
    {
      "auxiliary_loss_clip": 0.01050682,
      "auxiliary_loss_mlp": 0.01001491,
      "balance_loss_clip": 1.012797,
      "balance_loss_mlp": 1.0003581,
      "epoch": 0.6470269945289485,
      "flos": 65098002522240.0,
      "grad_norm": 0.7573884803135872,
      "language_loss": 0.57891333,
      "learning_rate": 1.1705229424774916e-06,
      "loss": 0.59943503,
      "num_input_tokens_seen": 115816285,
      "step": 5381,
      "time_per_iteration": 3.771580457687378
    },
    {
      "auxiliary_loss_clip": 0.01138316,
      "auxiliary_loss_mlp": 0.01034499,
      "balance_loss_clip": 1.04495907,
      "balance_loss_mlp": 1.02702153,
      "epoch": 0.6471472374195876,
      "flos": 30696943639680.0,
      "grad_norm": 1.7029346590767969,
      "language_loss": 0.63955659,
      "learning_rate": 1.1698141874582867e-06,
      "loss": 0.66128474,
      "num_input_tokens_seen": 115837330,
      "step": 5382,
      "time_per_iteration": 2.6193666458129883
    },
    {
      "auxiliary_loss_clip": 0.0116766,
      "auxiliary_loss_mlp": 0.01023194,
      "balance_loss_clip": 1.05160201,
      "balance_loss_mlp": 1.01642537,
      "epoch": 0.6472674803102266,
      "flos": 20521835487360.0,
      "grad_norm": 2.0011978223460902,
      "language_loss": 0.7210477,
      "learning_rate": 1.169105558377609e-06,
      "loss": 0.74295622,
      "num_input_tokens_seen": 115857420,
      "step": 5383,
      "time_per_iteration": 2.4630322456359863
    },
    {
      "auxiliary_loss_clip": 0.0110784,
      "auxiliary_loss_mlp": 0.00761117,
      "balance_loss_clip": 1.04874921,
      "balance_loss_mlp": 1.00021172,
      "epoch": 0.6473877232008658,
      "flos": 24715447320960.0,
      "grad_norm": 1.6018677460296296,
      "language_loss": 0.78458279,
      "learning_rate": 1.1683970553429587e-06,
      "loss": 0.80327237,
      "num_input_tokens_seen": 115878875,
      "step": 5384,
      "time_per_iteration": 2.655327558517456
    },
    {
      "auxiliary_loss_clip": 0.01133669,
      "auxiliary_loss_mlp": 0.01028447,
      "balance_loss_clip": 1.04844522,
      "balance_loss_mlp": 1.02101719,
      "epoch": 0.6475079660915048,
      "flos": 15885552441600.0,
      "grad_norm": 1.9621136299950048,
      "language_loss": 0.82528836,
      "learning_rate": 1.1676886784618128e-06,
      "loss": 0.84690952,
      "num_input_tokens_seen": 115895540,
      "step": 5385,
      "time_per_iteration": 2.5378496646881104
    },
    {
      "auxiliary_loss_clip": 0.01154535,
      "auxiliary_loss_mlp": 0.01025949,
      "balance_loss_clip": 1.05010974,
      "balance_loss_mlp": 1.0174849,
      "epoch": 0.6476282089821439,
      "flos": 17381590922880.0,
      "grad_norm": 2.0194653023193334,
      "language_loss": 0.84103811,
      "learning_rate": 1.1669804278416332e-06,
      "loss": 0.86284292,
      "num_input_tokens_seen": 115910265,
      "step": 5386,
      "time_per_iteration": 2.491683006286621
    },
    {
      "auxiliary_loss_clip": 0.01146968,
      "auxiliary_loss_mlp": 0.01031282,
      "balance_loss_clip": 1.0502553,
      "balance_loss_mlp": 1.02423656,
      "epoch": 0.6477484518727831,
      "flos": 20194078861440.0,
      "grad_norm": 2.136906632592689,
      "language_loss": 0.71161228,
      "learning_rate": 1.1662723035898602e-06,
      "loss": 0.73339486,
      "num_input_tokens_seen": 115930025,
      "step": 5387,
      "time_per_iteration": 2.520371437072754
    },
    {
      "auxiliary_loss_clip": 0.011551,
      "auxiliary_loss_mlp": 0.01028264,
      "balance_loss_clip": 1.04968417,
      "balance_loss_mlp": 1.02049172,
      "epoch": 0.6478686947634221,
      "flos": 25410426641280.0,
      "grad_norm": 1.5729017838627806,
      "language_loss": 0.81829178,
      "learning_rate": 1.165564305813915e-06,
      "loss": 0.84012544,
      "num_input_tokens_seen": 115949025,
      "step": 5388,
      "time_per_iteration": 2.5456418991088867
    },
    {
      "auxiliary_loss_clip": 0.01154641,
      "auxiliary_loss_mlp": 0.01026778,
      "balance_loss_clip": 1.04931998,
      "balance_loss_mlp": 1.01965487,
      "epoch": 0.6479889376540612,
      "flos": 20083581648000.0,
      "grad_norm": 1.7666772639743735,
      "language_loss": 0.8065573,
      "learning_rate": 1.1648564346212019e-06,
      "loss": 0.82837141,
      "num_input_tokens_seen": 115968145,
      "step": 5389,
      "time_per_iteration": 2.483328104019165
    },
    {
      "auxiliary_loss_clip": 0.011511,
      "auxiliary_loss_mlp": 0.01029335,
      "balance_loss_clip": 1.05027759,
      "balance_loss_mlp": 1.02178049,
      "epoch": 0.6481091805447003,
      "flos": 26758082039040.0,
      "grad_norm": 1.8870826808943437,
      "language_loss": 0.76030773,
      "learning_rate": 1.164148690119104e-06,
      "loss": 0.78211206,
      "num_input_tokens_seen": 115989425,
      "step": 5390,
      "time_per_iteration": 3.345306873321533
    },
    {
      "auxiliary_loss_clip": 0.01166615,
      "auxiliary_loss_mlp": 0.01025718,
      "balance_loss_clip": 1.04954386,
      "balance_loss_mlp": 1.01864314,
      "epoch": 0.6482294234353394,
      "flos": 23952094462080.0,
      "grad_norm": 1.6891317099687795,
      "language_loss": 0.73865342,
      "learning_rate": 1.163441072414985e-06,
      "loss": 0.76057673,
      "num_input_tokens_seen": 116009630,
      "step": 5391,
      "time_per_iteration": 2.4691548347473145
    },
    {
      "auxiliary_loss_clip": 0.01155966,
      "auxiliary_loss_mlp": 0.01029089,
      "balance_loss_clip": 1.05210853,
      "balance_loss_mlp": 1.02157032,
      "epoch": 0.6483496663259785,
      "flos": 26209833776640.0,
      "grad_norm": 1.9334896599943103,
      "language_loss": 0.69779927,
      "learning_rate": 1.16273358161619e-06,
      "loss": 0.71964985,
      "num_input_tokens_seen": 116029965,
      "step": 5392,
      "time_per_iteration": 3.3189268112182617
    },
    {
      "auxiliary_loss_clip": 0.01146759,
      "auxiliary_loss_mlp": 0.01031508,
      "balance_loss_clip": 1.04942632,
      "balance_loss_mlp": 1.02381659,
      "epoch": 0.6484699092166175,
      "flos": 20922239370240.0,
      "grad_norm": 1.7630131578834347,
      "language_loss": 0.8403827,
      "learning_rate": 1.1620262178300446e-06,
      "loss": 0.86216539,
      "num_input_tokens_seen": 116048580,
      "step": 5393,
      "time_per_iteration": 3.458660840988159
    },
    {
      "auxiliary_loss_clip": 0.01124631,
      "auxiliary_loss_mlp": 0.01023857,
      "balance_loss_clip": 1.04270959,
      "balance_loss_mlp": 1.01620984,
      "epoch": 0.6485901521072567,
      "flos": 33072865678080.0,
      "grad_norm": 1.7327327100683878,
      "language_loss": 0.75736743,
      "learning_rate": 1.1613189811638563e-06,
      "loss": 0.77885234,
      "num_input_tokens_seen": 116070305,
      "step": 5394,
      "time_per_iteration": 2.64713454246521
    },
    {
      "auxiliary_loss_clip": 0.01155944,
      "auxiliary_loss_mlp": 0.01032709,
      "balance_loss_clip": 1.04936719,
      "balance_loss_mlp": 1.02552652,
      "epoch": 0.6487103949978957,
      "flos": 22274060745600.0,
      "grad_norm": 1.5644936170146648,
      "language_loss": 0.77938628,
      "learning_rate": 1.1606118717249117e-06,
      "loss": 0.80127281,
      "num_input_tokens_seen": 116090405,
      "step": 5395,
      "time_per_iteration": 2.5174057483673096
    },
    {
      "auxiliary_loss_clip": 0.0117224,
      "auxiliary_loss_mlp": 0.01026917,
      "balance_loss_clip": 1.05043483,
      "balance_loss_mlp": 1.01934767,
      "epoch": 0.6488306378885348,
      "flos": 22930400010240.0,
      "grad_norm": 1.6827983373764588,
      "language_loss": 0.67827934,
      "learning_rate": 1.1599048896204787e-06,
      "loss": 0.70027089,
      "num_input_tokens_seen": 116110285,
      "step": 5396,
      "time_per_iteration": 2.4828875064849854
    },
    {
      "auxiliary_loss_clip": 0.01126596,
      "auxiliary_loss_mlp": 0.01025089,
      "balance_loss_clip": 1.04460692,
      "balance_loss_mlp": 1.01805007,
      "epoch": 0.648950880779174,
      "flos": 20376110010240.0,
      "grad_norm": 1.672253267194269,
      "language_loss": 0.80777907,
      "learning_rate": 1.1591980349578061e-06,
      "loss": 0.82929593,
      "num_input_tokens_seen": 116128955,
      "step": 5397,
      "time_per_iteration": 2.570786237716675
    },
    {
      "auxiliary_loss_clip": 0.01029758,
      "auxiliary_loss_mlp": 0.01005511,
      "balance_loss_clip": 1.01211357,
      "balance_loss_mlp": 1.00435448,
      "epoch": 0.649071123669813,
      "flos": 59930889310080.0,
      "grad_norm": 0.732164568434597,
      "language_loss": 0.54363608,
      "learning_rate": 1.158491307844123e-06,
      "loss": 0.5639888,
      "num_input_tokens_seen": 116188875,
      "step": 5398,
      "time_per_iteration": 3.1408488750457764
    },
    {
      "auxiliary_loss_clip": 0.01139231,
      "auxiliary_loss_mlp": 0.01024503,
      "balance_loss_clip": 1.04820168,
      "balance_loss_mlp": 1.01740146,
      "epoch": 0.6491913665604521,
      "flos": 20446566537600.0,
      "grad_norm": 1.58154504178718,
      "language_loss": 0.83944654,
      "learning_rate": 1.1577847083866387e-06,
      "loss": 0.86108387,
      "num_input_tokens_seen": 116207910,
      "step": 5399,
      "time_per_iteration": 2.51816725730896
    },
    {
      "auxiliary_loss_clip": 0.0112982,
      "auxiliary_loss_mlp": 0.0102552,
      "balance_loss_clip": 1.04548097,
      "balance_loss_mlp": 1.01805723,
      "epoch": 0.6493116094510912,
      "flos": 16946820702720.0,
      "grad_norm": 1.6907219860000646,
      "language_loss": 0.71921897,
      "learning_rate": 1.1570782366925453e-06,
      "loss": 0.74077231,
      "num_input_tokens_seen": 116226425,
      "step": 5400,
      "time_per_iteration": 2.5119876861572266
    },
    {
      "auxiliary_loss_clip": 0.01141029,
      "auxiliary_loss_mlp": 0.010286,
      "balance_loss_clip": 1.04288018,
      "balance_loss_mlp": 1.02078557,
      "epoch": 0.6494318523417303,
      "flos": 18802935072000.0,
      "grad_norm": 1.5891434170505723,
      "language_loss": 0.75683439,
      "learning_rate": 1.1563718928690132e-06,
      "loss": 0.7785306,
      "num_input_tokens_seen": 116243860,
      "step": 5401,
      "time_per_iteration": 2.5209498405456543
    },
    {
      "auxiliary_loss_clip": 0.01129183,
      "auxiliary_loss_mlp": 0.01030852,
      "balance_loss_clip": 1.04732728,
      "balance_loss_mlp": 1.02300453,
      "epoch": 0.6495520952323693,
      "flos": 18982847318400.0,
      "grad_norm": 2.1338126646907294,
      "language_loss": 0.71291423,
      "learning_rate": 1.1556656770231942e-06,
      "loss": 0.73451453,
      "num_input_tokens_seen": 116260055,
      "step": 5402,
      "time_per_iteration": 2.5493288040161133
    },
    {
      "auxiliary_loss_clip": 0.01154696,
      "auxiliary_loss_mlp": 0.0102483,
      "balance_loss_clip": 1.04731405,
      "balance_loss_mlp": 1.01768029,
      "epoch": 0.6496723381230085,
      "flos": 22745388032640.0,
      "grad_norm": 1.6662418293025814,
      "language_loss": 0.76284921,
      "learning_rate": 1.1549595892622207e-06,
      "loss": 0.78464448,
      "num_input_tokens_seen": 116278825,
      "step": 5403,
      "time_per_iteration": 2.508307456970215
    },
    {
      "auxiliary_loss_clip": 0.01013712,
      "auxiliary_loss_mlp": 0.01000794,
      "balance_loss_clip": 1.01367521,
      "balance_loss_mlp": 0.99973357,
      "epoch": 0.6497925810136476,
      "flos": 62145283887360.0,
      "grad_norm": 0.8243421932327678,
      "language_loss": 0.59035981,
      "learning_rate": 1.1542536296932047e-06,
      "loss": 0.61050487,
      "num_input_tokens_seen": 116342360,
      "step": 5404,
      "time_per_iteration": 3.1260552406311035
    },
    {
      "auxiliary_loss_clip": 0.01133865,
      "auxiliary_loss_mlp": 0.01035449,
      "balance_loss_clip": 1.04579735,
      "balance_loss_mlp": 1.02757549,
      "epoch": 0.6499128239042866,
      "flos": 20156731695360.0,
      "grad_norm": 2.087644935859129,
      "language_loss": 0.69894159,
      "learning_rate": 1.1535477984232414e-06,
      "loss": 0.72063476,
      "num_input_tokens_seen": 116362235,
      "step": 5405,
      "time_per_iteration": 2.5685324668884277
    },
    {
      "auxiliary_loss_clip": 0.01109882,
      "auxiliary_loss_mlp": 0.01023637,
      "balance_loss_clip": 1.03946042,
      "balance_loss_mlp": 1.01650524,
      "epoch": 0.6500330667949258,
      "flos": 24462420940800.0,
      "grad_norm": 1.8103330189807276,
      "language_loss": 0.76580441,
      "learning_rate": 1.152842095559404e-06,
      "loss": 0.78713965,
      "num_input_tokens_seen": 116382895,
      "step": 5406,
      "time_per_iteration": 2.6338613033294678
    },
    {
      "auxiliary_loss_clip": 0.01143032,
      "auxiliary_loss_mlp": 0.01025991,
      "balance_loss_clip": 1.04552042,
      "balance_loss_mlp": 1.01933038,
      "epoch": 0.6501533096855648,
      "flos": 25477399549440.0,
      "grad_norm": 1.6071461418306918,
      "language_loss": 0.76477838,
      "learning_rate": 1.1521365212087474e-06,
      "loss": 0.78646863,
      "num_input_tokens_seen": 116402880,
      "step": 5407,
      "time_per_iteration": 3.3873884677886963
    },
    {
      "auxiliary_loss_clip": 0.01154704,
      "auxiliary_loss_mlp": 0.01027634,
      "balance_loss_clip": 1.04838824,
      "balance_loss_mlp": 1.02023101,
      "epoch": 0.6502735525762039,
      "flos": 44819245347840.0,
      "grad_norm": 1.7393072622893384,
      "language_loss": 0.70943433,
      "learning_rate": 1.1514310754783062e-06,
      "loss": 0.73125768,
      "num_input_tokens_seen": 116425830,
      "step": 5408,
      "time_per_iteration": 2.697362184524536
    },
    {
      "auxiliary_loss_clip": 0.01143671,
      "auxiliary_loss_mlp": 0.01027476,
      "balance_loss_clip": 1.04888296,
      "balance_loss_mlp": 1.02031779,
      "epoch": 0.6503937954668431,
      "flos": 28658546726400.0,
      "grad_norm": 1.9356839266032662,
      "language_loss": 0.73135942,
      "learning_rate": 1.1507257584750964e-06,
      "loss": 0.75307095,
      "num_input_tokens_seen": 116446010,
      "step": 5409,
      "time_per_iteration": 2.577609062194824
    },
    {
      "auxiliary_loss_clip": 0.01169387,
      "auxiliary_loss_mlp": 0.01027761,
      "balance_loss_clip": 1.05169737,
      "balance_loss_mlp": 1.0200038,
      "epoch": 0.6505140383574821,
      "flos": 20922562592640.0,
      "grad_norm": 1.8243131288719279,
      "language_loss": 0.77342844,
      "learning_rate": 1.150020570306113e-06,
      "loss": 0.7953999,
      "num_input_tokens_seen": 116465150,
      "step": 5410,
      "time_per_iteration": 2.46199631690979
    },
    {
      "auxiliary_loss_clip": 0.01133742,
      "auxiliary_loss_mlp": 0.01026627,
      "balance_loss_clip": 1.0427382,
      "balance_loss_mlp": 1.01870835,
      "epoch": 0.6506342812481212,
      "flos": 20595236929920.0,
      "grad_norm": 1.8752421526738936,
      "language_loss": 0.74819291,
      "learning_rate": 1.1493155110783338e-06,
      "loss": 0.76979661,
      "num_input_tokens_seen": 116483675,
      "step": 5411,
      "time_per_iteration": 2.506695508956909
    },
    {
      "auxiliary_loss_clip": 0.01155279,
      "auxiliary_loss_mlp": 0.01029054,
      "balance_loss_clip": 1.04927921,
      "balance_loss_mlp": 1.02131724,
      "epoch": 0.6507545241387603,
      "flos": 30226478279040.0,
      "grad_norm": 1.9712439333586247,
      "language_loss": 0.70351505,
      "learning_rate": 1.1486105808987155e-06,
      "loss": 0.72535837,
      "num_input_tokens_seen": 116505165,
      "step": 5412,
      "time_per_iteration": 2.5535600185394287
    },
    {
      "auxiliary_loss_clip": 0.01156513,
      "auxiliary_loss_mlp": 0.0103346,
      "balance_loss_clip": 1.05036104,
      "balance_loss_mlp": 1.02601564,
      "epoch": 0.6508747670293994,
      "flos": 17128241320320.0,
      "grad_norm": 1.8008779568146736,
      "language_loss": 0.81114495,
      "learning_rate": 1.1479057798741947e-06,
      "loss": 0.83304471,
      "num_input_tokens_seen": 116523220,
      "step": 5413,
      "time_per_iteration": 2.4583325386047363
    },
    {
      "auxiliary_loss_clip": 0.01034833,
      "auxiliary_loss_mlp": 0.01007086,
      "balance_loss_clip": 1.01075816,
      "balance_loss_mlp": 1.00601327,
      "epoch": 0.6509950099200384,
      "flos": 68559826573440.0,
      "grad_norm": 0.7860396829514378,
      "language_loss": 0.53366607,
      "learning_rate": 1.14720110811169e-06,
      "loss": 0.55408525,
      "num_input_tokens_seen": 116580450,
      "step": 5414,
      "time_per_iteration": 3.122953414916992
    },
    {
      "auxiliary_loss_clip": 0.01160988,
      "auxiliary_loss_mlp": 0.01026675,
      "balance_loss_clip": 1.05307531,
      "balance_loss_mlp": 1.01900434,
      "epoch": 0.6511152528106776,
      "flos": 22347462188160.0,
      "grad_norm": 2.2926688040353382,
      "language_loss": 0.76902199,
      "learning_rate": 1.146496565718098e-06,
      "loss": 0.79089868,
      "num_input_tokens_seen": 116601020,
      "step": 5415,
      "time_per_iteration": 2.5184547901153564
    },
    {
      "auxiliary_loss_clip": 0.01139415,
      "auxiliary_loss_mlp": 0.01027309,
      "balance_loss_clip": 1.04830551,
      "balance_loss_mlp": 1.01961088,
      "epoch": 0.6512354957013167,
      "flos": 20522158709760.0,
      "grad_norm": 1.9195947343853785,
      "language_loss": 0.75613767,
      "learning_rate": 1.1457921528002996e-06,
      "loss": 0.77780485,
      "num_input_tokens_seen": 116619455,
      "step": 5416,
      "time_per_iteration": 3.29231858253479
    },
    {
      "auxiliary_loss_clip": 0.01169698,
      "auxiliary_loss_mlp": 0.0076143,
      "balance_loss_clip": 1.0506258,
      "balance_loss_mlp": 1.00020623,
      "epoch": 0.6513557385919557,
      "flos": 32337342881280.0,
      "grad_norm": 6.078759950101584,
      "language_loss": 0.72211665,
      "learning_rate": 1.1450878694651522e-06,
      "loss": 0.7414279,
      "num_input_tokens_seen": 116640020,
      "step": 5417,
      "time_per_iteration": 2.5563583374023438
    },
    {
      "auxiliary_loss_clip": 0.0111162,
      "auxiliary_loss_mlp": 0.0102937,
      "balance_loss_clip": 1.04125071,
      "balance_loss_mlp": 1.02232468,
      "epoch": 0.6514759814825949,
      "flos": 12093206417280.0,
      "grad_norm": 2.760350890286075,
      "language_loss": 0.63058913,
      "learning_rate": 1.1443837158194954e-06,
      "loss": 0.651999,
      "num_input_tokens_seen": 116655165,
      "step": 5418,
      "time_per_iteration": 3.373089075088501
    },
    {
      "auxiliary_loss_clip": 0.01128283,
      "auxiliary_loss_mlp": 0.01035143,
      "balance_loss_clip": 1.04861832,
      "balance_loss_mlp": 1.02765083,
      "epoch": 0.651596224373234,
      "flos": 22526907557760.0,
      "grad_norm": 1.7389972844986408,
      "language_loss": 0.74407196,
      "learning_rate": 1.1436796919701484e-06,
      "loss": 0.76570618,
      "num_input_tokens_seen": 116673880,
      "step": 5419,
      "time_per_iteration": 3.258052349090576
    },
    {
      "auxiliary_loss_clip": 0.01139662,
      "auxiliary_loss_mlp": 0.0103273,
      "balance_loss_clip": 1.04817224,
      "balance_loss_mlp": 1.02555394,
      "epoch": 0.651716467263873,
      "flos": 27818955250560.0,
      "grad_norm": 1.7637940008279633,
      "language_loss": 0.6185689,
      "learning_rate": 1.1429757980239115e-06,
      "loss": 0.64029282,
      "num_input_tokens_seen": 116694305,
      "step": 5420,
      "time_per_iteration": 2.5779592990875244
    },
    {
      "auxiliary_loss_clip": 0.01170673,
      "auxiliary_loss_mlp": 0.01029374,
      "balance_loss_clip": 1.04985023,
      "balance_loss_mlp": 1.02125919,
      "epoch": 0.6518367101545122,
      "flos": 24316300414080.0,
      "grad_norm": 2.468568569870674,
      "language_loss": 0.81537086,
      "learning_rate": 1.1422720340875636e-06,
      "loss": 0.83737129,
      "num_input_tokens_seen": 116713055,
      "step": 5421,
      "time_per_iteration": 2.4728310108184814
    },
    {
      "auxiliary_loss_clip": 0.01159162,
      "auxiliary_loss_mlp": 0.01034495,
      "balance_loss_clip": 1.04809153,
      "balance_loss_mlp": 1.02647817,
      "epoch": 0.6519569530451512,
      "flos": 20011939971840.0,
      "grad_norm": 2.140041212003667,
      "language_loss": 0.79481304,
      "learning_rate": 1.1415684002678671e-06,
      "loss": 0.81674957,
      "num_input_tokens_seen": 116731815,
      "step": 5422,
      "time_per_iteration": 2.4942705631256104
    },
    {
      "auxiliary_loss_clip": 0.01143053,
      "auxiliary_loss_mlp": 0.01026741,
      "balance_loss_clip": 1.0451498,
      "balance_loss_mlp": 1.01829755,
      "epoch": 0.6520771959357903,
      "flos": 21576064682880.0,
      "grad_norm": 1.9603442988834594,
      "language_loss": 0.7765432,
      "learning_rate": 1.1408648966715617e-06,
      "loss": 0.79824114,
      "num_input_tokens_seen": 116749335,
      "step": 5423,
      "time_per_iteration": 2.5174977779388428
    },
    {
      "auxiliary_loss_clip": 0.01139841,
      "auxiliary_loss_mlp": 0.0102888,
      "balance_loss_clip": 1.04374146,
      "balance_loss_mlp": 1.02083921,
      "epoch": 0.6521974388264293,
      "flos": 22711021695360.0,
      "grad_norm": 1.8823858862340872,
      "language_loss": 0.72615111,
      "learning_rate": 1.1401615234053683e-06,
      "loss": 0.74783826,
      "num_input_tokens_seen": 116768155,
      "step": 5424,
      "time_per_iteration": 2.5169951915740967
    },
    {
      "auxiliary_loss_clip": 0.01143244,
      "auxiliary_loss_mlp": 0.01024849,
      "balance_loss_clip": 1.04653502,
      "balance_loss_mlp": 1.01754737,
      "epoch": 0.6523176817170685,
      "flos": 23002939526400.0,
      "grad_norm": 2.1650782425964494,
      "language_loss": 0.75934827,
      "learning_rate": 1.1394582805759885e-06,
      "loss": 0.78102911,
      "num_input_tokens_seen": 116787435,
      "step": 5425,
      "time_per_iteration": 2.522138833999634
    },
    {
      "auxiliary_loss_clip": 0.0115476,
      "auxiliary_loss_mlp": 0.01026721,
      "balance_loss_clip": 1.04967237,
      "balance_loss_mlp": 1.01869273,
      "epoch": 0.6524379246077076,
      "flos": 21688249835520.0,
      "grad_norm": 1.688368495305262,
      "language_loss": 0.75497508,
      "learning_rate": 1.1387551682901022e-06,
      "loss": 0.77678984,
      "num_input_tokens_seen": 116808040,
      "step": 5426,
      "time_per_iteration": 2.500398635864258
    },
    {
      "auxiliary_loss_clip": 0.01120939,
      "auxiliary_loss_mlp": 0.01027841,
      "balance_loss_clip": 1.04183626,
      "balance_loss_mlp": 1.02014041,
      "epoch": 0.6525581674983466,
      "flos": 19390936711680.0,
      "grad_norm": 1.87369360923161,
      "language_loss": 0.70748276,
      "learning_rate": 1.138052186654373e-06,
      "loss": 0.72897059,
      "num_input_tokens_seen": 116825510,
      "step": 5427,
      "time_per_iteration": 2.529946804046631
    },
    {
      "auxiliary_loss_clip": 0.01143561,
      "auxiliary_loss_mlp": 0.01031055,
      "balance_loss_clip": 1.04792464,
      "balance_loss_mlp": 1.02350664,
      "epoch": 0.6526784103889858,
      "flos": 17165444832000.0,
      "grad_norm": 2.3213033898250934,
      "language_loss": 0.87857103,
      "learning_rate": 1.1373493357754417e-06,
      "loss": 0.90031719,
      "num_input_tokens_seen": 116844415,
      "step": 5428,
      "time_per_iteration": 2.5116307735443115
    },
    {
      "auxiliary_loss_clip": 0.01167631,
      "auxiliary_loss_mlp": 0.01025377,
      "balance_loss_clip": 1.04855597,
      "balance_loss_mlp": 1.01847494,
      "epoch": 0.6527986532796248,
      "flos": 18989168112000.0,
      "grad_norm": 2.1333095800533988,
      "language_loss": 0.77011192,
      "learning_rate": 1.1366466157599303e-06,
      "loss": 0.79204202,
      "num_input_tokens_seen": 116863690,
      "step": 5429,
      "time_per_iteration": 2.491112232208252
    },
    {
      "auxiliary_loss_clip": 0.01107497,
      "auxiliary_loss_mlp": 0.00760976,
      "balance_loss_clip": 1.04170346,
      "balance_loss_mlp": 1.00025845,
      "epoch": 0.6529188961702639,
      "flos": 14238581011200.0,
      "grad_norm": 2.2298835995679043,
      "language_loss": 0.76055115,
      "learning_rate": 1.1359440267144412e-06,
      "loss": 0.7792359,
      "num_input_tokens_seen": 116881145,
      "step": 5430,
      "time_per_iteration": 2.652953624725342
    },
    {
      "auxiliary_loss_clip": 0.01155448,
      "auxiliary_loss_mlp": 0.01033553,
      "balance_loss_clip": 1.04789424,
      "balance_loss_mlp": 1.02664232,
      "epoch": 0.653039139060903,
      "flos": 36682929158400.0,
      "grad_norm": 1.9587974030745092,
      "language_loss": 0.74171352,
      "learning_rate": 1.1352415687455556e-06,
      "loss": 0.76360351,
      "num_input_tokens_seen": 116902405,
      "step": 5431,
      "time_per_iteration": 2.643061637878418
    },
    {
      "auxiliary_loss_clip": 0.01156006,
      "auxiliary_loss_mlp": 0.01029152,
      "balance_loss_clip": 1.05111766,
      "balance_loss_mlp": 1.02173424,
      "epoch": 0.6531593819515421,
      "flos": 25376275785600.0,
      "grad_norm": 2.2817222488333306,
      "language_loss": 0.63780856,
      "learning_rate": 1.1345392419598362e-06,
      "loss": 0.65966022,
      "num_input_tokens_seen": 116921285,
      "step": 5432,
      "time_per_iteration": 2.5335605144500732
    },
    {
      "auxiliary_loss_clip": 0.0114947,
      "auxiliary_loss_mlp": 0.01026864,
      "balance_loss_clip": 1.04698825,
      "balance_loss_mlp": 1.01926184,
      "epoch": 0.6532796248421812,
      "flos": 21178533888000.0,
      "grad_norm": 1.5162610583897513,
      "language_loss": 0.7174319,
      "learning_rate": 1.1338370464638263e-06,
      "loss": 0.73919523,
      "num_input_tokens_seen": 116940685,
      "step": 5433,
      "time_per_iteration": 3.2798705101013184
    },
    {
      "auxiliary_loss_clip": 0.01167552,
      "auxiliary_loss_mlp": 0.01023884,
      "balance_loss_clip": 1.04757571,
      "balance_loss_mlp": 1.01657355,
      "epoch": 0.6533998677328203,
      "flos": 17675950878720.0,
      "grad_norm": 2.003982115099924,
      "language_loss": 0.63702309,
      "learning_rate": 1.1331349823640474e-06,
      "loss": 0.65893745,
      "num_input_tokens_seen": 116958115,
      "step": 5434,
      "time_per_iteration": 2.5335097312927246
    },
    {
      "auxiliary_loss_clip": 0.01153997,
      "auxiliary_loss_mlp": 0.00760446,
      "balance_loss_clip": 1.04788756,
      "balance_loss_mlp": 1.0002594,
      "epoch": 0.6535201106234594,
      "flos": 28400384701440.0,
      "grad_norm": 2.065219746491407,
      "language_loss": 0.77803862,
      "learning_rate": 1.132433049767003e-06,
      "loss": 0.79718304,
      "num_input_tokens_seen": 116976030,
      "step": 5435,
      "time_per_iteration": 2.5605647563934326
    },
    {
      "auxiliary_loss_clip": 0.01140338,
      "auxiliary_loss_mlp": 0.01027394,
      "balance_loss_clip": 1.04840446,
      "balance_loss_mlp": 1.02054799,
      "epoch": 0.6536403535140984,
      "flos": 23586667447680.0,
      "grad_norm": 1.7827630882289778,
      "language_loss": 0.80858266,
      "learning_rate": 1.1317312487791748e-06,
      "loss": 0.83025998,
      "num_input_tokens_seen": 116997680,
      "step": 5436,
      "time_per_iteration": 2.5478522777557373
    },
    {
      "auxiliary_loss_clip": 0.01148374,
      "auxiliary_loss_mlp": 0.01026691,
      "balance_loss_clip": 1.04568648,
      "balance_loss_mlp": 1.01904392,
      "epoch": 0.6537605964047376,
      "flos": 21579476474880.0,
      "grad_norm": 2.010792245561881,
      "language_loss": 0.73236609,
      "learning_rate": 1.1310295795070253e-06,
      "loss": 0.75411677,
      "num_input_tokens_seen": 117017620,
      "step": 5437,
      "time_per_iteration": 2.476300001144409
    },
    {
      "auxiliary_loss_clip": 0.01115559,
      "auxiliary_loss_mlp": 0.01033477,
      "balance_loss_clip": 1.04492652,
      "balance_loss_mlp": 1.02648878,
      "epoch": 0.6538808392953767,
      "flos": 26833997433600.0,
      "grad_norm": 1.7651852704788737,
      "language_loss": 0.80499816,
      "learning_rate": 1.1303280420569982e-06,
      "loss": 0.82648849,
      "num_input_tokens_seen": 117039505,
      "step": 5438,
      "time_per_iteration": 2.6357927322387695
    },
    {
      "auxiliary_loss_clip": 0.01148789,
      "auxiliary_loss_mlp": 0.01028934,
      "balance_loss_clip": 1.04653907,
      "balance_loss_mlp": 1.02183855,
      "epoch": 0.6540010821860157,
      "flos": 30738241301760.0,
      "grad_norm": 1.6062057564808938,
      "language_loss": 0.77092338,
      "learning_rate": 1.1296266365355158e-06,
      "loss": 0.79270059,
      "num_input_tokens_seen": 117062890,
      "step": 5439,
      "time_per_iteration": 2.581784963607788
    },
    {
      "auxiliary_loss_clip": 0.01134941,
      "auxiliary_loss_mlp": 0.01029823,
      "balance_loss_clip": 1.04998517,
      "balance_loss_mlp": 1.02166033,
      "epoch": 0.6541213250766549,
      "flos": 26907147480960.0,
      "grad_norm": 1.9048859174638555,
      "language_loss": 0.74205482,
      "learning_rate": 1.1289253630489806e-06,
      "loss": 0.76370239,
      "num_input_tokens_seen": 117083940,
      "step": 5440,
      "time_per_iteration": 2.651430368423462
    },
    {
      "auxiliary_loss_clip": 0.01160215,
      "auxiliary_loss_mlp": 0.01030818,
      "balance_loss_clip": 1.04779172,
      "balance_loss_mlp": 1.02292013,
      "epoch": 0.6542415679672939,
      "flos": 19172384409600.0,
      "grad_norm": 1.9277055197498014,
      "language_loss": 0.72163033,
      "learning_rate": 1.1282242217037753e-06,
      "loss": 0.74354064,
      "num_input_tokens_seen": 117101440,
      "step": 5441,
      "time_per_iteration": 3.2570526599884033
    },
    {
      "auxiliary_loss_clip": 0.0110901,
      "auxiliary_loss_mlp": 0.01032171,
      "balance_loss_clip": 1.04170024,
      "balance_loss_mlp": 1.02380252,
      "epoch": 0.654361810857933,
      "flos": 48173517100800.0,
      "grad_norm": 2.694827500064338,
      "language_loss": 0.62048841,
      "learning_rate": 1.127523212606262e-06,
      "loss": 0.64190018,
      "num_input_tokens_seen": 117124265,
      "step": 5442,
      "time_per_iteration": 2.8150174617767334
    },
    {
      "auxiliary_loss_clip": 0.01153988,
      "auxiliary_loss_mlp": 0.01031856,
      "balance_loss_clip": 1.04760766,
      "balance_loss_mlp": 1.02466822,
      "epoch": 0.6544820537485722,
      "flos": 26943165843840.0,
      "grad_norm": 1.567245481564848,
      "language_loss": 0.73010802,
      "learning_rate": 1.1268223358627835e-06,
      "loss": 0.75196648,
      "num_input_tokens_seen": 117146755,
      "step": 5443,
      "time_per_iteration": 2.5406782627105713
    },
    {
      "auxiliary_loss_clip": 0.01169415,
      "auxiliary_loss_mlp": 0.01026769,
      "balance_loss_clip": 1.04970324,
      "balance_loss_mlp": 1.01906836,
      "epoch": 0.6546022966392112,
      "flos": 20886328748160.0,
      "grad_norm": 1.7671875198862104,
      "language_loss": 0.71524739,
      "learning_rate": 1.126121591579663e-06,
      "loss": 0.7372092,
      "num_input_tokens_seen": 117165960,
      "step": 5444,
      "time_per_iteration": 3.2688372135162354
    },
    {
      "auxiliary_loss_clip": 0.01152696,
      "auxiliary_loss_mlp": 0.01024446,
      "balance_loss_clip": 1.04932737,
      "balance_loss_mlp": 1.0173974,
      "epoch": 0.6547225395298503,
      "flos": 24936693143040.0,
      "grad_norm": 1.5705647188111869,
      "language_loss": 0.6875428,
      "learning_rate": 1.1254209798632018e-06,
      "loss": 0.70931417,
      "num_input_tokens_seen": 117186980,
      "step": 5445,
      "time_per_iteration": 3.2837698459625244
    },
    {
      "auxiliary_loss_clip": 0.01086055,
      "auxiliary_loss_mlp": 0.01030177,
      "balance_loss_clip": 1.03774285,
      "balance_loss_mlp": 1.02331638,
      "epoch": 0.6548427824204894,
      "flos": 22565942663040.0,
      "grad_norm": 1.7924037191174962,
      "language_loss": 0.84595871,
      "learning_rate": 1.124720500819683e-06,
      "loss": 0.8671211,
      "num_input_tokens_seen": 117205135,
      "step": 5446,
      "time_per_iteration": 2.6200053691864014
    },
    {
      "auxiliary_loss_clip": 0.01171968,
      "auxiliary_loss_mlp": 0.01030582,
      "balance_loss_clip": 1.05258346,
      "balance_loss_mlp": 1.02263689,
      "epoch": 0.6549630253111285,
      "flos": 18442500048000.0,
      "grad_norm": 1.727811063686695,
      "language_loss": 0.8218323,
      "learning_rate": 1.1240201545553682e-06,
      "loss": 0.84385788,
      "num_input_tokens_seen": 117222935,
      "step": 5447,
      "time_per_iteration": 2.4540579319000244
    },
    {
      "auxiliary_loss_clip": 0.01125208,
      "auxiliary_loss_mlp": 0.01029786,
      "balance_loss_clip": 1.0449276,
      "balance_loss_mlp": 1.02231145,
      "epoch": 0.6550832682017675,
      "flos": 25187313312000.0,
      "grad_norm": 9.527498504966,
      "language_loss": 0.72851151,
      "learning_rate": 1.1233199411764987e-06,
      "loss": 0.75006151,
      "num_input_tokens_seen": 117242370,
      "step": 5448,
      "time_per_iteration": 2.5746569633483887
    },
    {
      "auxiliary_loss_clip": 0.01114819,
      "auxiliary_loss_mlp": 0.01021628,
      "balance_loss_clip": 1.04276121,
      "balance_loss_mlp": 1.01424587,
      "epoch": 0.6552035110924067,
      "flos": 22748153379840.0,
      "grad_norm": 1.7209953386803591,
      "language_loss": 0.68918955,
      "learning_rate": 1.1226198607892978e-06,
      "loss": 0.710554,
      "num_input_tokens_seen": 117262930,
      "step": 5449,
      "time_per_iteration": 2.567357063293457
    },
    {
      "auxiliary_loss_clip": 0.01116992,
      "auxiliary_loss_mlp": 0.01026723,
      "balance_loss_clip": 1.04650557,
      "balance_loss_mlp": 1.01950812,
      "epoch": 0.6553237539830458,
      "flos": 21799178012160.0,
      "grad_norm": 2.066297570341196,
      "language_loss": 0.79976583,
      "learning_rate": 1.1219199134999664e-06,
      "loss": 0.82120299,
      "num_input_tokens_seen": 117281430,
      "step": 5450,
      "time_per_iteration": 2.5963332653045654
    },
    {
      "auxiliary_loss_clip": 0.01137605,
      "auxiliary_loss_mlp": 0.0102835,
      "balance_loss_clip": 1.04673088,
      "balance_loss_mlp": 1.02061927,
      "epoch": 0.6554439968736848,
      "flos": 20887226588160.0,
      "grad_norm": 4.870540980115804,
      "language_loss": 0.78874111,
      "learning_rate": 1.1212200994146863e-06,
      "loss": 0.81040072,
      "num_input_tokens_seen": 117299185,
      "step": 5451,
      "time_per_iteration": 2.510100841522217
    },
    {
      "auxiliary_loss_clip": 0.01120896,
      "auxiliary_loss_mlp": 0.01025758,
      "balance_loss_clip": 1.03885198,
      "balance_loss_mlp": 1.01792049,
      "epoch": 0.655564239764324,
      "flos": 16139045698560.0,
      "grad_norm": 1.716750942511835,
      "language_loss": 0.75635278,
      "learning_rate": 1.120520418639618e-06,
      "loss": 0.77781934,
      "num_input_tokens_seen": 117317720,
      "step": 5452,
      "time_per_iteration": 2.5658695697784424
    },
    {
      "auxiliary_loss_clip": 0.01155711,
      "auxiliary_loss_mlp": 0.01026105,
      "balance_loss_clip": 1.05017447,
      "balance_loss_mlp": 1.0190239,
      "epoch": 0.655684482654963,
      "flos": 29570354496000.0,
      "grad_norm": 2.4190950596145067,
      "language_loss": 0.83451569,
      "learning_rate": 1.119820871280903e-06,
      "loss": 0.85633379,
      "num_input_tokens_seen": 117338795,
      "step": 5453,
      "time_per_iteration": 2.581064462661743
    },
    {
      "auxiliary_loss_clip": 0.01154464,
      "auxiliary_loss_mlp": 0.01027794,
      "balance_loss_clip": 1.04772854,
      "balance_loss_mlp": 1.02018261,
      "epoch": 0.6558047255456021,
      "flos": 29789409588480.0,
      "grad_norm": 1.935642870233234,
      "language_loss": 0.73469055,
      "learning_rate": 1.1191214574446614e-06,
      "loss": 0.75651312,
      "num_input_tokens_seen": 117359040,
      "step": 5454,
      "time_per_iteration": 2.5680932998657227
    },
    {
      "auxiliary_loss_clip": 0.01136118,
      "auxiliary_loss_mlp": 0.01028248,
      "balance_loss_clip": 1.04620004,
      "balance_loss_mlp": 1.02102351,
      "epoch": 0.6559249684362413,
      "flos": 29059166090880.0,
      "grad_norm": 1.5937095356830602,
      "language_loss": 0.80105704,
      "learning_rate": 1.118422177236995e-06,
      "loss": 0.82270074,
      "num_input_tokens_seen": 117380865,
      "step": 5455,
      "time_per_iteration": 2.593764066696167
    },
    {
      "auxiliary_loss_clip": 0.01140722,
      "auxiliary_loss_mlp": 0.0102548,
      "balance_loss_clip": 1.04595661,
      "balance_loss_mlp": 1.0174278,
      "epoch": 0.6560452113268803,
      "flos": 20225464369920.0,
      "grad_norm": 2.3005430970029312,
      "language_loss": 0.85712254,
      "learning_rate": 1.1177230307639835e-06,
      "loss": 0.87878454,
      "num_input_tokens_seen": 117398405,
      "step": 5456,
      "time_per_iteration": 2.5446839332580566
    },
    {
      "auxiliary_loss_clip": 0.01121491,
      "auxiliary_loss_mlp": 0.01030874,
      "balance_loss_clip": 1.04395103,
      "balance_loss_mlp": 1.02365828,
      "epoch": 0.6561654542175194,
      "flos": 25045538330880.0,
      "grad_norm": 1.6508019213303118,
      "language_loss": 0.78711754,
      "learning_rate": 1.1170240181316865e-06,
      "loss": 0.8086412,
      "num_input_tokens_seen": 117419850,
      "step": 5457,
      "time_per_iteration": 2.607171058654785
    },
    {
      "auxiliary_loss_clip": 0.0111954,
      "auxiliary_loss_mlp": 0.01028517,
      "balance_loss_clip": 1.04087067,
      "balance_loss_mlp": 1.02064323,
      "epoch": 0.6562856971081584,
      "flos": 22856711258880.0,
      "grad_norm": 2.0198146436567277,
      "language_loss": 0.7964766,
      "learning_rate": 1.1163251394461442e-06,
      "loss": 0.81795728,
      "num_input_tokens_seen": 117438330,
      "step": 5458,
      "time_per_iteration": 2.5665955543518066
    },
    {
      "auxiliary_loss_clip": 0.01152936,
      "auxiliary_loss_mlp": 0.01026815,
      "balance_loss_clip": 1.04834151,
      "balance_loss_mlp": 1.01862788,
      "epoch": 0.6564059399987976,
      "flos": 18872565586560.0,
      "grad_norm": 1.81068764104059,
      "language_loss": 0.82065606,
      "learning_rate": 1.1156263948133746e-06,
      "loss": 0.8424536,
      "num_input_tokens_seen": 117454985,
      "step": 5459,
      "time_per_iteration": 3.52496337890625
    },
    {
      "auxiliary_loss_clip": 0.01106005,
      "auxiliary_loss_mlp": 0.00761399,
      "balance_loss_clip": 1.04359102,
      "balance_loss_mlp": 1.00023341,
      "epoch": 0.6565261828894366,
      "flos": 25484187219840.0,
      "grad_norm": 1.593804037533109,
      "language_loss": 0.77661335,
      "learning_rate": 1.1149277843393787e-06,
      "loss": 0.79528743,
      "num_input_tokens_seen": 117476145,
      "step": 5460,
      "time_per_iteration": 2.6693572998046875
    },
    {
      "auxiliary_loss_clip": 0.01090887,
      "auxiliary_loss_mlp": 0.00761061,
      "balance_loss_clip": 1.03721797,
      "balance_loss_mlp": 1.00020564,
      "epoch": 0.6566464257800757,
      "flos": 19683500987520.0,
      "grad_norm": 2.6396831404628385,
      "language_loss": 0.63463479,
      "learning_rate": 1.1142293081301342e-06,
      "loss": 0.65315425,
      "num_input_tokens_seen": 117494025,
      "step": 5461,
      "time_per_iteration": 2.6189284324645996
    },
    {
      "auxiliary_loss_clip": 0.01135655,
      "auxiliary_loss_mlp": 0.01026732,
      "balance_loss_clip": 1.04522777,
      "balance_loss_mlp": 1.02011836,
      "epoch": 0.6567666686707149,
      "flos": 23514127931520.0,
      "grad_norm": 1.5642342096534616,
      "language_loss": 0.67855269,
      "learning_rate": 1.1135309662915995e-06,
      "loss": 0.70017654,
      "num_input_tokens_seen": 117514190,
      "step": 5462,
      "time_per_iteration": 2.554874897003174
    },
    {
      "auxiliary_loss_clip": 0.01115321,
      "auxiliary_loss_mlp": 0.01028641,
      "balance_loss_clip": 1.04260707,
      "balance_loss_mlp": 1.02133381,
      "epoch": 0.6568869115613539,
      "flos": 32781342896640.0,
      "grad_norm": 1.8994390730297561,
      "language_loss": 0.60381019,
      "learning_rate": 1.112832758929712e-06,
      "loss": 0.62524986,
      "num_input_tokens_seen": 117536800,
      "step": 5463,
      "time_per_iteration": 2.6939549446105957
    },
    {
      "auxiliary_loss_clip": 0.01151627,
      "auxiliary_loss_mlp": 0.010263,
      "balance_loss_clip": 1.04787576,
      "balance_loss_mlp": 1.01877809,
      "epoch": 0.657007154451993,
      "flos": 18442428220800.0,
      "grad_norm": 1.819818643638633,
      "language_loss": 0.7472226,
      "learning_rate": 1.11213468615039e-06,
      "loss": 0.76900184,
      "num_input_tokens_seen": 117556230,
      "step": 5464,
      "time_per_iteration": 2.528226375579834
    },
    {
      "auxiliary_loss_clip": 0.01092843,
      "auxiliary_loss_mlp": 0.01029503,
      "balance_loss_clip": 1.04146051,
      "balance_loss_mlp": 1.02217174,
      "epoch": 0.6571273973426321,
      "flos": 25156717902720.0,
      "grad_norm": 1.9163785361743049,
      "language_loss": 0.75357556,
      "learning_rate": 1.1114367480595292e-06,
      "loss": 0.77479899,
      "num_input_tokens_seen": 117577310,
      "step": 5465,
      "time_per_iteration": 2.663393974304199
    },
    {
      "auxiliary_loss_clip": 0.01094516,
      "auxiliary_loss_mlp": 0.01026922,
      "balance_loss_clip": 1.04042733,
      "balance_loss_mlp": 1.01851153,
      "epoch": 0.6572476402332712,
      "flos": 17529830352000.0,
      "grad_norm": 1.8865834511206034,
      "language_loss": 0.81377321,
      "learning_rate": 1.1107389447630086e-06,
      "loss": 0.83498758,
      "num_input_tokens_seen": 117596010,
      "step": 5466,
      "time_per_iteration": 2.600276470184326
    },
    {
      "auxiliary_loss_clip": 0.01134162,
      "auxiliary_loss_mlp": 0.00761249,
      "balance_loss_clip": 1.04283714,
      "balance_loss_mlp": 1.00027466,
      "epoch": 0.6573678831239103,
      "flos": 17014260487680.0,
      "grad_norm": 2.0144183661632904,
      "language_loss": 0.78312302,
      "learning_rate": 1.1100412763666818e-06,
      "loss": 0.80207717,
      "num_input_tokens_seen": 117611270,
      "step": 5467,
      "time_per_iteration": 3.35496187210083
    },
    {
      "auxiliary_loss_clip": 0.01145732,
      "auxiliary_loss_mlp": 0.01027043,
      "balance_loss_clip": 1.04922211,
      "balance_loss_mlp": 1.0198164,
      "epoch": 0.6574881260145494,
      "flos": 23910078528000.0,
      "grad_norm": 1.3931207845913098,
      "language_loss": 0.79986113,
      "learning_rate": 1.1093437429763865e-06,
      "loss": 0.82158887,
      "num_input_tokens_seen": 117631535,
      "step": 5468,
      "time_per_iteration": 2.574359655380249
    },
    {
      "auxiliary_loss_clip": 0.01154368,
      "auxiliary_loss_mlp": 0.01025516,
      "balance_loss_clip": 1.0480845,
      "balance_loss_mlp": 1.01878953,
      "epoch": 0.6576083689051885,
      "flos": 11218458504960.0,
      "grad_norm": 3.984499960184166,
      "language_loss": 0.73648316,
      "learning_rate": 1.1086463446979361e-06,
      "loss": 0.75828201,
      "num_input_tokens_seen": 117649885,
      "step": 5469,
      "time_per_iteration": 2.484447956085205
    },
    {
      "auxiliary_loss_clip": 0.01160195,
      "auxiliary_loss_mlp": 0.0102956,
      "balance_loss_clip": 1.0531857,
      "balance_loss_mlp": 1.02185035,
      "epoch": 0.6577286117958275,
      "flos": 22455553190400.0,
      "grad_norm": 1.8967664563748463,
      "language_loss": 0.77706891,
      "learning_rate": 1.1079490816371277e-06,
      "loss": 0.79896641,
      "num_input_tokens_seen": 117669650,
      "step": 5470,
      "time_per_iteration": 4.071303844451904
    },
    {
      "auxiliary_loss_clip": 0.01153395,
      "auxiliary_loss_mlp": 0.00761596,
      "balance_loss_clip": 1.04656589,
      "balance_loss_mlp": 1.0002358,
      "epoch": 0.6578488546864667,
      "flos": 21872184405120.0,
      "grad_norm": 2.2430748855138205,
      "language_loss": 0.74732447,
      "learning_rate": 1.1072519538997352e-06,
      "loss": 0.76647437,
      "num_input_tokens_seen": 117688790,
      "step": 5471,
      "time_per_iteration": 2.511948823928833
    },
    {
      "auxiliary_loss_clip": 0.01139892,
      "auxiliary_loss_mlp": 0.01025539,
      "balance_loss_clip": 1.04393494,
      "balance_loss_mlp": 1.01784444,
      "epoch": 0.6579690975771058,
      "flos": 23543753673600.0,
      "grad_norm": 1.6127733408443248,
      "language_loss": 0.82329786,
      "learning_rate": 1.1065549615915095e-06,
      "loss": 0.84495223,
      "num_input_tokens_seen": 117708620,
      "step": 5472,
      "time_per_iteration": 2.538252115249634
    },
    {
      "auxiliary_loss_clip": 0.01155381,
      "auxiliary_loss_mlp": 0.01028864,
      "balance_loss_clip": 1.05169392,
      "balance_loss_mlp": 1.02131772,
      "epoch": 0.6580893404677448,
      "flos": 32743995730560.0,
      "grad_norm": 2.347671575462026,
      "language_loss": 0.78691113,
      "learning_rate": 1.105858104818187e-06,
      "loss": 0.80875361,
      "num_input_tokens_seen": 117729775,
      "step": 5473,
      "time_per_iteration": 2.5723114013671875
    },
    {
      "auxiliary_loss_clip": 0.0116008,
      "auxiliary_loss_mlp": 0.01030076,
      "balance_loss_clip": 1.0501225,
      "balance_loss_mlp": 1.02244616,
      "epoch": 0.658209583358384,
      "flos": 15888138220800.0,
      "grad_norm": 3.5841497416169923,
      "language_loss": 0.74833161,
      "learning_rate": 1.105161383685478e-06,
      "loss": 0.77023315,
      "num_input_tokens_seen": 117746160,
      "step": 5474,
      "time_per_iteration": 2.4700145721435547
    },
    {
      "auxiliary_loss_clip": 0.01028577,
      "auxiliary_loss_mlp": 0.01003436,
      "balance_loss_clip": 1.01329851,
      "balance_loss_mlp": 1.00232184,
      "epoch": 0.658329826249023,
      "flos": 62695902447360.0,
      "grad_norm": 0.7354307146301086,
      "language_loss": 0.56343257,
      "learning_rate": 1.1044647982990771e-06,
      "loss": 0.58375275,
      "num_input_tokens_seen": 117808045,
      "step": 5475,
      "time_per_iteration": 3.1064419746398926
    },
    {
      "auxiliary_loss_clip": 0.01144448,
      "auxiliary_loss_mlp": 0.01029261,
      "balance_loss_clip": 1.05015469,
      "balance_loss_mlp": 1.02149796,
      "epoch": 0.6584500691396621,
      "flos": 31722624501120.0,
      "grad_norm": 2.2579426366155144,
      "language_loss": 0.64310187,
      "learning_rate": 1.1037683487646536e-06,
      "loss": 0.66483897,
      "num_input_tokens_seen": 117828330,
      "step": 5476,
      "time_per_iteration": 2.6103577613830566
    },
    {
      "auxiliary_loss_clip": 0.01140627,
      "auxiliary_loss_mlp": 0.00760739,
      "balance_loss_clip": 1.05014277,
      "balance_loss_mlp": 1.00023413,
      "epoch": 0.6585703120303013,
      "flos": 18406086635520.0,
      "grad_norm": 2.257420017352675,
      "language_loss": 0.77272469,
      "learning_rate": 1.1030720351878583e-06,
      "loss": 0.79173833,
      "num_input_tokens_seen": 117846450,
      "step": 5477,
      "time_per_iteration": 2.5073328018188477
    },
    {
      "auxiliary_loss_clip": 0.01043743,
      "auxiliary_loss_mlp": 0.01001973,
      "balance_loss_clip": 1.01486254,
      "balance_loss_mlp": 1.00071514,
      "epoch": 0.6586905549209403,
      "flos": 58309880434560.0,
      "grad_norm": 0.80673535201615,
      "language_loss": 0.57681525,
      "learning_rate": 1.102375857674323e-06,
      "loss": 0.5972724,
      "num_input_tokens_seen": 117908365,
      "step": 5478,
      "time_per_iteration": 3.1118485927581787
    },
    {
      "auxiliary_loss_clip": 0.01137635,
      "auxiliary_loss_mlp": 0.01025045,
      "balance_loss_clip": 1.04238844,
      "balance_loss_mlp": 1.01817203,
      "epoch": 0.6588107978115794,
      "flos": 22782627457920.0,
      "grad_norm": 2.1183017210842854,
      "language_loss": 0.90540743,
      "learning_rate": 1.1016798163296561e-06,
      "loss": 0.9270342,
      "num_input_tokens_seen": 117927565,
      "step": 5479,
      "time_per_iteration": 2.5368309020996094
    },
    {
      "auxiliary_loss_clip": 0.01154024,
      "auxiliary_loss_mlp": 0.01033565,
      "balance_loss_clip": 1.04910254,
      "balance_loss_mlp": 1.02613211,
      "epoch": 0.6589310407022185,
      "flos": 20667525050880.0,
      "grad_norm": 1.9692509847570074,
      "language_loss": 0.66329944,
      "learning_rate": 1.1009839112594471e-06,
      "loss": 0.6851753,
      "num_input_tokens_seen": 117945590,
      "step": 5480,
      "time_per_iteration": 2.486555576324463
    },
    {
      "auxiliary_loss_clip": 0.01157369,
      "auxiliary_loss_mlp": 0.01029651,
      "balance_loss_clip": 1.04892313,
      "balance_loss_mlp": 1.02274275,
      "epoch": 0.6590512835928576,
      "flos": 25630595055360.0,
      "grad_norm": 2.3429498123510197,
      "language_loss": 0.71643865,
      "learning_rate": 1.1002881425692638e-06,
      "loss": 0.73830879,
      "num_input_tokens_seen": 117966020,
      "step": 5481,
      "time_per_iteration": 2.529355049133301
    },
    {
      "auxiliary_loss_clip": 0.01147378,
      "auxiliary_loss_mlp": 0.0102554,
      "balance_loss_clip": 1.04474592,
      "balance_loss_mlp": 1.01773763,
      "epoch": 0.6591715264834966,
      "flos": 23726108044800.0,
      "grad_norm": 1.6230101669003636,
      "language_loss": 0.75278509,
      "learning_rate": 1.0995925103646532e-06,
      "loss": 0.77451432,
      "num_input_tokens_seen": 117984620,
      "step": 5482,
      "time_per_iteration": 2.515502691268921
    },
    {
      "auxiliary_loss_clip": 0.01124454,
      "auxiliary_loss_mlp": 0.01024281,
      "balance_loss_clip": 1.04895043,
      "balance_loss_mlp": 1.01713181,
      "epoch": 0.6592917693741358,
      "flos": 35773850822400.0,
      "grad_norm": 1.4859234464756599,
      "language_loss": 0.6662966,
      "learning_rate": 1.0988970147511437e-06,
      "loss": 0.68778396,
      "num_input_tokens_seen": 118006500,
      "step": 5483,
      "time_per_iteration": 2.6869187355041504
    },
    {
      "auxiliary_loss_clip": 0.01140094,
      "auxiliary_loss_mlp": 0.01025743,
      "balance_loss_clip": 1.04891944,
      "balance_loss_mlp": 1.01834583,
      "epoch": 0.6594120122647749,
      "flos": 21396834794880.0,
      "grad_norm": 2.1569933747270658,
      "language_loss": 0.80189937,
      "learning_rate": 1.0982016558342405e-06,
      "loss": 0.82355773,
      "num_input_tokens_seen": 118025470,
      "step": 5484,
      "time_per_iteration": 2.5434372425079346
    },
    {
      "auxiliary_loss_clip": 0.01170963,
      "auxiliary_loss_mlp": 0.01025558,
      "balance_loss_clip": 1.05180871,
      "balance_loss_mlp": 1.01866436,
      "epoch": 0.6595322551554139,
      "flos": 19351829779200.0,
      "grad_norm": 2.0389595193234387,
      "language_loss": 0.71062863,
      "learning_rate": 1.0975064337194291e-06,
      "loss": 0.73259377,
      "num_input_tokens_seen": 118043515,
      "step": 5485,
      "time_per_iteration": 3.2565596103668213
    },
    {
      "auxiliary_loss_clip": 0.01117643,
      "auxiliary_loss_mlp": 0.01024576,
      "balance_loss_clip": 1.04235244,
      "balance_loss_mlp": 1.01713133,
      "epoch": 0.6596524980460531,
      "flos": 16837113588480.0,
      "grad_norm": 1.3851498068770394,
      "language_loss": 0.70296061,
      "learning_rate": 1.0968113485121743e-06,
      "loss": 0.72438276,
      "num_input_tokens_seen": 118063105,
      "step": 5486,
      "time_per_iteration": 2.5696792602539062
    },
    {
      "auxiliary_loss_clip": 0.01153482,
      "auxiliary_loss_mlp": 0.00761081,
      "balance_loss_clip": 1.04522395,
      "balance_loss_mlp": 1.00023901,
      "epoch": 0.6597727409366921,
      "flos": 21798567480960.0,
      "grad_norm": 1.7493107588810082,
      "language_loss": 0.79910505,
      "learning_rate": 1.0961164003179185e-06,
      "loss": 0.81825078,
      "num_input_tokens_seen": 118081615,
      "step": 5487,
      "time_per_iteration": 2.5159542560577393
    },
    {
      "auxiliary_loss_clip": 0.01123016,
      "auxiliary_loss_mlp": 0.01024193,
      "balance_loss_clip": 1.04378653,
      "balance_loss_mlp": 1.01719499,
      "epoch": 0.6598929838273312,
      "flos": 23730704985600.0,
      "grad_norm": 1.7849862634196725,
      "language_loss": 0.83800447,
      "learning_rate": 1.0954215892420884e-06,
      "loss": 0.85947657,
      "num_input_tokens_seen": 118102315,
      "step": 5488,
      "time_per_iteration": 2.572155237197876
    },
    {
      "auxiliary_loss_clip": 0.01131713,
      "auxiliary_loss_mlp": 0.01030527,
      "balance_loss_clip": 1.04833353,
      "balance_loss_mlp": 1.02312386,
      "epoch": 0.6600132267179702,
      "flos": 19974520978560.0,
      "grad_norm": 1.6634703109275866,
      "language_loss": 0.70291805,
      "learning_rate": 1.094726915390082e-06,
      "loss": 0.72454047,
      "num_input_tokens_seen": 118120650,
      "step": 5489,
      "time_per_iteration": 2.5478005409240723
    },
    {
      "auxiliary_loss_clip": 0.01154352,
      "auxiliary_loss_mlp": 0.01030077,
      "balance_loss_clip": 1.0488323,
      "balance_loss_mlp": 1.02267766,
      "epoch": 0.6601334696086094,
      "flos": 22342649765760.0,
      "grad_norm": 1.7971623772562064,
      "language_loss": 0.69608319,
      "learning_rate": 1.0940323788672836e-06,
      "loss": 0.71792746,
      "num_input_tokens_seen": 118139825,
      "step": 5490,
      "time_per_iteration": 2.520022392272949
    },
    {
      "auxiliary_loss_clip": 0.01148238,
      "auxiliary_loss_mlp": 0.0102621,
      "balance_loss_clip": 1.04734778,
      "balance_loss_mlp": 1.01920962,
      "epoch": 0.6602537124992485,
      "flos": 25703098657920.0,
      "grad_norm": 2.165453340651022,
      "language_loss": 0.73541296,
      "learning_rate": 1.0933379797790522e-06,
      "loss": 0.75715744,
      "num_input_tokens_seen": 118159240,
      "step": 5491,
      "time_per_iteration": 2.514840841293335
    },
    {
      "auxiliary_loss_clip": 0.01168169,
      "auxiliary_loss_mlp": 0.01029778,
      "balance_loss_clip": 1.0505718,
      "balance_loss_mlp": 1.02244675,
      "epoch": 0.6603739553898875,
      "flos": 25848572739840.0,
      "grad_norm": 3.1648905207737466,
      "language_loss": 0.71749568,
      "learning_rate": 1.0926437182307293e-06,
      "loss": 0.73947525,
      "num_input_tokens_seen": 118178050,
      "step": 5492,
      "time_per_iteration": 3.265697479248047
    },
    {
      "auxiliary_loss_clip": 0.01145126,
      "auxiliary_loss_mlp": 0.01026441,
      "balance_loss_clip": 1.04649067,
      "balance_loss_mlp": 1.01954174,
      "epoch": 0.6604941982805267,
      "flos": 24570296461440.0,
      "grad_norm": 1.7439931305005685,
      "language_loss": 0.78564936,
      "learning_rate": 1.0919495943276338e-06,
      "loss": 0.807365,
      "num_input_tokens_seen": 118199070,
      "step": 5493,
      "time_per_iteration": 2.5579118728637695
    },
    {
      "auxiliary_loss_clip": 0.01125637,
      "auxiliary_loss_mlp": 0.01026317,
      "balance_loss_clip": 1.04013062,
      "balance_loss_mlp": 1.01882493,
      "epoch": 0.6606144411711657,
      "flos": 13261775581440.0,
      "grad_norm": 2.2680750302643675,
      "language_loss": 0.76303947,
      "learning_rate": 1.0912556081750611e-06,
      "loss": 0.78455907,
      "num_input_tokens_seen": 118217000,
      "step": 5494,
      "time_per_iteration": 2.514146327972412
    },
    {
      "auxiliary_loss_clip": 0.01138998,
      "auxiliary_loss_mlp": 0.01024587,
      "balance_loss_clip": 1.04865861,
      "balance_loss_mlp": 1.01760173,
      "epoch": 0.6607346840618048,
      "flos": 25155281358720.0,
      "grad_norm": 2.2928014241872643,
      "language_loss": 0.76386249,
      "learning_rate": 1.0905617598782909e-06,
      "loss": 0.78549838,
      "num_input_tokens_seen": 118237205,
      "step": 5495,
      "time_per_iteration": 2.547835111618042
    },
    {
      "auxiliary_loss_clip": 0.01103215,
      "auxiliary_loss_mlp": 0.01026684,
      "balance_loss_clip": 1.04092574,
      "balance_loss_mlp": 1.01934385,
      "epoch": 0.660854926952444,
      "flos": 17638029095040.0,
      "grad_norm": 2.158673180055542,
      "language_loss": 0.81100333,
      "learning_rate": 1.0898680495425775e-06,
      "loss": 0.83230233,
      "num_input_tokens_seen": 118255495,
      "step": 5496,
      "time_per_iteration": 3.3456761837005615
    },
    {
      "auxiliary_loss_clip": 0.01141316,
      "auxiliary_loss_mlp": 0.01029356,
      "balance_loss_clip": 1.04704785,
      "balance_loss_mlp": 1.02192342,
      "epoch": 0.660975169843083,
      "flos": 16836000266880.0,
      "grad_norm": 1.6444940842124636,
      "language_loss": 0.80337656,
      "learning_rate": 1.0891744772731594e-06,
      "loss": 0.82508326,
      "num_input_tokens_seen": 118273310,
      "step": 5497,
      "time_per_iteration": 2.494062900543213
    },
    {
      "auxiliary_loss_clip": 0.01154229,
      "auxiliary_loss_mlp": 0.01029667,
      "balance_loss_clip": 1.04695964,
      "balance_loss_mlp": 1.02226448,
      "epoch": 0.6610954127337221,
      "flos": 26870410846080.0,
      "grad_norm": 2.3638076316297494,
      "language_loss": 0.65805721,
      "learning_rate": 1.088481043175248e-06,
      "loss": 0.67989612,
      "num_input_tokens_seen": 118293880,
      "step": 5498,
      "time_per_iteration": 2.5377020835876465
    },
    {
      "auxiliary_loss_clip": 0.01130318,
      "auxiliary_loss_mlp": 0.01027656,
      "balance_loss_clip": 1.04349136,
      "balance_loss_mlp": 1.02036619,
      "epoch": 0.6612156556243612,
      "flos": 26465697331200.0,
      "grad_norm": 1.668976575413938,
      "language_loss": 0.75558031,
      "learning_rate": 1.0877877473540368e-06,
      "loss": 0.77716005,
      "num_input_tokens_seen": 118314465,
      "step": 5499,
      "time_per_iteration": 2.578437328338623
    },
    {
      "auxiliary_loss_clip": 0.01168852,
      "auxiliary_loss_mlp": 0.01030795,
      "balance_loss_clip": 1.04905868,
      "balance_loss_mlp": 1.0236659,
      "epoch": 0.6613358985150003,
      "flos": 19791915212160.0,
      "grad_norm": 1.807928516946348,
      "language_loss": 0.72489107,
      "learning_rate": 1.0870945899147002e-06,
      "loss": 0.74688751,
      "num_input_tokens_seen": 118331110,
      "step": 5500,
      "time_per_iteration": 2.4338152408599854
    },
    {
      "auxiliary_loss_clip": 0.01153463,
      "auxiliary_loss_mlp": 0.01029353,
      "balance_loss_clip": 1.04943109,
      "balance_loss_mlp": 1.02202439,
      "epoch": 0.6614561414056394,
      "flos": 26831627136000.0,
      "grad_norm": 1.939863583808289,
      "language_loss": 0.76043272,
      "learning_rate": 1.0864015709623879e-06,
      "loss": 0.78226089,
      "num_input_tokens_seen": 118351980,
      "step": 5501,
      "time_per_iteration": 2.5715439319610596
    },
    {
      "auxiliary_loss_clip": 0.0116034,
      "auxiliary_loss_mlp": 0.0103032,
      "balance_loss_clip": 1.05044353,
      "balance_loss_mlp": 1.02279162,
      "epoch": 0.6615763842962785,
      "flos": 22894597128960.0,
      "grad_norm": 2.1650330374480795,
      "language_loss": 0.79818875,
      "learning_rate": 1.0857086906022313e-06,
      "loss": 0.8200953,
      "num_input_tokens_seen": 118370315,
      "step": 5502,
      "time_per_iteration": 2.552996873855591
    },
    {
      "auxiliary_loss_clip": 0.01087141,
      "auxiliary_loss_mlp": 0.01026437,
      "balance_loss_clip": 1.04243755,
      "balance_loss_mlp": 1.01879859,
      "epoch": 0.6616966271869176,
      "flos": 24790321221120.0,
      "grad_norm": 2.0613688894625057,
      "language_loss": 0.73479098,
      "learning_rate": 1.0850159489393388e-06,
      "loss": 0.75592679,
      "num_input_tokens_seen": 118389575,
      "step": 5503,
      "time_per_iteration": 2.6428065299987793
    },
    {
      "auxiliary_loss_clip": 0.01118949,
      "auxiliary_loss_mlp": 0.01027565,
      "balance_loss_clip": 1.04063892,
      "balance_loss_mlp": 1.01957202,
      "epoch": 0.6618168700775566,
      "flos": 17202109639680.0,
      "grad_norm": 1.7808163314410936,
      "language_loss": 0.8219468,
      "learning_rate": 1.0843233460787992e-06,
      "loss": 0.84341192,
      "num_input_tokens_seen": 118406790,
      "step": 5504,
      "time_per_iteration": 2.52597975730896
    },
    {
      "auxiliary_loss_clip": 0.01112492,
      "auxiliary_loss_mlp": 0.01029787,
      "balance_loss_clip": 1.04332244,
      "balance_loss_mlp": 1.022488,
      "epoch": 0.6619371129681958,
      "flos": 25447091448960.0,
      "grad_norm": 1.933018497589596,
      "language_loss": 0.77505088,
      "learning_rate": 1.0836308821256805e-06,
      "loss": 0.79647362,
      "num_input_tokens_seen": 118427590,
      "step": 5505,
      "time_per_iteration": 2.6026968955993652
    },
    {
      "auxiliary_loss_clip": 0.01154463,
      "auxiliary_loss_mlp": 0.01024677,
      "balance_loss_clip": 1.04975986,
      "balance_loss_mlp": 1.01818943,
      "epoch": 0.6620573558588349,
      "flos": 18040444139520.0,
      "grad_norm": 2.032497809118727,
      "language_loss": 0.77861512,
      "learning_rate": 1.0829385571850282e-06,
      "loss": 0.80040658,
      "num_input_tokens_seen": 118444570,
      "step": 5506,
      "time_per_iteration": 2.4761977195739746
    },
    {
      "auxiliary_loss_clip": 0.01174588,
      "auxiliary_loss_mlp": 0.01031584,
      "balance_loss_clip": 1.05231977,
      "balance_loss_mlp": 1.02336502,
      "epoch": 0.6621775987494739,
      "flos": 17785586165760.0,
      "grad_norm": 2.314740657880902,
      "language_loss": 0.8353042,
      "learning_rate": 1.0822463713618679e-06,
      "loss": 0.85736597,
      "num_input_tokens_seen": 118461425,
      "step": 5507,
      "time_per_iteration": 2.4326953887939453
    },
    {
      "auxiliary_loss_clip": 0.0112922,
      "auxiliary_loss_mlp": 0.01030417,
      "balance_loss_clip": 1.04468369,
      "balance_loss_mlp": 1.02280498,
      "epoch": 0.6622978416401131,
      "flos": 17492590926720.0,
      "grad_norm": 2.049302002158593,
      "language_loss": 0.84745061,
      "learning_rate": 1.0815543247612034e-06,
      "loss": 0.86904699,
      "num_input_tokens_seen": 118478495,
      "step": 5508,
      "time_per_iteration": 2.528991222381592
    },
    {
      "auxiliary_loss_clip": 0.01137456,
      "auxiliary_loss_mlp": 0.01025628,
      "balance_loss_clip": 1.04203105,
      "balance_loss_mlp": 1.01841617,
      "epoch": 0.6624180845307521,
      "flos": 21648352803840.0,
      "grad_norm": 1.5296939165763543,
      "language_loss": 0.82940531,
      "learning_rate": 1.0808624174880168e-06,
      "loss": 0.85103613,
      "num_input_tokens_seen": 118499145,
      "step": 5509,
      "time_per_iteration": 2.537257432937622
    },
    {
      "auxiliary_loss_clip": 0.01164657,
      "auxiliary_loss_mlp": 0.01023131,
      "balance_loss_clip": 1.04854012,
      "balance_loss_mlp": 1.01627612,
      "epoch": 0.6625383274213912,
      "flos": 23805902108160.0,
      "grad_norm": 1.8607091536186615,
      "language_loss": 0.79889202,
      "learning_rate": 1.080170649647272e-06,
      "loss": 0.82076991,
      "num_input_tokens_seen": 118518950,
      "step": 5510,
      "time_per_iteration": 2.500964879989624
    },
    {
      "auxiliary_loss_clip": 0.01165253,
      "auxiliary_loss_mlp": 0.01029622,
      "balance_loss_clip": 1.04846716,
      "balance_loss_mlp": 1.02227271,
      "epoch": 0.6626585703120303,
      "flos": 33262941473280.0,
      "grad_norm": 1.6104067821836114,
      "language_loss": 0.67571723,
      "learning_rate": 1.0794790213439068e-06,
      "loss": 0.69766599,
      "num_input_tokens_seen": 118545850,
      "step": 5511,
      "time_per_iteration": 3.410719871520996
    },
    {
      "auxiliary_loss_clip": 0.01114764,
      "auxiliary_loss_mlp": 0.01028024,
      "balance_loss_clip": 1.04362428,
      "balance_loss_mlp": 1.02057886,
      "epoch": 0.6627788132026694,
      "flos": 22085780630400.0,
      "grad_norm": 2.7432983403994378,
      "language_loss": 0.78772312,
      "learning_rate": 1.078787532682843e-06,
      "loss": 0.80915105,
      "num_input_tokens_seen": 118563325,
      "step": 5512,
      "time_per_iteration": 2.5909414291381836
    },
    {
      "auxiliary_loss_clip": 0.01151746,
      "auxiliary_loss_mlp": 0.01024193,
      "balance_loss_clip": 1.04846525,
      "balance_loss_mlp": 1.01706421,
      "epoch": 0.6628990560933085,
      "flos": 36173608260480.0,
      "grad_norm": 2.1596213229518515,
      "language_loss": 0.75843787,
      "learning_rate": 1.0780961837689773e-06,
      "loss": 0.78019726,
      "num_input_tokens_seen": 118582835,
      "step": 5513,
      "time_per_iteration": 2.60988450050354
    },
    {
      "auxiliary_loss_clip": 0.01133478,
      "auxiliary_loss_mlp": 0.0102851,
      "balance_loss_clip": 1.04829741,
      "balance_loss_mlp": 1.02123809,
      "epoch": 0.6630192989839476,
      "flos": 18513567106560.0,
      "grad_norm": 1.8643411618227863,
      "language_loss": 0.69985747,
      "learning_rate": 1.0774049747071883e-06,
      "loss": 0.72147733,
      "num_input_tokens_seen": 118600715,
      "step": 5514,
      "time_per_iteration": 2.50639271736145
    },
    {
      "auxiliary_loss_clip": 0.01113779,
      "auxiliary_loss_mlp": 0.01021988,
      "balance_loss_clip": 1.04723322,
      "balance_loss_mlp": 1.01464462,
      "epoch": 0.6631395418745867,
      "flos": 35809510049280.0,
      "grad_norm": 4.165876233833878,
      "language_loss": 0.68201339,
      "learning_rate": 1.076713905602332e-06,
      "loss": 0.70337105,
      "num_input_tokens_seen": 118621290,
      "step": 5515,
      "time_per_iteration": 2.705028533935547
    },
    {
      "auxiliary_loss_clip": 0.01160533,
      "auxiliary_loss_mlp": 0.01032015,
      "balance_loss_clip": 1.05137312,
      "balance_loss_mlp": 1.02443373,
      "epoch": 0.6632597847652257,
      "flos": 20047742853120.0,
      "grad_norm": 1.767234490246254,
      "language_loss": 0.81218648,
      "learning_rate": 1.07602297655924e-06,
      "loss": 0.83411199,
      "num_input_tokens_seen": 118639610,
      "step": 5516,
      "time_per_iteration": 2.484509229660034
    },
    {
      "auxiliary_loss_clip": 0.01169995,
      "auxiliary_loss_mlp": 0.01023097,
      "balance_loss_clip": 1.05278468,
      "balance_loss_mlp": 1.01617074,
      "epoch": 0.6633800276558649,
      "flos": 21214480423680.0,
      "grad_norm": 1.6742899219487943,
      "language_loss": 0.8105191,
      "learning_rate": 1.0753321876827292e-06,
      "loss": 0.83244991,
      "num_input_tokens_seen": 118658895,
      "step": 5517,
      "time_per_iteration": 2.465320348739624
    },
    {
      "auxiliary_loss_clip": 0.0116881,
      "auxiliary_loss_mlp": 0.01028755,
      "balance_loss_clip": 1.04975533,
      "balance_loss_mlp": 1.02153695,
      "epoch": 0.663500270546504,
      "flos": 23987753688960.0,
      "grad_norm": 3.2276838761070024,
      "language_loss": 0.73799723,
      "learning_rate": 1.0746415390775893e-06,
      "loss": 0.75997287,
      "num_input_tokens_seen": 118677025,
      "step": 5518,
      "time_per_iteration": 3.277190923690796
    },
    {
      "auxiliary_loss_clip": 0.01167921,
      "auxiliary_loss_mlp": 0.01031745,
      "balance_loss_clip": 1.05128968,
      "balance_loss_mlp": 1.02448535,
      "epoch": 0.663620513437143,
      "flos": 17932389050880.0,
      "grad_norm": 2.4462773832948197,
      "language_loss": 0.7630775,
      "learning_rate": 1.0739510308485939e-06,
      "loss": 0.78507423,
      "num_input_tokens_seen": 118694240,
      "step": 5519,
      "time_per_iteration": 2.4362287521362305
    },
    {
      "auxiliary_loss_clip": 0.01031524,
      "auxiliary_loss_mlp": 0.01002315,
      "balance_loss_clip": 1.01451707,
      "balance_loss_mlp": 1.00111103,
      "epoch": 0.6637407563277821,
      "flos": 57840241086720.0,
      "grad_norm": 0.8112241909036088,
      "language_loss": 0.62560725,
      "learning_rate": 1.07326066310049e-06,
      "loss": 0.64594567,
      "num_input_tokens_seen": 118758365,
      "step": 5520,
      "time_per_iteration": 3.171503782272339
    },
    {
      "auxiliary_loss_clip": 0.0112347,
      "auxiliary_loss_mlp": 0.01030896,
      "balance_loss_clip": 1.04358244,
      "balance_loss_mlp": 1.02272391,
      "epoch": 0.6638609992184212,
      "flos": 27306007079040.0,
      "grad_norm": 1.78182835288331,
      "language_loss": 0.79073703,
      "learning_rate": 1.0725704359380059e-06,
      "loss": 0.81228065,
      "num_input_tokens_seen": 118778220,
      "step": 5521,
      "time_per_iteration": 2.6341795921325684
    },
    {
      "auxiliary_loss_clip": 0.01168623,
      "auxiliary_loss_mlp": 0.01031961,
      "balance_loss_clip": 1.05006671,
      "balance_loss_mlp": 1.02466583,
      "epoch": 0.6639812421090603,
      "flos": 18624854419200.0,
      "grad_norm": 1.8415030156218595,
      "language_loss": 0.7208572,
      "learning_rate": 1.0718803494658497e-06,
      "loss": 0.74286306,
      "num_input_tokens_seen": 118797110,
      "step": 5522,
      "time_per_iteration": 3.151150941848755
    },
    {
      "auxiliary_loss_clip": 0.01059058,
      "auxiliary_loss_mlp": 0.01030775,
      "balance_loss_clip": 1.0361861,
      "balance_loss_mlp": 1.02331877,
      "epoch": 0.6641014849996993,
      "flos": 15924479806080.0,
      "grad_norm": 2.0456372258305198,
      "language_loss": 0.84107578,
      "learning_rate": 1.071190403788707e-06,
      "loss": 0.86197412,
      "num_input_tokens_seen": 118812415,
      "step": 5523,
      "time_per_iteration": 3.5614545345306396
    },
    {
      "auxiliary_loss_clip": 0.01134931,
      "auxiliary_loss_mlp": 0.01027932,
      "balance_loss_clip": 1.04885173,
      "balance_loss_mlp": 1.02079189,
      "epoch": 0.6642217278903385,
      "flos": 26505486622080.0,
      "grad_norm": 1.962634767268292,
      "language_loss": 0.75613731,
      "learning_rate": 1.0705005990112415e-06,
      "loss": 0.77776593,
      "num_input_tokens_seen": 118832195,
      "step": 5524,
      "time_per_iteration": 2.8410847187042236
    },
    {
      "auxiliary_loss_clip": 0.01101889,
      "auxiliary_loss_mlp": 0.01025927,
      "balance_loss_clip": 1.04534936,
      "balance_loss_mlp": 1.01834559,
      "epoch": 0.6643419707809776,
      "flos": 15377308951680.0,
      "grad_norm": 4.6554250175982785,
      "language_loss": 0.74557507,
      "learning_rate": 1.0698109352380957e-06,
      "loss": 0.76685327,
      "num_input_tokens_seen": 118849795,
      "step": 5525,
      "time_per_iteration": 2.5994455814361572
    },
    {
      "auxiliary_loss_clip": 0.01165847,
      "auxiliary_loss_mlp": 0.01026272,
      "balance_loss_clip": 1.04871643,
      "balance_loss_mlp": 1.01907766,
      "epoch": 0.6644622136716166,
      "flos": 25117610970240.0,
      "grad_norm": 1.7386773514304639,
      "language_loss": 0.77790105,
      "learning_rate": 1.0691214125738909e-06,
      "loss": 0.79982221,
      "num_input_tokens_seen": 118870000,
      "step": 5526,
      "time_per_iteration": 2.5054798126220703
    },
    {
      "auxiliary_loss_clip": 0.01061387,
      "auxiliary_loss_mlp": 0.01002138,
      "balance_loss_clip": 1.013659,
      "balance_loss_mlp": 1.00094032,
      "epoch": 0.6645824565622558,
      "flos": 66201717680640.0,
      "grad_norm": 0.791641891176091,
      "language_loss": 0.57523292,
      "learning_rate": 1.0684320311232287e-06,
      "loss": 0.59586817,
      "num_input_tokens_seen": 118932905,
      "step": 5527,
      "time_per_iteration": 3.2169415950775146
    },
    {
      "auxiliary_loss_clip": 0.01135224,
      "auxiliary_loss_mlp": 0.01024632,
      "balance_loss_clip": 1.0454042,
      "balance_loss_mlp": 1.01684177,
      "epoch": 0.6647026994528948,
      "flos": 25082131311360.0,
      "grad_norm": 1.6500292091141693,
      "language_loss": 0.81289381,
      "learning_rate": 1.0677427909906865e-06,
      "loss": 0.83449233,
      "num_input_tokens_seen": 118953355,
      "step": 5528,
      "time_per_iteration": 2.6502134799957275
    },
    {
      "auxiliary_loss_clip": 0.01172388,
      "auxiliary_loss_mlp": 0.01030413,
      "balance_loss_clip": 1.0509243,
      "balance_loss_mlp": 1.02305245,
      "epoch": 0.6648229423435339,
      "flos": 18222187979520.0,
      "grad_norm": 1.928909200883684,
      "language_loss": 0.72334909,
      "learning_rate": 1.0670536922808216e-06,
      "loss": 0.74537706,
      "num_input_tokens_seen": 118973480,
      "step": 5529,
      "time_per_iteration": 2.565969228744507
    },
    {
      "auxiliary_loss_clip": 0.01142592,
      "auxiliary_loss_mlp": 0.01028578,
      "balance_loss_clip": 1.04839218,
      "balance_loss_mlp": 1.02149677,
      "epoch": 0.6649431852341731,
      "flos": 18296882311680.0,
      "grad_norm": 2.0330468833455746,
      "language_loss": 0.71671939,
      "learning_rate": 1.06636473509817e-06,
      "loss": 0.73843116,
      "num_input_tokens_seen": 118989860,
      "step": 5530,
      "time_per_iteration": 2.5809764862060547
    },
    {
      "auxiliary_loss_clip": 0.01137972,
      "auxiliary_loss_mlp": 0.00761254,
      "balance_loss_clip": 1.04583156,
      "balance_loss_mlp": 1.00023961,
      "epoch": 0.6650634281248121,
      "flos": 17019575700480.0,
      "grad_norm": 2.117131494032988,
      "language_loss": 0.80721557,
      "learning_rate": 1.0656759195472447e-06,
      "loss": 0.82620788,
      "num_input_tokens_seen": 119007150,
      "step": 5531,
      "time_per_iteration": 2.511514902114868
    },
    {
      "auxiliary_loss_clip": 0.01040683,
      "auxiliary_loss_mlp": 0.01002618,
      "balance_loss_clip": 1.01448774,
      "balance_loss_mlp": 1.00149715,
      "epoch": 0.6651836710154512,
      "flos": 69294810666240.0,
      "grad_norm": 0.8015071945037564,
      "language_loss": 0.59731883,
      "learning_rate": 1.0649872457325414e-06,
      "loss": 0.61775184,
      "num_input_tokens_seen": 119068435,
      "step": 5532,
      "time_per_iteration": 3.075063705444336
    },
    {
      "auxiliary_loss_clip": 0.01052046,
      "auxiliary_loss_mlp": 0.01001531,
      "balance_loss_clip": 1.01303077,
      "balance_loss_mlp": 1.00029755,
      "epoch": 0.6653039139060903,
      "flos": 66883444882560.0,
      "grad_norm": 0.8483726450148077,
      "language_loss": 0.55121338,
      "learning_rate": 1.0642987137585278e-06,
      "loss": 0.57174921,
      "num_input_tokens_seen": 119127960,
      "step": 5533,
      "time_per_iteration": 3.0424134731292725
    },
    {
      "auxiliary_loss_clip": 0.01137593,
      "auxiliary_loss_mlp": 0.0102605,
      "balance_loss_clip": 1.04639876,
      "balance_loss_mlp": 1.01865935,
      "epoch": 0.6654241567967294,
      "flos": 21470056669440.0,
      "grad_norm": 1.6894615229284786,
      "language_loss": 0.82278341,
      "learning_rate": 1.0636103237296561e-06,
      "loss": 0.84441984,
      "num_input_tokens_seen": 119146885,
      "step": 5534,
      "time_per_iteration": 2.5111546516418457
    },
    {
      "auxiliary_loss_clip": 0.01154523,
      "auxiliary_loss_mlp": 0.01028138,
      "balance_loss_clip": 1.05258727,
      "balance_loss_mlp": 1.02159357,
      "epoch": 0.6655443996873684,
      "flos": 25119514391040.0,
      "grad_norm": 1.789725700514809,
      "language_loss": 0.84607613,
      "learning_rate": 1.062922075750353e-06,
      "loss": 0.86790276,
      "num_input_tokens_seen": 119166900,
      "step": 5535,
      "time_per_iteration": 2.519035816192627
    },
    {
      "auxiliary_loss_clip": 0.01128742,
      "auxiliary_loss_mlp": 0.01023192,
      "balance_loss_clip": 1.04626155,
      "balance_loss_mlp": 1.01642966,
      "epoch": 0.6656646425780076,
      "flos": 17457326749440.0,
      "grad_norm": 1.9207645013029482,
      "language_loss": 0.72056508,
      "learning_rate": 1.0622339699250267e-06,
      "loss": 0.74208444,
      "num_input_tokens_seen": 119184820,
      "step": 5536,
      "time_per_iteration": 3.3568038940429688
    },
    {
      "auxiliary_loss_clip": 0.01125732,
      "auxiliary_loss_mlp": 0.01027251,
      "balance_loss_clip": 1.04463577,
      "balance_loss_mlp": 1.02078438,
      "epoch": 0.6657848854686467,
      "flos": 23434190213760.0,
      "grad_norm": 2.3385865048339887,
      "language_loss": 0.79227579,
      "learning_rate": 1.0615460063580624e-06,
      "loss": 0.81380564,
      "num_input_tokens_seen": 119203295,
      "step": 5537,
      "time_per_iteration": 2.5803029537200928
    },
    {
      "auxiliary_loss_clip": 0.01143539,
      "auxiliary_loss_mlp": 0.01027818,
      "balance_loss_clip": 1.0485642,
      "balance_loss_mlp": 1.0210793,
      "epoch": 0.6659051283592857,
      "flos": 11509909459200.0,
      "grad_norm": 1.761456125878147,
      "language_loss": 0.72898674,
      "learning_rate": 1.060858185153821e-06,
      "loss": 0.75070035,
      "num_input_tokens_seen": 119221395,
      "step": 5538,
      "time_per_iteration": 2.500131607055664
    },
    {
      "auxiliary_loss_clip": 0.01144439,
      "auxiliary_loss_mlp": 0.01031035,
      "balance_loss_clip": 1.04721081,
      "balance_loss_mlp": 1.02286375,
      "epoch": 0.6660253712499249,
      "flos": 20594554571520.0,
      "grad_norm": 2.6321393073114376,
      "language_loss": 0.75560683,
      "learning_rate": 1.0601705064166474e-06,
      "loss": 0.77736157,
      "num_input_tokens_seen": 119239790,
      "step": 5539,
      "time_per_iteration": 2.520533800125122
    },
    {
      "auxiliary_loss_clip": 0.01133567,
      "auxiliary_loss_mlp": 0.01029114,
      "balance_loss_clip": 1.04562283,
      "balance_loss_mlp": 1.02150583,
      "epoch": 0.666145614140564,
      "flos": 21251504367360.0,
      "grad_norm": 1.9766448351208366,
      "language_loss": 0.73434651,
      "learning_rate": 1.0594829702508596e-06,
      "loss": 0.75597334,
      "num_input_tokens_seen": 119257505,
      "step": 5540,
      "time_per_iteration": 2.5094213485717773
    },
    {
      "auxiliary_loss_clip": 0.01129497,
      "auxiliary_loss_mlp": 0.0102637,
      "balance_loss_clip": 1.04587245,
      "balance_loss_mlp": 1.01968837,
      "epoch": 0.666265857031203,
      "flos": 33726188200320.0,
      "grad_norm": 1.6295548485481133,
      "language_loss": 0.55307102,
      "learning_rate": 1.0587955767607592e-06,
      "loss": 0.57462966,
      "num_input_tokens_seen": 119279365,
      "step": 5541,
      "time_per_iteration": 2.672281265258789
    },
    {
      "auxiliary_loss_clip": 0.01167411,
      "auxiliary_loss_mlp": 0.01023456,
      "balance_loss_clip": 1.04973793,
      "balance_loss_mlp": 1.01610649,
      "epoch": 0.6663860999218422,
      "flos": 17456644391040.0,
      "grad_norm": 2.3371178143932294,
      "language_loss": 0.77486169,
      "learning_rate": 1.0581083260506206e-06,
      "loss": 0.79677033,
      "num_input_tokens_seen": 119296150,
      "step": 5542,
      "time_per_iteration": 2.425196409225464
    },
    {
      "auxiliary_loss_clip": 0.01137748,
      "auxiliary_loss_mlp": 0.01025085,
      "balance_loss_clip": 1.04424655,
      "balance_loss_mlp": 1.01790833,
      "epoch": 0.6665063428124812,
      "flos": 17676740977920.0,
      "grad_norm": 2.1096521898504177,
      "language_loss": 0.76581764,
      "learning_rate": 1.0574212182246993e-06,
      "loss": 0.78744602,
      "num_input_tokens_seen": 119314845,
      "step": 5543,
      "time_per_iteration": 2.533027410507202
    },
    {
      "auxiliary_loss_clip": 0.01145583,
      "auxiliary_loss_mlp": 0.01027514,
      "balance_loss_clip": 1.04761624,
      "balance_loss_mlp": 1.0198642,
      "epoch": 0.6666265857031203,
      "flos": 27673265687040.0,
      "grad_norm": 2.490136574284313,
      "language_loss": 0.75864154,
      "learning_rate": 1.0567342533872303e-06,
      "loss": 0.7803725,
      "num_input_tokens_seen": 119334875,
      "step": 5544,
      "time_per_iteration": 2.5899250507354736
    },
    {
      "auxiliary_loss_clip": 0.01142512,
      "auxiliary_loss_mlp": 0.01031238,
      "balance_loss_clip": 1.04852867,
      "balance_loss_mlp": 1.02363825,
      "epoch": 0.6667468285937594,
      "flos": 25046831220480.0,
      "grad_norm": 1.603866791244231,
      "language_loss": 0.80763471,
      "learning_rate": 1.0560474316424255e-06,
      "loss": 0.82937217,
      "num_input_tokens_seen": 119354635,
      "step": 5545,
      "time_per_iteration": 3.3497142791748047
    },
    {
      "auxiliary_loss_clip": 0.01140372,
      "auxiliary_loss_mlp": 0.01025842,
      "balance_loss_clip": 1.04483199,
      "balance_loss_mlp": 1.01828444,
      "epoch": 0.6668670714843985,
      "flos": 22780472641920.0,
      "grad_norm": 2.1078059633374666,
      "language_loss": 0.73561567,
      "learning_rate": 1.0553607530944746e-06,
      "loss": 0.75727785,
      "num_input_tokens_seen": 119372690,
      "step": 5546,
      "time_per_iteration": 2.5736465454101562
    },
    {
      "auxiliary_loss_clip": 0.01126833,
      "auxiliary_loss_mlp": 0.01026327,
      "balance_loss_clip": 1.04361987,
      "balance_loss_mlp": 1.01900721,
      "epoch": 0.6669873143750376,
      "flos": 22163886754560.0,
      "grad_norm": 2.011932618377912,
      "language_loss": 0.89484924,
      "learning_rate": 1.0546742178475463e-06,
      "loss": 0.91638088,
      "num_input_tokens_seen": 119391685,
      "step": 5547,
      "time_per_iteration": 2.5795228481292725
    },
    {
      "auxiliary_loss_clip": 0.01116287,
      "auxiliary_loss_mlp": 0.01024101,
      "balance_loss_clip": 1.04724228,
      "balance_loss_mlp": 1.01746094,
      "epoch": 0.6671075572656767,
      "flos": 20514832335360.0,
      "grad_norm": 1.817005994615816,
      "language_loss": 0.86707288,
      "learning_rate": 1.0539878260057868e-06,
      "loss": 0.88847685,
      "num_input_tokens_seen": 119410725,
      "step": 5548,
      "time_per_iteration": 4.004964590072632
    },
    {
      "auxiliary_loss_clip": 0.01155837,
      "auxiliary_loss_mlp": 0.01031563,
      "balance_loss_clip": 1.05071735,
      "balance_loss_mlp": 1.02368307,
      "epoch": 0.6672278001563158,
      "flos": 17931203902080.0,
      "grad_norm": 2.403607840916262,
      "language_loss": 0.68259037,
      "learning_rate": 1.0533015776733226e-06,
      "loss": 0.70446438,
      "num_input_tokens_seen": 119426875,
      "step": 5549,
      "time_per_iteration": 2.450758695602417
    },
    {
      "auxiliary_loss_clip": 0.01136191,
      "auxiliary_loss_mlp": 0.01030155,
      "balance_loss_clip": 1.04714096,
      "balance_loss_mlp": 1.02224565,
      "epoch": 0.6673480430469548,
      "flos": 22342146975360.0,
      "grad_norm": 2.202187835943687,
      "language_loss": 0.78510875,
      "learning_rate": 1.0526154729542566e-06,
      "loss": 0.80677223,
      "num_input_tokens_seen": 119446935,
      "step": 5550,
      "time_per_iteration": 2.5340051651000977
    },
    {
      "auxiliary_loss_clip": 0.01128042,
      "auxiliary_loss_mlp": 0.01023529,
      "balance_loss_clip": 1.0485692,
      "balance_loss_mlp": 1.01639771,
      "epoch": 0.6674682859375939,
      "flos": 20703830722560.0,
      "grad_norm": 2.365660936632226,
      "language_loss": 0.80379283,
      "learning_rate": 1.0519295119526699e-06,
      "loss": 0.82530856,
      "num_input_tokens_seen": 119463240,
      "step": 5551,
      "time_per_iteration": 2.56020450592041
    },
    {
      "auxiliary_loss_clip": 0.01143908,
      "auxiliary_loss_mlp": 0.01023808,
      "balance_loss_clip": 1.04758704,
      "balance_loss_mlp": 1.01656914,
      "epoch": 0.667588528828233,
      "flos": 26206673379840.0,
      "grad_norm": 1.5905477189425241,
      "language_loss": 0.82656413,
      "learning_rate": 1.0512436947726227e-06,
      "loss": 0.84824127,
      "num_input_tokens_seen": 119484655,
      "step": 5552,
      "time_per_iteration": 2.5629281997680664
    },
    {
      "auxiliary_loss_clip": 0.0112667,
      "auxiliary_loss_mlp": 0.01025164,
      "balance_loss_clip": 1.04469204,
      "balance_loss_mlp": 1.01786256,
      "epoch": 0.6677087717188721,
      "flos": 23071025756160.0,
      "grad_norm": 2.5903207221299533,
      "language_loss": 0.65251601,
      "learning_rate": 1.0505580215181517e-06,
      "loss": 0.67403436,
      "num_input_tokens_seen": 119502895,
      "step": 5553,
      "time_per_iteration": 2.5495095252990723
    },
    {
      "auxiliary_loss_clip": 0.01020605,
      "auxiliary_loss_mlp": 0.01004987,
      "balance_loss_clip": 1.01227486,
      "balance_loss_mlp": 1.00391376,
      "epoch": 0.6678290146095112,
      "flos": 70941315219840.0,
      "grad_norm": 0.7965459120548644,
      "language_loss": 0.56627905,
      "learning_rate": 1.0498724922932753e-06,
      "loss": 0.58653498,
      "num_input_tokens_seen": 119561010,
      "step": 5554,
      "time_per_iteration": 3.0812716484069824
    },
    {
      "auxiliary_loss_clip": 0.01174021,
      "auxiliary_loss_mlp": 0.01028143,
      "balance_loss_clip": 1.05307794,
      "balance_loss_mlp": 1.02000129,
      "epoch": 0.6679492575001503,
      "flos": 18661088263680.0,
      "grad_norm": 2.0352660915695853,
      "language_loss": 0.86421895,
      "learning_rate": 1.0491871072019851e-06,
      "loss": 0.8862406,
      "num_input_tokens_seen": 119578900,
      "step": 5555,
      "time_per_iteration": 2.4442546367645264
    },
    {
      "auxiliary_loss_clip": 0.01129697,
      "auxiliary_loss_mlp": 0.01029453,
      "balance_loss_clip": 1.04370177,
      "balance_loss_mlp": 1.02268767,
      "epoch": 0.6680695003907894,
      "flos": 29711985822720.0,
      "grad_norm": 1.639524166169494,
      "language_loss": 0.64358938,
      "learning_rate": 1.0485018663482555e-06,
      "loss": 0.66518092,
      "num_input_tokens_seen": 119598920,
      "step": 5556,
      "time_per_iteration": 2.6116418838500977
    },
    {
      "auxiliary_loss_clip": 0.0115045,
      "auxiliary_loss_mlp": 0.01026927,
      "balance_loss_clip": 1.04860258,
      "balance_loss_mlp": 1.018857,
      "epoch": 0.6681897432814284,
      "flos": 28218964083840.0,
      "grad_norm": 2.461462032615497,
      "language_loss": 0.70598614,
      "learning_rate": 1.0478167698360354e-06,
      "loss": 0.72775996,
      "num_input_tokens_seen": 119618220,
      "step": 5557,
      "time_per_iteration": 2.5278279781341553
    },
    {
      "auxiliary_loss_clip": 0.0114518,
      "auxiliary_loss_mlp": 0.01029748,
      "balance_loss_clip": 1.04520202,
      "balance_loss_mlp": 1.02213073,
      "epoch": 0.6683099861720676,
      "flos": 25046543911680.0,
      "grad_norm": 2.021330036682852,
      "language_loss": 0.70470637,
      "learning_rate": 1.0471318177692556e-06,
      "loss": 0.72645569,
      "num_input_tokens_seen": 119638520,
      "step": 5558,
      "time_per_iteration": 2.5291197299957275
    },
    {
      "auxiliary_loss_clip": 0.01113417,
      "auxiliary_loss_mlp": 0.01023587,
      "balance_loss_clip": 1.04172218,
      "balance_loss_mlp": 1.01666713,
      "epoch": 0.6684302290627067,
      "flos": 22996977868800.0,
      "grad_norm": 2.2224188221020924,
      "language_loss": 0.75946176,
      "learning_rate": 1.046447010251821e-06,
      "loss": 0.78083169,
      "num_input_tokens_seen": 119655850,
      "step": 5559,
      "time_per_iteration": 2.604618549346924
    },
    {
      "auxiliary_loss_clip": 0.0114089,
      "auxiliary_loss_mlp": 0.01029535,
      "balance_loss_clip": 1.04927588,
      "balance_loss_mlp": 1.02283847,
      "epoch": 0.6685504719533457,
      "flos": 26573824247040.0,
      "grad_norm": 1.635632401923903,
      "language_loss": 0.75770414,
      "learning_rate": 1.0457623473876157e-06,
      "loss": 0.77940834,
      "num_input_tokens_seen": 119675355,
      "step": 5560,
      "time_per_iteration": 2.5775182247161865
    },
    {
      "auxiliary_loss_clip": 0.01165832,
      "auxiliary_loss_mlp": 0.01026843,
      "balance_loss_clip": 1.04904246,
      "balance_loss_mlp": 1.01992249,
      "epoch": 0.6686707148439849,
      "flos": 28986087870720.0,
      "grad_norm": 1.7753152940242976,
      "language_loss": 0.70947427,
      "learning_rate": 1.0450778292805046e-06,
      "loss": 0.73140103,
      "num_input_tokens_seen": 119695340,
      "step": 5561,
      "time_per_iteration": 2.49786114692688
    },
    {
      "auxiliary_loss_clip": 0.01157808,
      "auxiliary_loss_mlp": 0.01033439,
      "balance_loss_clip": 1.04860365,
      "balance_loss_mlp": 1.02635765,
      "epoch": 0.6687909577346239,
      "flos": 23623152687360.0,
      "grad_norm": 1.5127376380851054,
      "language_loss": 0.78669751,
      "learning_rate": 1.0443934560343267e-06,
      "loss": 0.80860996,
      "num_input_tokens_seen": 119716750,
      "step": 5562,
      "time_per_iteration": 3.342372179031372
    },
    {
      "auxiliary_loss_clip": 0.0111153,
      "auxiliary_loss_mlp": 0.01028714,
      "balance_loss_clip": 1.0415771,
      "balance_loss_mlp": 1.02182984,
      "epoch": 0.668911200625263,
      "flos": 23148593176320.0,
      "grad_norm": 1.9350987321996647,
      "language_loss": 0.7811296,
      "learning_rate": 1.0437092277529034e-06,
      "loss": 0.80253208,
      "num_input_tokens_seen": 119736005,
      "step": 5563,
      "time_per_iteration": 2.5839102268218994
    },
    {
      "auxiliary_loss_clip": 0.01135455,
      "auxiliary_loss_mlp": 0.01032186,
      "balance_loss_clip": 1.04546261,
      "balance_loss_mlp": 1.02523923,
      "epoch": 0.6690314435159022,
      "flos": 18551919853440.0,
      "grad_norm": 2.274694452106577,
      "language_loss": 0.73593533,
      "learning_rate": 1.0430251445400292e-06,
      "loss": 0.75761175,
      "num_input_tokens_seen": 119754050,
      "step": 5564,
      "time_per_iteration": 2.4953529834747314
    },
    {
      "auxiliary_loss_clip": 0.01068526,
      "auxiliary_loss_mlp": 0.01026962,
      "balance_loss_clip": 1.04126334,
      "balance_loss_mlp": 1.02013087,
      "epoch": 0.6691516864065412,
      "flos": 31759540704000.0,
      "grad_norm": 2.0371121135412666,
      "language_loss": 0.62822533,
      "learning_rate": 1.0423412064994787e-06,
      "loss": 0.64918017,
      "num_input_tokens_seen": 119774820,
      "step": 5565,
      "time_per_iteration": 2.9091737270355225
    },
    {
      "auxiliary_loss_clip": 0.01126024,
      "auxiliary_loss_mlp": 0.01024362,
      "balance_loss_clip": 1.04286289,
      "balance_loss_mlp": 1.0178113,
      "epoch": 0.6692719292971803,
      "flos": 34933864296960.0,
      "grad_norm": 1.7925151324497715,
      "language_loss": 0.73881519,
      "learning_rate": 1.0416574137350064e-06,
      "loss": 0.76031911,
      "num_input_tokens_seen": 119795525,
      "step": 5566,
      "time_per_iteration": 2.8942360877990723
    },
    {
      "auxiliary_loss_clip": 0.01147847,
      "auxiliary_loss_mlp": 0.01029205,
      "balance_loss_clip": 1.04772437,
      "balance_loss_mlp": 1.02204311,
      "epoch": 0.6693921721878194,
      "flos": 20449188230400.0,
      "grad_norm": 2.326676562983499,
      "language_loss": 0.80631942,
      "learning_rate": 1.0409737663503428e-06,
      "loss": 0.82808995,
      "num_input_tokens_seen": 119813905,
      "step": 5567,
      "time_per_iteration": 2.5580177307128906
    },
    {
      "auxiliary_loss_clip": 0.01150615,
      "auxiliary_loss_mlp": 0.01024114,
      "balance_loss_clip": 1.04476333,
      "balance_loss_mlp": 1.0167706,
      "epoch": 0.6695124150784585,
      "flos": 16614538963200.0,
      "grad_norm": 1.7200582538463487,
      "language_loss": 0.82848376,
      "learning_rate": 1.040290264449196e-06,
      "loss": 0.85023105,
      "num_input_tokens_seen": 119832010,
      "step": 5568,
      "time_per_iteration": 2.5544047355651855
    },
    {
      "auxiliary_loss_clip": 0.01152235,
      "auxiliary_loss_mlp": 0.01022461,
      "balance_loss_clip": 1.05041766,
      "balance_loss_mlp": 1.01566601,
      "epoch": 0.6696326579690975,
      "flos": 26652145852800.0,
      "grad_norm": 1.808884667167318,
      "language_loss": 0.64315307,
      "learning_rate": 1.0396069081352532e-06,
      "loss": 0.6649,
      "num_input_tokens_seen": 119851165,
      "step": 5569,
      "time_per_iteration": 2.6171913146972656
    },
    {
      "auxiliary_loss_clip": 0.01059124,
      "auxiliary_loss_mlp": 0.01001158,
      "balance_loss_clip": 1.01173174,
      "balance_loss_mlp": 0.9999842,
      "epoch": 0.6697529008597367,
      "flos": 66964603662720.0,
      "grad_norm": 0.7818850517624751,
      "language_loss": 0.56076711,
      "learning_rate": 1.0389236975121782e-06,
      "loss": 0.58136988,
      "num_input_tokens_seen": 119906015,
      "step": 5570,
      "time_per_iteration": 3.004574775695801
    },
    {
      "auxiliary_loss_clip": 0.01169897,
      "auxiliary_loss_mlp": 0.0102822,
      "balance_loss_clip": 1.05038977,
      "balance_loss_mlp": 1.02121341,
      "epoch": 0.6698731437503758,
      "flos": 20886939279360.0,
      "grad_norm": 2.2131459592124334,
      "language_loss": 0.71234721,
      "learning_rate": 1.0382406326836147e-06,
      "loss": 0.73432839,
      "num_input_tokens_seen": 119925160,
      "step": 5571,
      "time_per_iteration": 3.8365190029144287
    },
    {
      "auxiliary_loss_clip": 0.01160506,
      "auxiliary_loss_mlp": 0.01027352,
      "balance_loss_clip": 1.05019534,
      "balance_loss_mlp": 1.01958227,
      "epoch": 0.6699933866410148,
      "flos": 20409470766720.0,
      "grad_norm": 1.855072805673821,
      "language_loss": 0.75773358,
      "learning_rate": 1.0375577137531828e-06,
      "loss": 0.77961218,
      "num_input_tokens_seen": 119943720,
      "step": 5572,
      "time_per_iteration": 2.4996402263641357
    },
    {
      "auxiliary_loss_clip": 0.01144964,
      "auxiliary_loss_mlp": 0.01028823,
      "balance_loss_clip": 1.04877305,
      "balance_loss_mlp": 1.02191186,
      "epoch": 0.670113629531654,
      "flos": 29023075900800.0,
      "grad_norm": 1.516798380498256,
      "language_loss": 0.71746212,
      "learning_rate": 1.0368749408244802e-06,
      "loss": 0.73919994,
      "num_input_tokens_seen": 119966640,
      "step": 5573,
      "time_per_iteration": 2.609740972518921
    },
    {
      "auxiliary_loss_clip": 0.0114874,
      "auxiliary_loss_mlp": 0.01028435,
      "balance_loss_clip": 1.0479461,
      "balance_loss_mlp": 1.02145791,
      "epoch": 0.670233872422293,
      "flos": 19791699730560.0,
      "grad_norm": 1.7979942128597632,
      "language_loss": 0.78789556,
      "learning_rate": 1.0361923140010836e-06,
      "loss": 0.80966723,
      "num_input_tokens_seen": 119985125,
      "step": 5574,
      "time_per_iteration": 4.084025621414185
    },
    {
      "auxiliary_loss_clip": 0.0115747,
      "auxiliary_loss_mlp": 0.01028597,
      "balance_loss_clip": 1.04720938,
      "balance_loss_mlp": 1.02112842,
      "epoch": 0.6703541153129321,
      "flos": 24243689070720.0,
      "grad_norm": 1.8746330906275772,
      "language_loss": 0.6320945,
      "learning_rate": 1.0355098333865455e-06,
      "loss": 0.65395516,
      "num_input_tokens_seen": 120004355,
      "step": 5575,
      "time_per_iteration": 2.503755807876587
    },
    {
      "auxiliary_loss_clip": 0.01148403,
      "auxiliary_loss_mlp": 0.010303,
      "balance_loss_clip": 1.04984331,
      "balance_loss_mlp": 1.0238744,
      "epoch": 0.6704743582035713,
      "flos": 26688523351680.0,
      "grad_norm": 1.7896202502925787,
      "language_loss": 0.69266891,
      "learning_rate": 1.0348274990844006e-06,
      "loss": 0.71445596,
      "num_input_tokens_seen": 120027115,
      "step": 5576,
      "time_per_iteration": 2.558018207550049
    },
    {
      "auxiliary_loss_clip": 0.01153887,
      "auxiliary_loss_mlp": 0.01028922,
      "balance_loss_clip": 1.05035889,
      "balance_loss_mlp": 1.02213049,
      "epoch": 0.6705946010942103,
      "flos": 23514379326720.0,
      "grad_norm": 1.9743005460654202,
      "language_loss": 0.72356331,
      "learning_rate": 1.034145311198155e-06,
      "loss": 0.74539137,
      "num_input_tokens_seen": 120047130,
      "step": 5577,
      "time_per_iteration": 2.5072097778320312
    },
    {
      "auxiliary_loss_clip": 0.01165712,
      "auxiliary_loss_mlp": 0.01027056,
      "balance_loss_clip": 1.04989886,
      "balance_loss_mlp": 1.02000439,
      "epoch": 0.6707148439848494,
      "flos": 24061011477120.0,
      "grad_norm": 2.024801785567019,
      "language_loss": 0.64129579,
      "learning_rate": 1.0334632698312989e-06,
      "loss": 0.66322351,
      "num_input_tokens_seen": 120067925,
      "step": 5578,
      "time_per_iteration": 2.4741666316986084
    },
    {
      "auxiliary_loss_clip": 0.01132905,
      "auxiliary_loss_mlp": 0.0102743,
      "balance_loss_clip": 1.04525304,
      "balance_loss_mlp": 1.02024126,
      "epoch": 0.6708350868754885,
      "flos": 22528667324160.0,
      "grad_norm": 2.068881561404667,
      "language_loss": 0.75310671,
      "learning_rate": 1.032781375087295e-06,
      "loss": 0.77471006,
      "num_input_tokens_seen": 120087825,
      "step": 5579,
      "time_per_iteration": 2.53546142578125
    },
    {
      "auxiliary_loss_clip": 0.01142093,
      "auxiliary_loss_mlp": 0.01024688,
      "balance_loss_clip": 1.04964995,
      "balance_loss_mlp": 1.01820338,
      "epoch": 0.6709553297661276,
      "flos": 25227749047680.0,
      "grad_norm": 1.358191038607805,
      "language_loss": 0.67313826,
      "learning_rate": 1.0320996270695891e-06,
      "loss": 0.69480604,
      "num_input_tokens_seen": 120108895,
      "step": 5580,
      "time_per_iteration": 2.5693914890289307
    },
    {
      "auxiliary_loss_clip": 0.01126307,
      "auxiliary_loss_mlp": 0.01025477,
      "balance_loss_clip": 1.043571,
      "balance_loss_mlp": 1.01826215,
      "epoch": 0.6710755726567667,
      "flos": 20448757267200.0,
      "grad_norm": 1.6442564703658582,
      "language_loss": 0.73293674,
      "learning_rate": 1.0314180258815998e-06,
      "loss": 0.75445461,
      "num_input_tokens_seen": 120127535,
      "step": 5581,
      "time_per_iteration": 2.585735321044922
    },
    {
      "auxiliary_loss_clip": 0.01114345,
      "auxiliary_loss_mlp": 0.01028044,
      "balance_loss_clip": 1.04184961,
      "balance_loss_mlp": 1.02121592,
      "epoch": 0.6711958155474057,
      "flos": 25995411538560.0,
      "grad_norm": 1.5125267838184913,
      "language_loss": 0.74144101,
      "learning_rate": 1.0307365716267247e-06,
      "loss": 0.76286489,
      "num_input_tokens_seen": 120147980,
      "step": 5582,
      "time_per_iteration": 2.6718838214874268
    },
    {
      "auxiliary_loss_clip": 0.01152481,
      "auxiliary_loss_mlp": 0.01025938,
      "balance_loss_clip": 1.04799795,
      "balance_loss_mlp": 1.01854157,
      "epoch": 0.6713160584380449,
      "flos": 19937712516480.0,
      "grad_norm": 2.089826249419647,
      "language_loss": 0.78489351,
      "learning_rate": 1.0300552644083423e-06,
      "loss": 0.8066777,
      "num_input_tokens_seen": 120166905,
      "step": 5583,
      "time_per_iteration": 2.520772933959961
    },
    {
      "auxiliary_loss_clip": 0.01127301,
      "auxiliary_loss_mlp": 0.01033137,
      "balance_loss_clip": 1.0463779,
      "balance_loss_mlp": 1.02587712,
      "epoch": 0.6714363013286839,
      "flos": 18223373128320.0,
      "grad_norm": 2.133466655843785,
      "language_loss": 0.72036791,
      "learning_rate": 1.0293741043298036e-06,
      "loss": 0.74197233,
      "num_input_tokens_seen": 120185255,
      "step": 5584,
      "time_per_iteration": 2.5382838249206543
    },
    {
      "auxiliary_loss_clip": 0.01129243,
      "auxiliary_loss_mlp": 0.01029458,
      "balance_loss_clip": 1.05068684,
      "balance_loss_mlp": 1.02186418,
      "epoch": 0.671556544219323,
      "flos": 25812374808960.0,
      "grad_norm": 2.041853360570037,
      "language_loss": 0.71572477,
      "learning_rate": 1.0286930914944436e-06,
      "loss": 0.73731172,
      "num_input_tokens_seen": 120205070,
      "step": 5585,
      "time_per_iteration": 2.590501546859741
    },
    {
      "auxiliary_loss_clip": 0.01165305,
      "auxiliary_loss_mlp": 0.01024664,
      "balance_loss_clip": 1.04581845,
      "balance_loss_mlp": 1.01771164,
      "epoch": 0.6716767871099621,
      "flos": 15850431918720.0,
      "grad_norm": 2.599492601200961,
      "language_loss": 0.76944029,
      "learning_rate": 1.0280122260055684e-06,
      "loss": 0.79133999,
      "num_input_tokens_seen": 120220780,
      "step": 5586,
      "time_per_iteration": 2.4142444133758545
    },
    {
      "auxiliary_loss_clip": 0.01170369,
      "auxiliary_loss_mlp": 0.01033155,
      "balance_loss_clip": 1.05141199,
      "balance_loss_mlp": 1.02549613,
      "epoch": 0.6717970300006012,
      "flos": 19756112330880.0,
      "grad_norm": 2.4971675567811187,
      "language_loss": 0.82084751,
      "learning_rate": 1.0273315079664652e-06,
      "loss": 0.84288275,
      "num_input_tokens_seen": 120238735,
      "step": 5587,
      "time_per_iteration": 2.4471116065979004
    },
    {
      "auxiliary_loss_clip": 0.01158748,
      "auxiliary_loss_mlp": 0.01029917,
      "balance_loss_clip": 1.05074167,
      "balance_loss_mlp": 1.02257049,
      "epoch": 0.6719172728912403,
      "flos": 25485049146240.0,
      "grad_norm": 2.72391156285851,
      "language_loss": 0.74218911,
      "learning_rate": 1.0266509374803992e-06,
      "loss": 0.76407576,
      "num_input_tokens_seen": 120259895,
      "step": 5588,
      "time_per_iteration": 3.269397497177124
    },
    {
      "auxiliary_loss_clip": 0.01168759,
      "auxiliary_loss_mlp": 0.00761233,
      "balance_loss_clip": 1.05024743,
      "balance_loss_mlp": 1.00026107,
      "epoch": 0.6720375157818794,
      "flos": 15880344969600.0,
      "grad_norm": 2.501138013440124,
      "language_loss": 0.84629524,
      "learning_rate": 1.0259705146506123e-06,
      "loss": 0.86559522,
      "num_input_tokens_seen": 120274790,
      "step": 5589,
      "time_per_iteration": 2.433943748474121
    },
    {
      "auxiliary_loss_clip": 0.01156213,
      "auxiliary_loss_mlp": 0.01032152,
      "balance_loss_clip": 1.04870033,
      "balance_loss_mlp": 1.02483535,
      "epoch": 0.6721577586725185,
      "flos": 32010843231360.0,
      "grad_norm": 1.818213075841815,
      "language_loss": 0.77502263,
      "learning_rate": 1.025290239580324e-06,
      "loss": 0.79690629,
      "num_input_tokens_seen": 120295460,
      "step": 5590,
      "time_per_iteration": 2.5599887371063232
    },
    {
      "auxiliary_loss_clip": 0.01110054,
      "auxiliary_loss_mlp": 0.01029976,
      "balance_loss_clip": 1.04209065,
      "balance_loss_mlp": 1.02238846,
      "epoch": 0.6722780015631575,
      "flos": 20737873837440.0,
      "grad_norm": 1.590220842833504,
      "language_loss": 0.75230956,
      "learning_rate": 1.0246101123727313e-06,
      "loss": 0.77370989,
      "num_input_tokens_seen": 120314440,
      "step": 5591,
      "time_per_iteration": 2.5965328216552734
    },
    {
      "auxiliary_loss_clip": 0.011515,
      "auxiliary_loss_mlp": 0.01029133,
      "balance_loss_clip": 1.0459609,
      "balance_loss_mlp": 1.02253735,
      "epoch": 0.6723982444537967,
      "flos": 16909617191040.0,
      "grad_norm": 3.1481547306513478,
      "language_loss": 0.78647184,
      "learning_rate": 1.0239301331310085e-06,
      "loss": 0.8082782,
      "num_input_tokens_seen": 120332060,
      "step": 5592,
      "time_per_iteration": 2.453963279724121
    },
    {
      "auxiliary_loss_clip": 0.01152495,
      "auxiliary_loss_mlp": 0.01028078,
      "balance_loss_clip": 1.0494355,
      "balance_loss_mlp": 1.02129769,
      "epoch": 0.6725184873444358,
      "flos": 20667812359680.0,
      "grad_norm": 1.6689004464227335,
      "language_loss": 0.88513166,
      "learning_rate": 1.0232503019583088e-06,
      "loss": 0.90693736,
      "num_input_tokens_seen": 120351670,
      "step": 5593,
      "time_per_iteration": 2.508453130722046
    },
    {
      "auxiliary_loss_clip": 0.01149579,
      "auxiliary_loss_mlp": 0.01030868,
      "balance_loss_clip": 1.04842234,
      "balance_loss_mlp": 1.02366138,
      "epoch": 0.6726387302350748,
      "flos": 23727616416000.0,
      "grad_norm": 1.6506809819498072,
      "language_loss": 0.69840574,
      "learning_rate": 1.0225706189577619e-06,
      "loss": 0.72021019,
      "num_input_tokens_seen": 120370195,
      "step": 5594,
      "time_per_iteration": 2.5085299015045166
    },
    {
      "auxiliary_loss_clip": 0.01154909,
      "auxiliary_loss_mlp": 0.01021518,
      "balance_loss_clip": 1.05016541,
      "balance_loss_mlp": 1.01399946,
      "epoch": 0.672758973125714,
      "flos": 15188274650880.0,
      "grad_norm": 1.8722159157473302,
      "language_loss": 0.746508,
      "learning_rate": 1.021891084232475e-06,
      "loss": 0.76827222,
      "num_input_tokens_seen": 120388130,
      "step": 5595,
      "time_per_iteration": 2.47799015045166
    },
    {
      "auxiliary_loss_clip": 0.01153065,
      "auxiliary_loss_mlp": 0.01027947,
      "balance_loss_clip": 1.04717815,
      "balance_loss_mlp": 1.02005827,
      "epoch": 0.672879216016353,
      "flos": 18077252601600.0,
      "grad_norm": 2.740297985665008,
      "language_loss": 0.79495478,
      "learning_rate": 1.0212116978855325e-06,
      "loss": 0.81676483,
      "num_input_tokens_seen": 120406145,
      "step": 5596,
      "time_per_iteration": 2.4593453407287598
    },
    {
      "auxiliary_loss_clip": 0.01125024,
      "auxiliary_loss_mlp": 0.01022758,
      "balance_loss_clip": 1.04653907,
      "balance_loss_mlp": 1.01580167,
      "epoch": 0.6729994589069921,
      "flos": 23476349802240.0,
      "grad_norm": 1.6393770773005563,
      "language_loss": 0.78784627,
      "learning_rate": 1.020532460019997e-06,
      "loss": 0.80932409,
      "num_input_tokens_seen": 120425395,
      "step": 5597,
      "time_per_iteration": 3.361593246459961
    },
    {
      "auxiliary_loss_clip": 0.01087678,
      "auxiliary_loss_mlp": 0.01033343,
      "balance_loss_clip": 1.0431869,
      "balance_loss_mlp": 1.02625871,
      "epoch": 0.6731197017976313,
      "flos": 26322018929280.0,
      "grad_norm": 1.815399924540102,
      "language_loss": 0.70913053,
      "learning_rate": 1.0198533707389096e-06,
      "loss": 0.73034072,
      "num_input_tokens_seen": 120446270,
      "step": 5598,
      "time_per_iteration": 2.6984219551086426
    },
    {
      "auxiliary_loss_clip": 0.01150891,
      "auxiliary_loss_mlp": 0.00761015,
      "balance_loss_clip": 1.04896367,
      "balance_loss_mlp": 1.00022411,
      "epoch": 0.6732399446882703,
      "flos": 21616428591360.0,
      "grad_norm": 1.8458204456577636,
      "language_loss": 0.73297977,
      "learning_rate": 1.0191744301452853e-06,
      "loss": 0.75209892,
      "num_input_tokens_seen": 120465570,
      "step": 5599,
      "time_per_iteration": 2.5050244331359863
    },
    {
      "auxiliary_loss_clip": 0.0116548,
      "auxiliary_loss_mlp": 0.01029399,
      "balance_loss_clip": 1.04830146,
      "balance_loss_mlp": 1.0227828,
      "epoch": 0.6733601875789094,
      "flos": 25880173729920.0,
      "grad_norm": 1.8568092586394664,
      "language_loss": 0.70204842,
      "learning_rate": 1.0184956383421208e-06,
      "loss": 0.72399724,
      "num_input_tokens_seen": 120484220,
      "step": 5600,
      "time_per_iteration": 3.33374285697937
    },
    {
      "auxiliary_loss_clip": 0.01156555,
      "auxiliary_loss_mlp": 0.01022352,
      "balance_loss_clip": 1.04980803,
      "balance_loss_mlp": 1.01524127,
      "epoch": 0.6734804304695485,
      "flos": 22929573997440.0,
      "grad_norm": 1.8940662409910956,
      "language_loss": 0.65173364,
      "learning_rate": 1.017816995432387e-06,
      "loss": 0.67352271,
      "num_input_tokens_seen": 120503320,
      "step": 5601,
      "time_per_iteration": 2.4926583766937256
    },
    {
      "auxiliary_loss_clip": 0.01139294,
      "auxiliary_loss_mlp": 0.01026941,
      "balance_loss_clip": 1.04849255,
      "balance_loss_mlp": 1.01992226,
      "epoch": 0.6736006733601876,
      "flos": 18697968552960.0,
      "grad_norm": 1.819085015517038,
      "language_loss": 0.7397477,
      "learning_rate": 1.0171385015190353e-06,
      "loss": 0.76141006,
      "num_input_tokens_seen": 120523180,
      "step": 5602,
      "time_per_iteration": 2.548186779022217
    },
    {
      "auxiliary_loss_clip": 0.01131634,
      "auxiliary_loss_mlp": 0.00760118,
      "balance_loss_clip": 1.04677713,
      "balance_loss_mlp": 1.0002197,
      "epoch": 0.6737209162508266,
      "flos": 19427745173760.0,
      "grad_norm": 2.025685980620832,
      "language_loss": 0.73151445,
      "learning_rate": 1.0164601567049908e-06,
      "loss": 0.75043201,
      "num_input_tokens_seen": 120541710,
      "step": 5603,
      "time_per_iteration": 2.5246217250823975
    },
    {
      "auxiliary_loss_clip": 0.011387,
      "auxiliary_loss_mlp": 0.0103013,
      "balance_loss_clip": 1.04737306,
      "balance_loss_mlp": 1.02271533,
      "epoch": 0.6738411591414658,
      "flos": 20158060498560.0,
      "grad_norm": 1.646592162129762,
      "language_loss": 0.80185032,
      "learning_rate": 1.015781961093158e-06,
      "loss": 0.8235386,
      "num_input_tokens_seen": 120561030,
      "step": 5604,
      "time_per_iteration": 2.5261034965515137
    },
    {
      "auxiliary_loss_clip": 0.01141742,
      "auxiliary_loss_mlp": 0.01027064,
      "balance_loss_clip": 1.04388463,
      "balance_loss_mlp": 1.02017331,
      "epoch": 0.6739614020321049,
      "flos": 21653847584640.0,
      "grad_norm": 1.60725864575364,
      "language_loss": 0.76845413,
      "learning_rate": 1.0151039147864197e-06,
      "loss": 0.79014218,
      "num_input_tokens_seen": 120581005,
      "step": 5605,
      "time_per_iteration": 2.5183680057525635
    },
    {
      "auxiliary_loss_clip": 0.0107893,
      "auxiliary_loss_mlp": 0.01030354,
      "balance_loss_clip": 1.04428983,
      "balance_loss_mlp": 1.02276301,
      "epoch": 0.6740816449227439,
      "flos": 19171702051200.0,
      "grad_norm": 2.200586647834062,
      "language_loss": 0.65946084,
      "learning_rate": 1.0144260178876336e-06,
      "loss": 0.68055367,
      "num_input_tokens_seen": 120600350,
      "step": 5606,
      "time_per_iteration": 2.8401858806610107
    },
    {
      "auxiliary_loss_clip": 0.01145269,
      "auxiliary_loss_mlp": 0.01028649,
      "balance_loss_clip": 1.0471251,
      "balance_loss_mlp": 1.02190518,
      "epoch": 0.6742018878133831,
      "flos": 21097015971840.0,
      "grad_norm": 2.739664168652078,
      "language_loss": 0.67630696,
      "learning_rate": 1.0137482704996388e-06,
      "loss": 0.69804615,
      "num_input_tokens_seen": 120614700,
      "step": 5607,
      "time_per_iteration": 2.7237963676452637
    },
    {
      "auxiliary_loss_clip": 0.01131641,
      "auxiliary_loss_mlp": 0.01029267,
      "balance_loss_clip": 1.0475843,
      "balance_loss_mlp": 1.02177787,
      "epoch": 0.6743221307040221,
      "flos": 23549966726400.0,
      "grad_norm": 3.0630658034956038,
      "language_loss": 0.78620422,
      "learning_rate": 1.0130706727252461e-06,
      "loss": 0.80781329,
      "num_input_tokens_seen": 120631755,
      "step": 5608,
      "time_per_iteration": 2.571286916732788
    },
    {
      "auxiliary_loss_clip": 0.0112788,
      "auxiliary_loss_mlp": 0.01026673,
      "balance_loss_clip": 1.04521418,
      "balance_loss_mlp": 1.0195291,
      "epoch": 0.6744423735946612,
      "flos": 16249542912000.0,
      "grad_norm": 5.649036952668872,
      "language_loss": 0.68030095,
      "learning_rate": 1.0123932246672468e-06,
      "loss": 0.70184642,
      "num_input_tokens_seen": 120645900,
      "step": 5609,
      "time_per_iteration": 2.511312246322632
    },
    {
      "auxiliary_loss_clip": 0.01019744,
      "auxiliary_loss_mlp": 0.00751181,
      "balance_loss_clip": 1.01267087,
      "balance_loss_mlp": 1.00007105,
      "epoch": 0.6745626164853004,
      "flos": 57843257829120.0,
      "grad_norm": 0.7500119426160425,
      "language_loss": 0.55882072,
      "learning_rate": 1.0117159264284114e-06,
      "loss": 0.57652998,
      "num_input_tokens_seen": 120709070,
      "step": 5610,
      "time_per_iteration": 3.135748863220215
    },
    {
      "auxiliary_loss_clip": 0.01141336,
      "auxiliary_loss_mlp": 0.01031122,
      "balance_loss_clip": 1.04784787,
      "balance_loss_mlp": 1.02424979,
      "epoch": 0.6746828593759394,
      "flos": 20485027025280.0,
      "grad_norm": 1.6642417962332647,
      "language_loss": 0.77243227,
      "learning_rate": 1.0110387781114837e-06,
      "loss": 0.79415685,
      "num_input_tokens_seen": 120727685,
      "step": 5611,
      "time_per_iteration": 2.5182440280914307
    },
    {
      "auxiliary_loss_clip": 0.01166848,
      "auxiliary_loss_mlp": 0.01026538,
      "balance_loss_clip": 1.04969597,
      "balance_loss_mlp": 1.0194931,
      "epoch": 0.6748031022665785,
      "flos": 19208223204480.0,
      "grad_norm": 1.9051573700446875,
      "language_loss": 0.77147752,
      "learning_rate": 1.0103617798191872e-06,
      "loss": 0.79341137,
      "num_input_tokens_seen": 120747160,
      "step": 5612,
      "time_per_iteration": 2.4525206089019775
    },
    {
      "auxiliary_loss_clip": 0.0113383,
      "auxiliary_loss_mlp": 0.01026264,
      "balance_loss_clip": 1.0474391,
      "balance_loss_mlp": 1.01931977,
      "epoch": 0.6749233451572175,
      "flos": 15195026407680.0,
      "grad_norm": 2.231671946629498,
      "language_loss": 0.82251477,
      "learning_rate": 1.0096849316542217e-06,
      "loss": 0.84411573,
      "num_input_tokens_seen": 120763710,
      "step": 5613,
      "time_per_iteration": 3.5578079223632812
    },
    {
      "auxiliary_loss_clip": 0.01071236,
      "auxiliary_loss_mlp": 0.01025955,
      "balance_loss_clip": 1.03774261,
      "balance_loss_mlp": 1.01886189,
      "epoch": 0.6750435880478567,
      "flos": 26499489050880.0,
      "grad_norm": 2.2122038519486784,
      "language_loss": 0.74620622,
      "learning_rate": 1.0090082337192643e-06,
      "loss": 0.76717818,
      "num_input_tokens_seen": 120783355,
      "step": 5614,
      "time_per_iteration": 2.7532026767730713
    },
    {
      "auxiliary_loss_clip": 0.01091725,
      "auxiliary_loss_mlp": 0.01029368,
      "balance_loss_clip": 1.03738201,
      "balance_loss_mlp": 1.02271271,
      "epoch": 0.6751638309384957,
      "flos": 23404313076480.0,
      "grad_norm": 2.344904097156709,
      "language_loss": 0.78604239,
      "learning_rate": 1.0083316861169705e-06,
      "loss": 0.80725336,
      "num_input_tokens_seen": 120802090,
      "step": 5615,
      "time_per_iteration": 2.7281432151794434
    },
    {
      "auxiliary_loss_clip": 0.01130988,
      "auxiliary_loss_mlp": 0.01026237,
      "balance_loss_clip": 1.04426432,
      "balance_loss_mlp": 1.01837492,
      "epoch": 0.6752840738291348,
      "flos": 23441408847360.0,
      "grad_norm": 2.1742020250792655,
      "language_loss": 0.71401656,
      "learning_rate": 1.0076552889499713e-06,
      "loss": 0.73558879,
      "num_input_tokens_seen": 120822855,
      "step": 5616,
      "time_per_iteration": 2.615874767303467
    },
    {
      "auxiliary_loss_clip": 0.01153011,
      "auxiliary_loss_mlp": 0.01026031,
      "balance_loss_clip": 1.05025864,
      "balance_loss_mlp": 1.01874471,
      "epoch": 0.675404316719774,
      "flos": 30335826257280.0,
      "grad_norm": 1.8079908130409788,
      "language_loss": 0.73238385,
      "learning_rate": 1.006979042320876e-06,
      "loss": 0.75417423,
      "num_input_tokens_seen": 120843070,
      "step": 5617,
      "time_per_iteration": 2.561631917953491
    },
    {
      "auxiliary_loss_clip": 0.01133962,
      "auxiliary_loss_mlp": 0.01025517,
      "balance_loss_clip": 1.04316115,
      "balance_loss_mlp": 1.01877856,
      "epoch": 0.675524559610413,
      "flos": 23622613983360.0,
      "grad_norm": 1.9926272994884273,
      "language_loss": 0.62958562,
      "learning_rate": 1.0063029463322702e-06,
      "loss": 0.65118039,
      "num_input_tokens_seen": 120863345,
      "step": 5618,
      "time_per_iteration": 2.5540215969085693
    },
    {
      "auxiliary_loss_clip": 0.01103814,
      "auxiliary_loss_mlp": 0.00761173,
      "balance_loss_clip": 1.04059279,
      "balance_loss_mlp": 1.00025094,
      "epoch": 0.6756448025010521,
      "flos": 21248631279360.0,
      "grad_norm": 2.00728576785247,
      "language_loss": 0.75259781,
      "learning_rate": 1.0056270010867164e-06,
      "loss": 0.77124774,
      "num_input_tokens_seen": 120880915,
      "step": 5619,
      "time_per_iteration": 2.592349052429199
    },
    {
      "auxiliary_loss_clip": 0.01140104,
      "auxiliary_loss_mlp": 0.01026396,
      "balance_loss_clip": 1.04412746,
      "balance_loss_mlp": 1.01889455,
      "epoch": 0.6757650453916912,
      "flos": 21646521210240.0,
      "grad_norm": 2.3212290121060044,
      "language_loss": 0.78090179,
      "learning_rate": 1.004951206686758e-06,
      "loss": 0.80256677,
      "num_input_tokens_seen": 120899190,
      "step": 5620,
      "time_per_iteration": 2.5200958251953125
    },
    {
      "auxiliary_loss_clip": 0.01148898,
      "auxiliary_loss_mlp": 0.01028982,
      "balance_loss_clip": 1.04861355,
      "balance_loss_mlp": 1.02157283,
      "epoch": 0.6758852882823303,
      "flos": 21795658479360.0,
      "grad_norm": 1.9137869815505157,
      "language_loss": 0.71545345,
      "learning_rate": 1.0042755632349087e-06,
      "loss": 0.73723221,
      "num_input_tokens_seen": 120916080,
      "step": 5621,
      "time_per_iteration": 2.4919209480285645
    },
    {
      "auxiliary_loss_clip": 0.01126203,
      "auxiliary_loss_mlp": 0.01036462,
      "balance_loss_clip": 1.04683232,
      "balance_loss_mlp": 1.02886236,
      "epoch": 0.6760055311729694,
      "flos": 27088783580160.0,
      "grad_norm": 2.158112664638395,
      "language_loss": 0.62524354,
      "learning_rate": 1.0036000708336653e-06,
      "loss": 0.64687026,
      "num_input_tokens_seen": 120935210,
      "step": 5622,
      "time_per_iteration": 2.587871789932251
    },
    {
      "auxiliary_loss_clip": 0.01143138,
      "auxiliary_loss_mlp": 0.01029442,
      "balance_loss_clip": 1.04742992,
      "balance_loss_mlp": 1.02244687,
      "epoch": 0.6761257740636085,
      "flos": 17999792922240.0,
      "grad_norm": 2.0248863099934455,
      "language_loss": 0.79795134,
      "learning_rate": 1.0029247295854984e-06,
      "loss": 0.81967723,
      "num_input_tokens_seen": 120951830,
      "step": 5623,
      "time_per_iteration": 3.287555456161499
    },
    {
      "auxiliary_loss_clip": 0.01130323,
      "auxiliary_loss_mlp": 0.01027931,
      "balance_loss_clip": 1.0457654,
      "balance_loss_mlp": 1.02122295,
      "epoch": 0.6762460169542476,
      "flos": 15121912273920.0,
      "grad_norm": 1.8985606912453017,
      "language_loss": 0.71915501,
      "learning_rate": 1.0022495395928588e-06,
      "loss": 0.74073756,
      "num_input_tokens_seen": 120970310,
      "step": 5624,
      "time_per_iteration": 2.5237278938293457
    },
    {
      "auxiliary_loss_clip": 0.01058928,
      "auxiliary_loss_mlp": 0.01007389,
      "balance_loss_clip": 1.01137948,
      "balance_loss_mlp": 1.00629854,
      "epoch": 0.6763662598448866,
      "flos": 67886970030720.0,
      "grad_norm": 0.7941553470007131,
      "language_loss": 0.62379301,
      "learning_rate": 1.0015745009581697e-06,
      "loss": 0.64445621,
      "num_input_tokens_seen": 121031915,
      "step": 5625,
      "time_per_iteration": 3.103691577911377
    },
    {
      "auxiliary_loss_clip": 0.01151692,
      "auxiliary_loss_mlp": 0.01032366,
      "balance_loss_clip": 1.05043554,
      "balance_loss_mlp": 1.0252161,
      "epoch": 0.6764865027355258,
      "flos": 20631829910400.0,
      "grad_norm": 1.6626226566674922,
      "language_loss": 0.66980475,
      "learning_rate": 1.0008996137838343e-06,
      "loss": 0.69164538,
      "num_input_tokens_seen": 121050890,
      "step": 5626,
      "time_per_iteration": 3.266974687576294
    },
    {
      "auxiliary_loss_clip": 0.01171812,
      "auxiliary_loss_mlp": 0.0102789,
      "balance_loss_clip": 1.051368,
      "balance_loss_mlp": 1.0195899,
      "epoch": 0.6766067456261649,
      "flos": 21215809226880.0,
      "grad_norm": 2.382799619742517,
      "language_loss": 0.79655623,
      "learning_rate": 1.000224878172234e-06,
      "loss": 0.81855333,
      "num_input_tokens_seen": 121070015,
      "step": 5627,
      "time_per_iteration": 2.4623026847839355
    },
    {
      "auxiliary_loss_clip": 0.01153835,
      "auxiliary_loss_mlp": 0.01026899,
      "balance_loss_clip": 1.04951334,
      "balance_loss_mlp": 1.02007461,
      "epoch": 0.6767269885168039,
      "flos": 19938251220480.0,
      "grad_norm": 2.5316718897659283,
      "language_loss": 0.72528887,
      "learning_rate": 9.99550294225724e-07,
      "loss": 0.74709624,
      "num_input_tokens_seen": 121089170,
      "step": 5628,
      "time_per_iteration": 2.503897190093994
    },
    {
      "auxiliary_loss_clip": 0.01109311,
      "auxiliary_loss_mlp": 0.01024598,
      "balance_loss_clip": 1.03917241,
      "balance_loss_mlp": 1.01746976,
      "epoch": 0.6768472314074431,
      "flos": 20814076540800.0,
      "grad_norm": 3.2648796698543703,
      "language_loss": 0.72131795,
      "learning_rate": 9.988758620466402e-07,
      "loss": 0.74265707,
      "num_input_tokens_seen": 121108040,
      "step": 5629,
      "time_per_iteration": 2.570328712463379
    },
    {
      "auxiliary_loss_clip": 0.0109918,
      "auxiliary_loss_mlp": 0.01026491,
      "balance_loss_clip": 1.04356527,
      "balance_loss_mlp": 1.01983666,
      "epoch": 0.6769674742980821,
      "flos": 23186012169600.0,
      "grad_norm": 1.5500317886842934,
      "language_loss": 0.76286632,
      "learning_rate": 9.982015817372917e-07,
      "loss": 0.78412306,
      "num_input_tokens_seen": 121128480,
      "step": 5630,
      "time_per_iteration": 2.642570734024048
    },
    {
      "auxiliary_loss_clip": 0.01108442,
      "auxiliary_loss_mlp": 0.01028324,
      "balance_loss_clip": 1.04074121,
      "balance_loss_mlp": 1.02120769,
      "epoch": 0.6770877171887212,
      "flos": 24242934885120.0,
      "grad_norm": 1.8950674540593428,
      "language_loss": 0.82045722,
      "learning_rate": 9.975274533999657e-07,
      "loss": 0.84182489,
      "num_input_tokens_seen": 121148010,
      "step": 5631,
      "time_per_iteration": 2.597531318664551
    },
    {
      "auxiliary_loss_clip": 0.01166663,
      "auxiliary_loss_mlp": 0.01030726,
      "balance_loss_clip": 1.04847169,
      "balance_loss_mlp": 1.02273893,
      "epoch": 0.6772079600793603,
      "flos": 18141567903360.0,
      "grad_norm": 2.500745267480339,
      "language_loss": 0.83791614,
      "learning_rate": 9.96853477136929e-07,
      "loss": 0.8598901,
      "num_input_tokens_seen": 121162755,
      "step": 5632,
      "time_per_iteration": 2.421985387802124
    },
    {
      "auxiliary_loss_clip": 0.01114386,
      "auxiliary_loss_mlp": 0.01027447,
      "balance_loss_clip": 1.04152417,
      "balance_loss_mlp": 1.02066433,
      "epoch": 0.6773282029699994,
      "flos": 22452069571200.0,
      "grad_norm": 2.076387287775808,
      "language_loss": 0.7533766,
      "learning_rate": 9.96179653050422e-07,
      "loss": 0.77479494,
      "num_input_tokens_seen": 121182915,
      "step": 5633,
      "time_per_iteration": 2.5609376430511475
    },
    {
      "auxiliary_loss_clip": 0.01117954,
      "auxiliary_loss_mlp": 0.01034275,
      "balance_loss_clip": 1.04390585,
      "balance_loss_mlp": 1.02722406,
      "epoch": 0.6774484458606385,
      "flos": 18693730748160.0,
      "grad_norm": 2.107809326297555,
      "language_loss": 0.74360967,
      "learning_rate": 9.955059812426635e-07,
      "loss": 0.76513195,
      "num_input_tokens_seen": 121200445,
      "step": 5634,
      "time_per_iteration": 2.5468695163726807
    },
    {
      "auxiliary_loss_clip": 0.01166235,
      "auxiliary_loss_mlp": 0.01025762,
      "balance_loss_clip": 1.05083668,
      "balance_loss_mlp": 1.01894021,
      "epoch": 0.6775686887512776,
      "flos": 25994046821760.0,
      "grad_norm": 2.055614949299435,
      "language_loss": 0.82714647,
      "learning_rate": 9.948324618158493e-07,
      "loss": 0.84906644,
      "num_input_tokens_seen": 121220785,
      "step": 5635,
      "time_per_iteration": 2.494586944580078
    },
    {
      "auxiliary_loss_clip": 0.01153981,
      "auxiliary_loss_mlp": 0.01031252,
      "balance_loss_clip": 1.04545069,
      "balance_loss_mlp": 1.02315736,
      "epoch": 0.6776889316419167,
      "flos": 13587987922560.0,
      "grad_norm": 2.397582094042597,
      "language_loss": 0.77526677,
      "learning_rate": 9.941590948721502e-07,
      "loss": 0.79711914,
      "num_input_tokens_seen": 121237985,
      "step": 5636,
      "time_per_iteration": 2.4574460983276367
    },
    {
      "auxiliary_loss_clip": 0.01134762,
      "auxiliary_loss_mlp": 0.01024136,
      "balance_loss_clip": 1.0475297,
      "balance_loss_mlp": 1.01739156,
      "epoch": 0.6778091745325557,
      "flos": 27601121220480.0,
      "grad_norm": 1.7956688914882935,
      "language_loss": 0.76217723,
      "learning_rate": 9.934858805137188e-07,
      "loss": 0.78376621,
      "num_input_tokens_seen": 121258635,
      "step": 5637,
      "time_per_iteration": 2.5655081272125244
    },
    {
      "auxiliary_loss_clip": 0.01149741,
      "auxiliary_loss_mlp": 0.01029855,
      "balance_loss_clip": 1.04762554,
      "balance_loss_mlp": 1.02309632,
      "epoch": 0.6779294174231949,
      "flos": 18734058743040.0,
      "grad_norm": 1.6172305271493224,
      "language_loss": 0.80971402,
      "learning_rate": 9.92812818842677e-07,
      "loss": 0.83150995,
      "num_input_tokens_seen": 121277810,
      "step": 5638,
      "time_per_iteration": 2.5537209510803223
    },
    {
      "auxiliary_loss_clip": 0.01146534,
      "auxiliary_loss_mlp": 0.01026884,
      "balance_loss_clip": 1.04530311,
      "balance_loss_mlp": 1.02003813,
      "epoch": 0.678049660313834,
      "flos": 45873797765760.0,
      "grad_norm": 1.6179425309043678,
      "language_loss": 0.64164066,
      "learning_rate": 9.921399099611306e-07,
      "loss": 0.6633749,
      "num_input_tokens_seen": 121298975,
      "step": 5639,
      "time_per_iteration": 3.4619028568267822
    },
    {
      "auxiliary_loss_clip": 0.01141515,
      "auxiliary_loss_mlp": 0.01028127,
      "balance_loss_clip": 1.04769456,
      "balance_loss_mlp": 1.02082539,
      "epoch": 0.678169903204473,
      "flos": 19974556892160.0,
      "grad_norm": 1.6530889437661458,
      "language_loss": 0.68850207,
      "learning_rate": 9.914671539711588e-07,
      "loss": 0.71019852,
      "num_input_tokens_seen": 121318495,
      "step": 5640,
      "time_per_iteration": 2.562713146209717
    },
    {
      "auxiliary_loss_clip": 0.0107122,
      "auxiliary_loss_mlp": 0.00761056,
      "balance_loss_clip": 1.04218102,
      "balance_loss_mlp": 1.00023866,
      "epoch": 0.6782901460951122,
      "flos": 21395613732480.0,
      "grad_norm": 1.8342938812211296,
      "language_loss": 0.78478271,
      "learning_rate": 9.90794550974817e-07,
      "loss": 0.80310547,
      "num_input_tokens_seen": 121338890,
      "step": 5641,
      "time_per_iteration": 2.9581198692321777
    },
    {
      "auxiliary_loss_clip": 0.0112114,
      "auxiliary_loss_mlp": 0.01026589,
      "balance_loss_clip": 1.04469383,
      "balance_loss_mlp": 1.01955545,
      "epoch": 0.6784103889857512,
      "flos": 21434002392960.0,
      "grad_norm": 2.7032532320695712,
      "language_loss": 0.81421232,
      "learning_rate": 9.901221010741407e-07,
      "loss": 0.83568966,
      "num_input_tokens_seen": 121358210,
      "step": 5642,
      "time_per_iteration": 2.758780002593994
    },
    {
      "auxiliary_loss_clip": 0.01157657,
      "auxiliary_loss_mlp": 0.01025372,
      "balance_loss_clip": 1.04931164,
      "balance_loss_mlp": 1.01782298,
      "epoch": 0.6785306318763903,
      "flos": 32671923091200.0,
      "grad_norm": 2.1053609432870197,
      "language_loss": 0.74695045,
      "learning_rate": 9.894498043711375e-07,
      "loss": 0.76878071,
      "num_input_tokens_seen": 121379955,
      "step": 5643,
      "time_per_iteration": 2.588226079940796
    },
    {
      "auxiliary_loss_clip": 0.01135889,
      "auxiliary_loss_mlp": 0.01025074,
      "balance_loss_clip": 1.04364872,
      "balance_loss_mlp": 1.01797795,
      "epoch": 0.6786508747670293,
      "flos": 25632139340160.0,
      "grad_norm": 1.8348625494043846,
      "language_loss": 0.69369376,
      "learning_rate": 9.887776609677962e-07,
      "loss": 0.71530342,
      "num_input_tokens_seen": 121401325,
      "step": 5644,
      "time_per_iteration": 2.5913493633270264
    },
    {
      "auxiliary_loss_clip": 0.01114779,
      "auxiliary_loss_mlp": 0.01031371,
      "balance_loss_clip": 1.04068661,
      "balance_loss_mlp": 1.02455783,
      "epoch": 0.6787711176576685,
      "flos": 19171881619200.0,
      "grad_norm": 1.5922417622315646,
      "language_loss": 0.72441834,
      "learning_rate": 9.88105670966079e-07,
      "loss": 0.74587983,
      "num_input_tokens_seen": 121419785,
      "step": 5645,
      "time_per_iteration": 2.538856267929077
    },
    {
      "auxiliary_loss_clip": 0.01098953,
      "auxiliary_loss_mlp": 0.01020493,
      "balance_loss_clip": 1.04312825,
      "balance_loss_mlp": 1.0136919,
      "epoch": 0.6788913605483076,
      "flos": 13985159581440.0,
      "grad_norm": 1.8505555875579167,
      "language_loss": 0.78777945,
      "learning_rate": 9.874338344679283e-07,
      "loss": 0.80897391,
      "num_input_tokens_seen": 121435630,
      "step": 5646,
      "time_per_iteration": 2.5564157962799072
    },
    {
      "auxiliary_loss_clip": 0.01162691,
      "auxiliary_loss_mlp": 0.01024419,
      "balance_loss_clip": 1.04840326,
      "balance_loss_mlp": 1.01760042,
      "epoch": 0.6790116034389466,
      "flos": 22017586659840.0,
      "grad_norm": 1.7265046607893921,
      "language_loss": 0.73876077,
      "learning_rate": 9.86762151575259e-07,
      "loss": 0.76063186,
      "num_input_tokens_seen": 121455625,
      "step": 5647,
      "time_per_iteration": 2.471052408218384
    },
    {
      "auxiliary_loss_clip": 0.01111432,
      "auxiliary_loss_mlp": 0.00759518,
      "balance_loss_clip": 1.04590297,
      "balance_loss_mlp": 1.00021648,
      "epoch": 0.6791318463295858,
      "flos": 20922454851840.0,
      "grad_norm": 1.4570120984141495,
      "language_loss": 0.80214572,
      "learning_rate": 9.860906223899651e-07,
      "loss": 0.82085526,
      "num_input_tokens_seen": 121475020,
      "step": 5648,
      "time_per_iteration": 2.593214273452759
    },
    {
      "auxiliary_loss_clip": 0.01143552,
      "auxiliary_loss_mlp": 0.01028544,
      "balance_loss_clip": 1.04554212,
      "balance_loss_mlp": 1.02143335,
      "epoch": 0.6792520892202248,
      "flos": 28512749422080.0,
      "grad_norm": 1.5987802355111425,
      "language_loss": 0.75749528,
      "learning_rate": 9.854192470139184e-07,
      "loss": 0.77921629,
      "num_input_tokens_seen": 121496500,
      "step": 5649,
      "time_per_iteration": 3.3567605018615723
    },
    {
      "auxiliary_loss_clip": 0.01141105,
      "auxiliary_loss_mlp": 0.01038676,
      "balance_loss_clip": 1.04885244,
      "balance_loss_mlp": 1.03114772,
      "epoch": 0.6793723321108639,
      "flos": 20011904058240.0,
      "grad_norm": 2.2480835734300677,
      "language_loss": 0.71778071,
      "learning_rate": 9.847480255489645e-07,
      "loss": 0.73957849,
      "num_input_tokens_seen": 121515525,
      "step": 5650,
      "time_per_iteration": 2.5109007358551025
    },
    {
      "auxiliary_loss_clip": 0.01143977,
      "auxiliary_loss_mlp": 0.01028748,
      "balance_loss_clip": 1.0471828,
      "balance_loss_mlp": 1.02187002,
      "epoch": 0.6794925750015031,
      "flos": 26649488246400.0,
      "grad_norm": 1.6849769833165895,
      "language_loss": 0.68998837,
      "learning_rate": 9.840769580969295e-07,
      "loss": 0.71171564,
      "num_input_tokens_seen": 121535965,
      "step": 5651,
      "time_per_iteration": 2.559161424636841
    },
    {
      "auxiliary_loss_clip": 0.01143676,
      "auxiliary_loss_mlp": 0.01024288,
      "balance_loss_clip": 1.04555345,
      "balance_loss_mlp": 1.01742148,
      "epoch": 0.6796128178921421,
      "flos": 21580374314880.0,
      "grad_norm": 1.9116444023774068,
      "language_loss": 0.80077863,
      "learning_rate": 9.834060447596114e-07,
      "loss": 0.82245827,
      "num_input_tokens_seen": 121555235,
      "step": 5652,
      "time_per_iteration": 4.22889518737793
    },
    {
      "auxiliary_loss_clip": 0.01153827,
      "auxiliary_loss_mlp": 0.01025875,
      "balance_loss_clip": 1.0461266,
      "balance_loss_mlp": 1.01893377,
      "epoch": 0.6797330607827812,
      "flos": 22492002516480.0,
      "grad_norm": 2.2272921769026235,
      "language_loss": 0.78169626,
      "learning_rate": 9.827352856387868e-07,
      "loss": 0.80349326,
      "num_input_tokens_seen": 121574945,
      "step": 5653,
      "time_per_iteration": 2.547220468521118
    },
    {
      "auxiliary_loss_clip": 0.01014504,
      "auxiliary_loss_mlp": 0.01003005,
      "balance_loss_clip": 1.01182508,
      "balance_loss_mlp": 1.00196159,
      "epoch": 0.6798533036734203,
      "flos": 66306648286080.0,
      "grad_norm": 0.7823738579120485,
      "language_loss": 0.6431722,
      "learning_rate": 9.820646808362118e-07,
      "loss": 0.6633473,
      "num_input_tokens_seen": 121641200,
      "step": 5654,
      "time_per_iteration": 3.264039993286133
    },
    {
      "auxiliary_loss_clip": 0.01138134,
      "auxiliary_loss_mlp": 0.01029236,
      "balance_loss_clip": 1.04973948,
      "balance_loss_mlp": 1.02234256,
      "epoch": 0.6799735465640594,
      "flos": 16180163792640.0,
      "grad_norm": 2.6049060244326903,
      "language_loss": 0.72586119,
      "learning_rate": 9.813942304536154e-07,
      "loss": 0.74753493,
      "num_input_tokens_seen": 121659170,
      "step": 5655,
      "time_per_iteration": 2.501434803009033
    },
    {
      "auxiliary_loss_clip": 0.01140055,
      "auxiliary_loss_mlp": 0.01026507,
      "balance_loss_clip": 1.04576862,
      "balance_loss_mlp": 1.01952767,
      "epoch": 0.6800937894546984,
      "flos": 22125749489280.0,
      "grad_norm": 1.7989801559473582,
      "language_loss": 0.63698876,
      "learning_rate": 9.807239345927043e-07,
      "loss": 0.65865433,
      "num_input_tokens_seen": 121679180,
      "step": 5656,
      "time_per_iteration": 2.536226987838745
    },
    {
      "auxiliary_loss_clip": 0.01138536,
      "auxiliary_loss_mlp": 0.01028155,
      "balance_loss_clip": 1.04438472,
      "balance_loss_mlp": 1.02127934,
      "epoch": 0.6802140323453376,
      "flos": 31612953300480.0,
      "grad_norm": 2.0321165603182494,
      "language_loss": 0.72403836,
      "learning_rate": 9.80053793355162e-07,
      "loss": 0.74570525,
      "num_input_tokens_seen": 121697875,
      "step": 5657,
      "time_per_iteration": 2.5965347290039062
    },
    {
      "auxiliary_loss_clip": 0.01103326,
      "auxiliary_loss_mlp": 0.0102704,
      "balance_loss_clip": 1.04097557,
      "balance_loss_mlp": 1.01986408,
      "epoch": 0.6803342752359767,
      "flos": 17712938908800.0,
      "grad_norm": 2.2938597929919027,
      "language_loss": 0.75191844,
      "learning_rate": 9.793838068426472e-07,
      "loss": 0.77322221,
      "num_input_tokens_seen": 121715570,
      "step": 5658,
      "time_per_iteration": 2.5831503868103027
    },
    {
      "auxiliary_loss_clip": 0.01166918,
      "auxiliary_loss_mlp": 0.01026327,
      "balance_loss_clip": 1.05057955,
      "balance_loss_mlp": 1.01892436,
      "epoch": 0.6804545181266157,
      "flos": 11326800902400.0,
      "grad_norm": 2.7173849621934822,
      "language_loss": 0.61236942,
      "learning_rate": 9.78713975156799e-07,
      "loss": 0.63430178,
      "num_input_tokens_seen": 121731435,
      "step": 5659,
      "time_per_iteration": 2.4325311183929443
    },
    {
      "auxiliary_loss_clip": 0.01125207,
      "auxiliary_loss_mlp": 0.01029424,
      "balance_loss_clip": 1.04610777,
      "balance_loss_mlp": 1.02192533,
      "epoch": 0.6805747610172549,
      "flos": 29350976181120.0,
      "grad_norm": 1.8486801483951423,
      "language_loss": 0.71587217,
      "learning_rate": 9.780442983992273e-07,
      "loss": 0.73741853,
      "num_input_tokens_seen": 121749950,
      "step": 5660,
      "time_per_iteration": 2.6124396324157715
    },
    {
      "auxiliary_loss_clip": 0.01130801,
      "auxiliary_loss_mlp": 0.01032966,
      "balance_loss_clip": 1.04505086,
      "balance_loss_mlp": 1.02529788,
      "epoch": 0.680695003907894,
      "flos": 37631868612480.0,
      "grad_norm": 1.6071573782146757,
      "language_loss": 0.71866477,
      "learning_rate": 9.773747766715238e-07,
      "loss": 0.74030244,
      "num_input_tokens_seen": 121770770,
      "step": 5661,
      "time_per_iteration": 2.6625874042510986
    },
    {
      "auxiliary_loss_clip": 0.01139651,
      "auxiliary_loss_mlp": 0.01021775,
      "balance_loss_clip": 1.04577529,
      "balance_loss_mlp": 1.01432443,
      "epoch": 0.680815246798533,
      "flos": 22127365601280.0,
      "grad_norm": 1.7179918081846364,
      "language_loss": 0.79807717,
      "learning_rate": 9.767054100752536e-07,
      "loss": 0.81969142,
      "num_input_tokens_seen": 121790720,
      "step": 5662,
      "time_per_iteration": 2.525566577911377
    },
    {
      "auxiliary_loss_clip": 0.01123828,
      "auxiliary_loss_mlp": 0.01030101,
      "balance_loss_clip": 1.04616284,
      "balance_loss_mlp": 1.02283525,
      "epoch": 0.6809354896891722,
      "flos": 17201822330880.0,
      "grad_norm": 2.0901567309030833,
      "language_loss": 0.81161952,
      "learning_rate": 9.760361987119584e-07,
      "loss": 0.83315885,
      "num_input_tokens_seen": 121808455,
      "step": 5663,
      "time_per_iteration": 2.5406906604766846
    },
    {
      "auxiliary_loss_clip": 0.01136944,
      "auxiliary_loss_mlp": 0.01030694,
      "balance_loss_clip": 1.04802823,
      "balance_loss_mlp": 1.02250397,
      "epoch": 0.6810557325798112,
      "flos": 12458166554880.0,
      "grad_norm": 1.8382649777457272,
      "language_loss": 0.67841476,
      "learning_rate": 9.753671426831592e-07,
      "loss": 0.70009112,
      "num_input_tokens_seen": 121824470,
      "step": 5664,
      "time_per_iteration": 2.479050636291504
    },
    {
      "auxiliary_loss_clip": 0.01145355,
      "auxiliary_loss_mlp": 0.01026546,
      "balance_loss_clip": 1.04465723,
      "balance_loss_mlp": 1.01971579,
      "epoch": 0.6811759754704503,
      "flos": 22156165330560.0,
      "grad_norm": 1.7772225634655656,
      "language_loss": 0.79535538,
      "learning_rate": 9.746982420903483e-07,
      "loss": 0.81707436,
      "num_input_tokens_seen": 121842665,
      "step": 5665,
      "time_per_iteration": 3.391728639602661
    },
    {
      "auxiliary_loss_clip": 0.01150146,
      "auxiliary_loss_mlp": 0.01023564,
      "balance_loss_clip": 1.0494591,
      "balance_loss_mlp": 1.0164119,
      "epoch": 0.6812962183610894,
      "flos": 17525377065600.0,
      "grad_norm": 1.5682757718983629,
      "language_loss": 0.74863291,
      "learning_rate": 9.740294970349993e-07,
      "loss": 0.77037001,
      "num_input_tokens_seen": 121859080,
      "step": 5666,
      "time_per_iteration": 2.4695541858673096
    },
    {
      "auxiliary_loss_clip": 0.01039677,
      "auxiliary_loss_mlp": 0.01001153,
      "balance_loss_clip": 1.01083994,
      "balance_loss_mlp": 1.0000205,
      "epoch": 0.6814164612517285,
      "flos": 60274480855680.0,
      "grad_norm": 0.8836716635585617,
      "language_loss": 0.60894394,
      "learning_rate": 9.733609076185594e-07,
      "loss": 0.62935227,
      "num_input_tokens_seen": 121915485,
      "step": 5667,
      "time_per_iteration": 3.009120225906372
    },
    {
      "auxiliary_loss_clip": 0.01155378,
      "auxiliary_loss_mlp": 0.01027934,
      "balance_loss_clip": 1.04965425,
      "balance_loss_mlp": 1.02032852,
      "epoch": 0.6815367041423676,
      "flos": 19317750750720.0,
      "grad_norm": 1.7988610773594416,
      "language_loss": 0.83947885,
      "learning_rate": 9.72692473942455e-07,
      "loss": 0.86131203,
      "num_input_tokens_seen": 121932710,
      "step": 5668,
      "time_per_iteration": 2.473071813583374
    },
    {
      "auxiliary_loss_clip": 0.01114695,
      "auxiliary_loss_mlp": 0.01025102,
      "balance_loss_clip": 1.04689407,
      "balance_loss_mlp": 1.01764011,
      "epoch": 0.6816569470330067,
      "flos": 22161696024960.0,
      "grad_norm": 1.5675163728800097,
      "language_loss": 0.77683437,
      "learning_rate": 9.720241961080849e-07,
      "loss": 0.79823238,
      "num_input_tokens_seen": 121952025,
      "step": 5669,
      "time_per_iteration": 2.603996515274048
    },
    {
      "auxiliary_loss_clip": 0.01166815,
      "auxiliary_loss_mlp": 0.01028622,
      "balance_loss_clip": 1.04924011,
      "balance_loss_mlp": 1.02178514,
      "epoch": 0.6817771899236458,
      "flos": 41463501137280.0,
      "grad_norm": 1.947661232794189,
      "language_loss": 0.73007536,
      "learning_rate": 9.713560742168259e-07,
      "loss": 0.75202978,
      "num_input_tokens_seen": 121974650,
      "step": 5670,
      "time_per_iteration": 2.6379692554473877
    },
    {
      "auxiliary_loss_clip": 0.01118536,
      "auxiliary_loss_mlp": 0.01028356,
      "balance_loss_clip": 1.04423833,
      "balance_loss_mlp": 1.02146554,
      "epoch": 0.6818974328142848,
      "flos": 21106138026240.0,
      "grad_norm": 2.003643706207869,
      "language_loss": 0.71132737,
      "learning_rate": 9.706881083700333e-07,
      "loss": 0.73279631,
      "num_input_tokens_seen": 121994335,
      "step": 5671,
      "time_per_iteration": 2.570873975753784
    },
    {
      "auxiliary_loss_clip": 0.01095701,
      "auxiliary_loss_mlp": 0.01025766,
      "balance_loss_clip": 1.04605794,
      "balance_loss_mlp": 1.01819587,
      "epoch": 0.682017675704924,
      "flos": 20441897769600.0,
      "grad_norm": 2.1700822220827387,
      "language_loss": 0.82479972,
      "learning_rate": 9.700202986690357e-07,
      "loss": 0.84601438,
      "num_input_tokens_seen": 122012635,
      "step": 5672,
      "time_per_iteration": 2.6114470958709717
    },
    {
      "auxiliary_loss_clip": 0.01151184,
      "auxiliary_loss_mlp": 0.00761161,
      "balance_loss_clip": 1.04810286,
      "balance_loss_mlp": 1.00023961,
      "epoch": 0.682137918595563,
      "flos": 20044438801920.0,
      "grad_norm": 4.131740241173308,
      "language_loss": 0.66546011,
      "learning_rate": 9.693526452151413e-07,
      "loss": 0.68458354,
      "num_input_tokens_seen": 122031685,
      "step": 5673,
      "time_per_iteration": 2.5052146911621094
    },
    {
      "auxiliary_loss_clip": 0.01129369,
      "auxiliary_loss_mlp": 0.01024514,
      "balance_loss_clip": 1.04344487,
      "balance_loss_mlp": 1.01658916,
      "epoch": 0.6822581614862021,
      "flos": 31684559063040.0,
      "grad_norm": 1.6501846886148293,
      "language_loss": 0.75593495,
      "learning_rate": 9.686851481096305e-07,
      "loss": 0.77747369,
      "num_input_tokens_seen": 122052995,
      "step": 5674,
      "time_per_iteration": 3.4217255115509033
    },
    {
      "auxiliary_loss_clip": 0.01090648,
      "auxiliary_loss_mlp": 0.01026795,
      "balance_loss_clip": 1.04193497,
      "balance_loss_mlp": 1.02007174,
      "epoch": 0.6823784043768413,
      "flos": 23477570864640.0,
      "grad_norm": 2.038719204676254,
      "language_loss": 0.72020006,
      "learning_rate": 9.68017807453762e-07,
      "loss": 0.74137449,
      "num_input_tokens_seen": 122071740,
      "step": 5675,
      "time_per_iteration": 2.6241455078125
    },
    {
      "auxiliary_loss_clip": 0.01142604,
      "auxiliary_loss_mlp": 0.00760393,
      "balance_loss_clip": 1.04835248,
      "balance_loss_mlp": 1.00020611,
      "epoch": 0.6824986472674803,
      "flos": 14137134024960.0,
      "grad_norm": 1.8425440908073412,
      "language_loss": 0.73037696,
      "learning_rate": 9.673506233487721e-07,
      "loss": 0.74940693,
      "num_input_tokens_seen": 122089705,
      "step": 5676,
      "time_per_iteration": 2.5134241580963135
    },
    {
      "auxiliary_loss_clip": 0.011378,
      "auxiliary_loss_mlp": 0.00759719,
      "balance_loss_clip": 1.04486632,
      "balance_loss_mlp": 1.00021887,
      "epoch": 0.6826188901581194,
      "flos": 21504997624320.0,
      "grad_norm": 1.6220014649058054,
      "language_loss": 0.85952222,
      "learning_rate": 9.666835958958717e-07,
      "loss": 0.87849742,
      "num_input_tokens_seen": 122109025,
      "step": 5677,
      "time_per_iteration": 3.4203591346740723
    },
    {
      "auxiliary_loss_clip": 0.01165394,
      "auxiliary_loss_mlp": 0.01023423,
      "balance_loss_clip": 1.04916883,
      "balance_loss_mlp": 1.01656294,
      "epoch": 0.6827391330487584,
      "flos": 20810126044800.0,
      "grad_norm": 2.9023815656510132,
      "language_loss": 0.80719233,
      "learning_rate": 9.660167251962484e-07,
      "loss": 0.82908052,
      "num_input_tokens_seen": 122127385,
      "step": 5678,
      "time_per_iteration": 3.1669881343841553
    },
    {
      "auxiliary_loss_clip": 0.01126764,
      "auxiliary_loss_mlp": 0.01025955,
      "balance_loss_clip": 1.04494977,
      "balance_loss_mlp": 1.01855469,
      "epoch": 0.6828593759393976,
      "flos": 21688788539520.0,
      "grad_norm": 1.5755214379908924,
      "language_loss": 0.77823997,
      "learning_rate": 9.653500113510654e-07,
      "loss": 0.79976714,
      "num_input_tokens_seen": 122146500,
      "step": 5679,
      "time_per_iteration": 2.5586845874786377
    },
    {
      "auxiliary_loss_clip": 0.01133224,
      "auxiliary_loss_mlp": 0.01028674,
      "balance_loss_clip": 1.04401433,
      "balance_loss_mlp": 1.02114546,
      "epoch": 0.6829796188300367,
      "flos": 25337707557120.0,
      "grad_norm": 2.2016656460012305,
      "language_loss": 0.67460656,
      "learning_rate": 9.646834544614627e-07,
      "loss": 0.69622552,
      "num_input_tokens_seen": 122167000,
      "step": 5680,
      "time_per_iteration": 2.55533766746521
    },
    {
      "auxiliary_loss_clip": 0.0113169,
      "auxiliary_loss_mlp": 0.01025222,
      "balance_loss_clip": 1.04764128,
      "balance_loss_mlp": 1.01826,
      "epoch": 0.6830998617206757,
      "flos": 20704800389760.0,
      "grad_norm": 1.8578506392204794,
      "language_loss": 0.76461303,
      "learning_rate": 9.64017054628558e-07,
      "loss": 0.78618217,
      "num_input_tokens_seen": 122185825,
      "step": 5681,
      "time_per_iteration": 2.512789011001587
    },
    {
      "auxiliary_loss_clip": 0.01114414,
      "auxiliary_loss_mlp": 0.01025703,
      "balance_loss_clip": 1.04334009,
      "balance_loss_mlp": 1.01868176,
      "epoch": 0.6832201046113149,
      "flos": 21726638496000.0,
      "grad_norm": 1.5941441557914793,
      "language_loss": 0.78459799,
      "learning_rate": 9.63350811953441e-07,
      "loss": 0.8059991,
      "num_input_tokens_seen": 122206200,
      "step": 5682,
      "time_per_iteration": 2.6013052463531494
    },
    {
      "auxiliary_loss_clip": 0.01126072,
      "auxiliary_loss_mlp": 0.01024027,
      "balance_loss_clip": 1.04457724,
      "balance_loss_mlp": 1.01708031,
      "epoch": 0.6833403475019539,
      "flos": 19536554448000.0,
      "grad_norm": 2.0652471508641073,
      "language_loss": 0.7053473,
      "learning_rate": 9.626847265371826e-07,
      "loss": 0.72684824,
      "num_input_tokens_seen": 122225520,
      "step": 5683,
      "time_per_iteration": 2.5576019287109375
    },
    {
      "auxiliary_loss_clip": 0.01128974,
      "auxiliary_loss_mlp": 0.01030835,
      "balance_loss_clip": 1.04240894,
      "balance_loss_mlp": 1.02292871,
      "epoch": 0.683460590392593,
      "flos": 19352153001600.0,
      "grad_norm": 1.8999893229702642,
      "language_loss": 0.784688,
      "learning_rate": 9.620187984808262e-07,
      "loss": 0.8062861,
      "num_input_tokens_seen": 122244320,
      "step": 5684,
      "time_per_iteration": 2.5262887477874756
    },
    {
      "auxiliary_loss_clip": 0.01136191,
      "auxiliary_loss_mlp": 0.00760703,
      "balance_loss_clip": 1.04593861,
      "balance_loss_mlp": 1.00020552,
      "epoch": 0.6835808332832322,
      "flos": 23288500650240.0,
      "grad_norm": 1.9296193042927574,
      "language_loss": 0.85909057,
      "learning_rate": 9.613530278853919e-07,
      "loss": 0.87805951,
      "num_input_tokens_seen": 122264295,
      "step": 5685,
      "time_per_iteration": 2.54351544380188
    },
    {
      "auxiliary_loss_clip": 0.01148906,
      "auxiliary_loss_mlp": 0.01029945,
      "balance_loss_clip": 1.04742265,
      "balance_loss_mlp": 1.02292311,
      "epoch": 0.6837010761738712,
      "flos": 21653416621440.0,
      "grad_norm": 1.8982122784811597,
      "language_loss": 0.74644858,
      "learning_rate": 9.60687414851879e-07,
      "loss": 0.76823711,
      "num_input_tokens_seen": 122285300,
      "step": 5686,
      "time_per_iteration": 2.517876625061035
    },
    {
      "auxiliary_loss_clip": 0.01136863,
      "auxiliary_loss_mlp": 0.0102696,
      "balance_loss_clip": 1.04710639,
      "balance_loss_mlp": 1.01929474,
      "epoch": 0.6838213190645103,
      "flos": 17566387418880.0,
      "grad_norm": 2.136565312108994,
      "language_loss": 0.76912403,
      "learning_rate": 9.600219594812575e-07,
      "loss": 0.79076219,
      "num_input_tokens_seen": 122303240,
      "step": 5687,
      "time_per_iteration": 2.4824867248535156
    },
    {
      "auxiliary_loss_clip": 0.01163313,
      "auxiliary_loss_mlp": 0.01024775,
      "balance_loss_clip": 1.04771364,
      "balance_loss_mlp": 1.01788485,
      "epoch": 0.6839415619551494,
      "flos": 23112538899840.0,
      "grad_norm": 1.6122664594782505,
      "language_loss": 0.72600925,
      "learning_rate": 9.593566618744786e-07,
      "loss": 0.74789017,
      "num_input_tokens_seen": 122323390,
      "step": 5688,
      "time_per_iteration": 2.471177339553833
    },
    {
      "auxiliary_loss_clip": 0.01164777,
      "auxiliary_loss_mlp": 0.01024016,
      "balance_loss_clip": 1.04760373,
      "balance_loss_mlp": 1.01714611,
      "epoch": 0.6840618048457885,
      "flos": 22127868391680.0,
      "grad_norm": 1.5908140554140784,
      "language_loss": 0.74031466,
      "learning_rate": 9.58691522132466e-07,
      "loss": 0.76220262,
      "num_input_tokens_seen": 122342200,
      "step": 5689,
      "time_per_iteration": 2.4723119735717773
    },
    {
      "auxiliary_loss_clip": 0.01144566,
      "auxiliary_loss_mlp": 0.01028136,
      "balance_loss_clip": 1.04910791,
      "balance_loss_mlp": 1.02106667,
      "epoch": 0.6841820477364275,
      "flos": 22015898720640.0,
      "grad_norm": 1.96896194936858,
      "language_loss": 0.84506124,
      "learning_rate": 9.58026540356123e-07,
      "loss": 0.86678821,
      "num_input_tokens_seen": 122360465,
      "step": 5690,
      "time_per_iteration": 2.5415971279144287
    },
    {
      "auxiliary_loss_clip": 0.01151498,
      "auxiliary_loss_mlp": 0.01025976,
      "balance_loss_clip": 1.04520035,
      "balance_loss_mlp": 1.01837027,
      "epoch": 0.6843022906270667,
      "flos": 24900531125760.0,
      "grad_norm": 1.6137864958644448,
      "language_loss": 0.86558896,
      "learning_rate": 9.573617166463246e-07,
      "loss": 0.88736373,
      "num_input_tokens_seen": 122381680,
      "step": 5691,
      "time_per_iteration": 3.3195462226867676
    },
    {
      "auxiliary_loss_clip": 0.01138605,
      "auxiliary_loss_mlp": 0.01024419,
      "balance_loss_clip": 1.04456234,
      "balance_loss_mlp": 1.01693821,
      "epoch": 0.6844225335177058,
      "flos": 19969924037760.0,
      "grad_norm": 1.9053040072080132,
      "language_loss": 0.60345274,
      "learning_rate": 9.56697051103924e-07,
      "loss": 0.62508297,
      "num_input_tokens_seen": 122399120,
      "step": 5692,
      "time_per_iteration": 2.5237269401550293
    },
    {
      "auxiliary_loss_clip": 0.01135848,
      "auxiliary_loss_mlp": 0.01025861,
      "balance_loss_clip": 1.04533195,
      "balance_loss_mlp": 1.01875651,
      "epoch": 0.6845427764083448,
      "flos": 25883334126720.0,
      "grad_norm": 1.7779145595680794,
      "language_loss": 0.80785573,
      "learning_rate": 9.560325438297522e-07,
      "loss": 0.82947284,
      "num_input_tokens_seen": 122417430,
      "step": 5693,
      "time_per_iteration": 2.5646586418151855
    },
    {
      "auxiliary_loss_clip": 0.01138211,
      "auxiliary_loss_mlp": 0.01028227,
      "balance_loss_clip": 1.04869306,
      "balance_loss_mlp": 1.02169704,
      "epoch": 0.684663019298984,
      "flos": 18880143356160.0,
      "grad_norm": 1.9005157970986026,
      "language_loss": 0.86649841,
      "learning_rate": 9.553681949246127e-07,
      "loss": 0.88816285,
      "num_input_tokens_seen": 122435055,
      "step": 5694,
      "time_per_iteration": 2.502432346343994
    },
    {
      "auxiliary_loss_clip": 0.01133004,
      "auxiliary_loss_mlp": 0.01030335,
      "balance_loss_clip": 1.0486325,
      "balance_loss_mlp": 1.02204394,
      "epoch": 0.684783262189623,
      "flos": 54193725302400.0,
      "grad_norm": 2.150539961280656,
      "language_loss": 0.75007784,
      "learning_rate": 9.547040044892886e-07,
      "loss": 0.77171123,
      "num_input_tokens_seen": 122462570,
      "step": 5695,
      "time_per_iteration": 2.8399417400360107
    },
    {
      "auxiliary_loss_clip": 0.01050097,
      "auxiliary_loss_mlp": 0.01001401,
      "balance_loss_clip": 1.01063955,
      "balance_loss_mlp": 1.00034034,
      "epoch": 0.6849035050802621,
      "flos": 63970264143360.0,
      "grad_norm": 0.864901599711069,
      "language_loss": 0.60177034,
      "learning_rate": 9.540399726245354e-07,
      "loss": 0.62228531,
      "num_input_tokens_seen": 122519275,
      "step": 5696,
      "time_per_iteration": 2.968353509902954
    },
    {
      "auxiliary_loss_clip": 0.0113527,
      "auxiliary_loss_mlp": 0.01029819,
      "balance_loss_clip": 1.04552543,
      "balance_loss_mlp": 1.0219872,
      "epoch": 0.6850237479709013,
      "flos": 25224121774080.0,
      "grad_norm": 1.9013705695185932,
      "language_loss": 0.68832725,
      "learning_rate": 9.533760994310859e-07,
      "loss": 0.70997816,
      "num_input_tokens_seen": 122539675,
      "step": 5697,
      "time_per_iteration": 2.5716166496276855
    },
    {
      "auxiliary_loss_clip": 0.01168351,
      "auxiliary_loss_mlp": 0.01025756,
      "balance_loss_clip": 1.05062366,
      "balance_loss_mlp": 1.01847577,
      "epoch": 0.6851439908615403,
      "flos": 19354128249600.0,
      "grad_norm": 1.98070628751398,
      "language_loss": 0.7477383,
      "learning_rate": 9.527123850096508e-07,
      "loss": 0.76967937,
      "num_input_tokens_seen": 122558035,
      "step": 5698,
      "time_per_iteration": 2.4416208267211914
    },
    {
      "auxiliary_loss_clip": 0.01153577,
      "auxiliary_loss_mlp": 0.01028327,
      "balance_loss_clip": 1.04820704,
      "balance_loss_mlp": 1.02096891,
      "epoch": 0.6852642337521794,
      "flos": 23182133500800.0,
      "grad_norm": 1.7856231498541275,
      "language_loss": 0.71774113,
      "learning_rate": 9.520488294609142e-07,
      "loss": 0.73956019,
      "num_input_tokens_seen": 122576815,
      "step": 5699,
      "time_per_iteration": 2.5128629207611084
    },
    {
      "auxiliary_loss_clip": 0.01016131,
      "auxiliary_loss_mlp": 0.01002788,
      "balance_loss_clip": 1.01026869,
      "balance_loss_mlp": 1.0015955,
      "epoch": 0.6853844766428185,
      "flos": 62647206583680.0,
      "grad_norm": 0.7401045916684383,
      "language_loss": 0.53870606,
      "learning_rate": 9.513854328855368e-07,
      "loss": 0.55889529,
      "num_input_tokens_seen": 122634690,
      "step": 5700,
      "time_per_iteration": 3.1216955184936523
    },
    {
      "auxiliary_loss_clip": 0.0116241,
      "auxiliary_loss_mlp": 0.01021951,
      "balance_loss_clip": 1.04722857,
      "balance_loss_mlp": 1.01460147,
      "epoch": 0.6855047195334576,
      "flos": 23437242869760.0,
      "grad_norm": 1.8902997397583672,
      "language_loss": 0.8096053,
      "learning_rate": 9.507221953841558e-07,
      "loss": 0.83144891,
      "num_input_tokens_seen": 122652320,
      "step": 5701,
      "time_per_iteration": 3.239274263381958
    },
    {
      "auxiliary_loss_clip": 0.01155713,
      "auxiliary_loss_mlp": 0.01028233,
      "balance_loss_clip": 1.05078673,
      "balance_loss_mlp": 1.02047896,
      "epoch": 0.6856249624240967,
      "flos": 20664831530880.0,
      "grad_norm": 1.6413448193191258,
      "language_loss": 0.78001034,
      "learning_rate": 9.500591170573824e-07,
      "loss": 0.80184972,
      "num_input_tokens_seen": 122672340,
      "step": 5702,
      "time_per_iteration": 2.5046093463897705
    },
    {
      "auxiliary_loss_clip": 0.01110552,
      "auxiliary_loss_mlp": 0.01028393,
      "balance_loss_clip": 1.04462862,
      "balance_loss_mlp": 1.02124023,
      "epoch": 0.6857452053147358,
      "flos": 17087302794240.0,
      "grad_norm": 1.8557634551056017,
      "language_loss": 0.74102861,
      "learning_rate": 9.493961980058078e-07,
      "loss": 0.76241809,
      "num_input_tokens_seen": 122689935,
      "step": 5703,
      "time_per_iteration": 2.5681872367858887
    },
    {
      "auxiliary_loss_clip": 0.01081553,
      "auxiliary_loss_mlp": 0.01024726,
      "balance_loss_clip": 1.03793788,
      "balance_loss_mlp": 1.01807141,
      "epoch": 0.6858654482053749,
      "flos": 30847266057600.0,
      "grad_norm": 2.1610589259119592,
      "language_loss": 0.6733427,
      "learning_rate": 9.48733438329993e-07,
      "loss": 0.69440544,
      "num_input_tokens_seen": 122710200,
      "step": 5704,
      "time_per_iteration": 4.335917234420776
    },
    {
      "auxiliary_loss_clip": 0.01165531,
      "auxiliary_loss_mlp": 0.00760694,
      "balance_loss_clip": 1.05103517,
      "balance_loss_mlp": 1.00019825,
      "epoch": 0.6859856910960139,
      "flos": 28877314510080.0,
      "grad_norm": 1.6392451437394708,
      "language_loss": 0.74553013,
      "learning_rate": 9.480708381304807e-07,
      "loss": 0.76479232,
      "num_input_tokens_seen": 122731495,
      "step": 5705,
      "time_per_iteration": 2.540844678878784
    },
    {
      "auxiliary_loss_clip": 0.01108765,
      "auxiliary_loss_mlp": 0.01028648,
      "balance_loss_clip": 1.04675865,
      "balance_loss_mlp": 1.02131701,
      "epoch": 0.6861059339866531,
      "flos": 19354523299200.0,
      "grad_norm": 3.1038666377538555,
      "language_loss": 0.83632052,
      "learning_rate": 9.474083975077858e-07,
      "loss": 0.85769463,
      "num_input_tokens_seen": 122748620,
      "step": 5706,
      "time_per_iteration": 2.568183183670044
    },
    {
      "auxiliary_loss_clip": 0.01143979,
      "auxiliary_loss_mlp": 0.01021611,
      "balance_loss_clip": 1.04503477,
      "balance_loss_mlp": 1.01429141,
      "epoch": 0.6862261768772921,
      "flos": 22199976944640.0,
      "grad_norm": 6.695634033673204,
      "language_loss": 0.79853654,
      "learning_rate": 9.467461165623994e-07,
      "loss": 0.82019246,
      "num_input_tokens_seen": 122767670,
      "step": 5707,
      "time_per_iteration": 2.501483917236328
    },
    {
      "auxiliary_loss_clip": 0.01152054,
      "auxiliary_loss_mlp": 0.01026147,
      "balance_loss_clip": 1.04480195,
      "balance_loss_mlp": 1.01923549,
      "epoch": 0.6863464197679312,
      "flos": 26285677344000.0,
      "grad_norm": 1.8173769854706732,
      "language_loss": 0.79838681,
      "learning_rate": 9.46083995394791e-07,
      "loss": 0.82016879,
      "num_input_tokens_seen": 122785480,
      "step": 5708,
      "time_per_iteration": 2.519155263900757
    },
    {
      "auxiliary_loss_clip": 0.01152219,
      "auxiliary_loss_mlp": 0.00760361,
      "balance_loss_clip": 1.04709446,
      "balance_loss_mlp": 1.00021708,
      "epoch": 0.6864666626585703,
      "flos": 37815228564480.0,
      "grad_norm": 2.6910946360441725,
      "language_loss": 0.63014579,
      "learning_rate": 9.454220341054012e-07,
      "loss": 0.64927161,
      "num_input_tokens_seen": 122810265,
      "step": 5709,
      "time_per_iteration": 2.6425440311431885
    },
    {
      "auxiliary_loss_clip": 0.01125337,
      "auxiliary_loss_mlp": 0.01025609,
      "balance_loss_clip": 1.04563737,
      "balance_loss_mlp": 1.01806879,
      "epoch": 0.6865869055492094,
      "flos": 19391152193280.0,
      "grad_norm": 1.8980088562320436,
      "language_loss": 0.80772674,
      "learning_rate": 9.447602327946512e-07,
      "loss": 0.82923621,
      "num_input_tokens_seen": 122828905,
      "step": 5710,
      "time_per_iteration": 2.5407516956329346
    },
    {
      "auxiliary_loss_clip": 0.01135689,
      "auxiliary_loss_mlp": 0.0102988,
      "balance_loss_clip": 1.04427576,
      "balance_loss_mlp": 1.0226531,
      "epoch": 0.6867071484398485,
      "flos": 20375966355840.0,
      "grad_norm": 1.8879728530309865,
      "language_loss": 0.7636835,
      "learning_rate": 9.440985915629338e-07,
      "loss": 0.78533918,
      "num_input_tokens_seen": 122846235,
      "step": 5711,
      "time_per_iteration": 2.5184476375579834
    },
    {
      "auxiliary_loss_clip": 0.0116805,
      "auxiliary_loss_mlp": 0.01027186,
      "balance_loss_clip": 1.05247533,
      "balance_loss_mlp": 1.01987004,
      "epoch": 0.6868273913304875,
      "flos": 15889143801600.0,
      "grad_norm": 1.8122391798254844,
      "language_loss": 0.73123628,
      "learning_rate": 9.434371105106223e-07,
      "loss": 0.75318861,
      "num_input_tokens_seen": 122863835,
      "step": 5712,
      "time_per_iteration": 2.441215991973877
    },
    {
      "auxiliary_loss_clip": 0.0112163,
      "auxiliary_loss_mlp": 0.01024091,
      "balance_loss_clip": 1.04381871,
      "balance_loss_mlp": 1.0173502,
      "epoch": 0.6869476342211267,
      "flos": 24462492768000.0,
      "grad_norm": 1.81442563635407,
      "language_loss": 0.70659763,
      "learning_rate": 9.427757897380602e-07,
      "loss": 0.72805488,
      "num_input_tokens_seen": 122883235,
      "step": 5713,
      "time_per_iteration": 2.591341733932495
    },
    {
      "auxiliary_loss_clip": 0.01120486,
      "auxiliary_loss_mlp": 0.01024596,
      "balance_loss_clip": 1.04435217,
      "balance_loss_mlp": 1.01700854,
      "epoch": 0.6870678771117658,
      "flos": 18442571875200.0,
      "grad_norm": 2.326256140573583,
      "language_loss": 0.84764057,
      "learning_rate": 9.421146293455695e-07,
      "loss": 0.86909139,
      "num_input_tokens_seen": 122898975,
      "step": 5714,
      "time_per_iteration": 2.5438804626464844
    },
    {
      "auxiliary_loss_clip": 0.01136844,
      "auxiliary_loss_mlp": 0.01033631,
      "balance_loss_clip": 1.04657006,
      "balance_loss_mlp": 1.02611518,
      "epoch": 0.6871881200024048,
      "flos": 22200371994240.0,
      "grad_norm": 2.28277672256521,
      "language_loss": 0.68300164,
      "learning_rate": 9.414536294334489e-07,
      "loss": 0.70470643,
      "num_input_tokens_seen": 122918995,
      "step": 5715,
      "time_per_iteration": 2.5342860221862793
    },
    {
      "auxiliary_loss_clip": 0.01139129,
      "auxiliary_loss_mlp": 0.01025616,
      "balance_loss_clip": 1.04364014,
      "balance_loss_mlp": 1.01885438,
      "epoch": 0.687308362893044,
      "flos": 22127724737280.0,
      "grad_norm": 1.7969527618959578,
      "language_loss": 0.69725823,
      "learning_rate": 9.407927901019708e-07,
      "loss": 0.71890569,
      "num_input_tokens_seen": 122938125,
      "step": 5716,
      "time_per_iteration": 2.5146777629852295
    },
    {
      "auxiliary_loss_clip": 0.01152605,
      "auxiliary_loss_mlp": 0.01024543,
      "balance_loss_clip": 1.04751706,
      "balance_loss_mlp": 1.01755738,
      "epoch": 0.687428605783683,
      "flos": 25040546340480.0,
      "grad_norm": 1.973169604322113,
      "language_loss": 0.76846039,
      "learning_rate": 9.401321114513854e-07,
      "loss": 0.79023182,
      "num_input_tokens_seen": 122957020,
      "step": 5717,
      "time_per_iteration": 3.2731316089630127
    },
    {
      "auxiliary_loss_clip": 0.01166904,
      "auxiliary_loss_mlp": 0.01027616,
      "balance_loss_clip": 1.05015016,
      "balance_loss_mlp": 1.02019501,
      "epoch": 0.6875488486743221,
      "flos": 23770063313280.0,
      "grad_norm": 1.730916092835809,
      "language_loss": 0.75332522,
      "learning_rate": 9.394715935819155e-07,
      "loss": 0.7752704,
      "num_input_tokens_seen": 122977410,
      "step": 5718,
      "time_per_iteration": 2.480463743209839
    },
    {
      "auxiliary_loss_clip": 0.01156586,
      "auxiliary_loss_mlp": 0.01024006,
      "balance_loss_clip": 1.04826593,
      "balance_loss_mlp": 1.01699066,
      "epoch": 0.6876690915649613,
      "flos": 25516937445120.0,
      "grad_norm": 1.8256219986346391,
      "language_loss": 0.62707502,
      "learning_rate": 9.388112365937608e-07,
      "loss": 0.64888096,
      "num_input_tokens_seen": 122996875,
      "step": 5719,
      "time_per_iteration": 2.530275821685791
    },
    {
      "auxiliary_loss_clip": 0.01124894,
      "auxiliary_loss_mlp": 0.01042784,
      "balance_loss_clip": 1.04478407,
      "balance_loss_mlp": 1.03511298,
      "epoch": 0.6877893344556003,
      "flos": 19427996568960.0,
      "grad_norm": 2.229840447013276,
      "language_loss": 0.82707143,
      "learning_rate": 9.381510405870985e-07,
      "loss": 0.84874821,
      "num_input_tokens_seen": 123015890,
      "step": 5720,
      "time_per_iteration": 2.535473346710205
    },
    {
      "auxiliary_loss_clip": 0.01153736,
      "auxiliary_loss_mlp": 0.01026952,
      "balance_loss_clip": 1.04994202,
      "balance_loss_mlp": 1.0191381,
      "epoch": 0.6879095773462394,
      "flos": 18661303745280.0,
      "grad_norm": 3.4194654884094096,
      "language_loss": 0.7736553,
      "learning_rate": 9.374910056620791e-07,
      "loss": 0.79546225,
      "num_input_tokens_seen": 123034955,
      "step": 5721,
      "time_per_iteration": 2.477207660675049
    },
    {
      "auxiliary_loss_clip": 0.01155783,
      "auxiliary_loss_mlp": 0.0103076,
      "balance_loss_clip": 1.05028844,
      "balance_loss_mlp": 1.02347028,
      "epoch": 0.6880298202368785,
      "flos": 20883132437760.0,
      "grad_norm": 2.344875550638085,
      "language_loss": 0.81039751,
      "learning_rate": 9.368311319188293e-07,
      "loss": 0.83226287,
      "num_input_tokens_seen": 123052770,
      "step": 5722,
      "time_per_iteration": 2.486459493637085
    },
    {
      "auxiliary_loss_clip": 0.01123253,
      "auxiliary_loss_mlp": 0.01027025,
      "balance_loss_clip": 1.04438496,
      "balance_loss_mlp": 1.02020025,
      "epoch": 0.6881500631275176,
      "flos": 30153292318080.0,
      "grad_norm": 1.693252059409962,
      "language_loss": 0.7936244,
      "learning_rate": 9.361714194574515e-07,
      "loss": 0.81512719,
      "num_input_tokens_seen": 123075105,
      "step": 5723,
      "time_per_iteration": 2.624188184738159
    },
    {
      "auxiliary_loss_clip": 0.01057697,
      "auxiliary_loss_mlp": 0.01000958,
      "balance_loss_clip": 1.00993156,
      "balance_loss_mlp": 0.9998911,
      "epoch": 0.6882703060181566,
      "flos": 66181537215360.0,
      "grad_norm": 0.7629704581325396,
      "language_loss": 0.58348477,
      "learning_rate": 9.355118683780228e-07,
      "loss": 0.60407138,
      "num_input_tokens_seen": 123145175,
      "step": 5724,
      "time_per_iteration": 3.16556978225708
    },
    {
      "auxiliary_loss_clip": 0.0116783,
      "auxiliary_loss_mlp": 0.01031387,
      "balance_loss_clip": 1.04947019,
      "balance_loss_mlp": 1.02406788,
      "epoch": 0.6883905489087958,
      "flos": 18214646123520.0,
      "grad_norm": 2.8283392893316663,
      "language_loss": 0.79192966,
      "learning_rate": 9.348524787805987e-07,
      "loss": 0.81392181,
      "num_input_tokens_seen": 123160365,
      "step": 5725,
      "time_per_iteration": 2.425603151321411
    },
    {
      "auxiliary_loss_clip": 0.01125693,
      "auxiliary_loss_mlp": 0.01027318,
      "balance_loss_clip": 1.04237247,
      "balance_loss_mlp": 1.01985002,
      "epoch": 0.6885107917994349,
      "flos": 14056262553600.0,
      "grad_norm": 2.7898642324005967,
      "language_loss": 0.84876698,
      "learning_rate": 9.341932507652053e-07,
      "loss": 0.87029707,
      "num_input_tokens_seen": 123174855,
      "step": 5726,
      "time_per_iteration": 2.5247151851654053
    },
    {
      "auxiliary_loss_clip": 0.01134689,
      "auxiliary_loss_mlp": 0.01026672,
      "balance_loss_clip": 1.04238462,
      "balance_loss_mlp": 1.01879251,
      "epoch": 0.6886310346900739,
      "flos": 28690722334080.0,
      "grad_norm": 1.6825044361732022,
      "language_loss": 0.78562248,
      "learning_rate": 9.335341844318489e-07,
      "loss": 0.80723602,
      "num_input_tokens_seen": 123194995,
      "step": 5727,
      "time_per_iteration": 3.38222599029541
    },
    {
      "auxiliary_loss_clip": 0.01136486,
      "auxiliary_loss_mlp": 0.01027779,
      "balance_loss_clip": 1.04652894,
      "balance_loss_mlp": 1.02047145,
      "epoch": 0.6887512775807131,
      "flos": 24535319592960.0,
      "grad_norm": 1.7214054067080398,
      "language_loss": 0.73218048,
      "learning_rate": 9.328752798805091e-07,
      "loss": 0.7538231,
      "num_input_tokens_seen": 123213465,
      "step": 5728,
      "time_per_iteration": 2.5493662357330322
    },
    {
      "auxiliary_loss_clip": 0.01153602,
      "auxiliary_loss_mlp": 0.01029835,
      "balance_loss_clip": 1.0493803,
      "balance_loss_mlp": 1.02239609,
      "epoch": 0.6888715204713521,
      "flos": 22414363269120.0,
      "grad_norm": 2.1542192628929158,
      "language_loss": 0.75673175,
      "learning_rate": 9.322165372111399e-07,
      "loss": 0.77856612,
      "num_input_tokens_seen": 123231610,
      "step": 5729,
      "time_per_iteration": 3.3181333541870117
    },
    {
      "auxiliary_loss_clip": 0.01121442,
      "auxiliary_loss_mlp": 0.01028151,
      "balance_loss_clip": 1.04601169,
      "balance_loss_mlp": 1.02110255,
      "epoch": 0.6889917633619912,
      "flos": 22054323294720.0,
      "grad_norm": 1.9759455018304592,
      "language_loss": 0.75579858,
      "learning_rate": 9.315579565236747e-07,
      "loss": 0.77729446,
      "num_input_tokens_seen": 123250715,
      "step": 5730,
      "time_per_iteration": 3.3290390968322754
    },
    {
      "auxiliary_loss_clip": 0.01136338,
      "auxiliary_loss_mlp": 0.01030563,
      "balance_loss_clip": 1.04792428,
      "balance_loss_mlp": 1.02363372,
      "epoch": 0.6891120062526304,
      "flos": 23949724164480.0,
      "grad_norm": 2.1152947795277655,
      "language_loss": 0.74255574,
      "learning_rate": 9.308995379180162e-07,
      "loss": 0.76422471,
      "num_input_tokens_seen": 123270270,
      "step": 5731,
      "time_per_iteration": 2.5412368774414062
    },
    {
      "auxiliary_loss_clip": 0.01048876,
      "auxiliary_loss_mlp": 0.01000581,
      "balance_loss_clip": 1.0096215,
      "balance_loss_mlp": 0.99944896,
      "epoch": 0.6892322491432694,
      "flos": 64117354337280.0,
      "grad_norm": 0.7401446330248045,
      "language_loss": 0.5953989,
      "learning_rate": 9.302412814940488e-07,
      "loss": 0.61589348,
      "num_input_tokens_seen": 123333045,
      "step": 5732,
      "time_per_iteration": 3.1396541595458984
    },
    {
      "auxiliary_loss_clip": 0.01137474,
      "auxiliary_loss_mlp": 0.01024939,
      "balance_loss_clip": 1.0446943,
      "balance_loss_mlp": 1.01736915,
      "epoch": 0.6893524920339085,
      "flos": 23002436736000.0,
      "grad_norm": 2.1347819835232844,
      "language_loss": 0.71340138,
      "learning_rate": 9.295831873516276e-07,
      "loss": 0.73502553,
      "num_input_tokens_seen": 123352320,
      "step": 5733,
      "time_per_iteration": 2.541996955871582
    },
    {
      "auxiliary_loss_clip": 0.01166325,
      "auxiliary_loss_mlp": 0.01024185,
      "balance_loss_clip": 1.05103803,
      "balance_loss_mlp": 1.01703799,
      "epoch": 0.6894727349245476,
      "flos": 21396260177280.0,
      "grad_norm": 2.0102870533987036,
      "language_loss": 0.76134324,
      "learning_rate": 9.289252555905873e-07,
      "loss": 0.78324831,
      "num_input_tokens_seen": 123372400,
      "step": 5734,
      "time_per_iteration": 2.4620800018310547
    },
    {
      "auxiliary_loss_clip": 0.01155252,
      "auxiliary_loss_mlp": 0.01031587,
      "balance_loss_clip": 1.05111241,
      "balance_loss_mlp": 1.02428555,
      "epoch": 0.6895929778151867,
      "flos": 19865316654720.0,
      "grad_norm": 1.8569828080358672,
      "language_loss": 0.75686371,
      "learning_rate": 9.282674863107334e-07,
      "loss": 0.77873206,
      "num_input_tokens_seen": 123390215,
      "step": 5735,
      "time_per_iteration": 2.49055814743042
    },
    {
      "auxiliary_loss_clip": 0.01148919,
      "auxiliary_loss_mlp": 0.01026601,
      "balance_loss_clip": 1.04844213,
      "balance_loss_mlp": 1.01951408,
      "epoch": 0.6897132207058257,
      "flos": 18179166464640.0,
      "grad_norm": 2.477312700318787,
      "language_loss": 0.7578975,
      "learning_rate": 9.276098796118488e-07,
      "loss": 0.77965266,
      "num_input_tokens_seen": 123406700,
      "step": 5736,
      "time_per_iteration": 2.45066499710083
    },
    {
      "auxiliary_loss_clip": 0.01137122,
      "auxiliary_loss_mlp": 0.01022163,
      "balance_loss_clip": 1.04663515,
      "balance_loss_mlp": 1.01506722,
      "epoch": 0.6898334635964649,
      "flos": 32561641359360.0,
      "grad_norm": 1.732943225997359,
      "language_loss": 0.66242653,
      "learning_rate": 9.269524355936938e-07,
      "loss": 0.68401945,
      "num_input_tokens_seen": 123429880,
      "step": 5737,
      "time_per_iteration": 2.6173627376556396
    },
    {
      "auxiliary_loss_clip": 0.01131646,
      "auxiliary_loss_mlp": 0.01024587,
      "balance_loss_clip": 1.04287481,
      "balance_loss_mlp": 1.01760995,
      "epoch": 0.689953706487104,
      "flos": 22819004956800.0,
      "grad_norm": 1.7024815078908677,
      "language_loss": 0.85150766,
      "learning_rate": 9.262951543560002e-07,
      "loss": 0.87307,
      "num_input_tokens_seen": 123449105,
      "step": 5738,
      "time_per_iteration": 2.528764247894287
    },
    {
      "auxiliary_loss_clip": 0.01139218,
      "auxiliary_loss_mlp": 0.010349,
      "balance_loss_clip": 1.04997778,
      "balance_loss_mlp": 1.02735412,
      "epoch": 0.690073949377743,
      "flos": 18515362786560.0,
      "grad_norm": 2.2228202177294683,
      "language_loss": 0.85902929,
      "learning_rate": 9.256380359984795e-07,
      "loss": 0.88077044,
      "num_input_tokens_seen": 123466215,
      "step": 5739,
      "time_per_iteration": 2.4776101112365723
    },
    {
      "auxiliary_loss_clip": 0.01117374,
      "auxiliary_loss_mlp": 0.01024675,
      "balance_loss_clip": 1.04089355,
      "balance_loss_mlp": 1.01736116,
      "epoch": 0.6901941922683821,
      "flos": 34857194716800.0,
      "grad_norm": 1.77334945872231,
      "language_loss": 0.74368823,
      "learning_rate": 9.249810806208139e-07,
      "loss": 0.7651087,
      "num_input_tokens_seen": 123485480,
      "step": 5740,
      "time_per_iteration": 2.7054927349090576
    },
    {
      "auxiliary_loss_clip": 0.01107066,
      "auxiliary_loss_mlp": 0.00760415,
      "balance_loss_clip": 1.03845954,
      "balance_loss_mlp": 1.00023735,
      "epoch": 0.6903144351590212,
      "flos": 16253672976000.0,
      "grad_norm": 1.926815259565751,
      "language_loss": 0.80127066,
      "learning_rate": 9.243242883226627e-07,
      "loss": 0.81994551,
      "num_input_tokens_seen": 123504575,
      "step": 5741,
      "time_per_iteration": 2.5816588401794434
    },
    {
      "auxiliary_loss_clip": 0.0115456,
      "auxiliary_loss_mlp": 0.01026079,
      "balance_loss_clip": 1.04450166,
      "balance_loss_mlp": 1.01834249,
      "epoch": 0.6904346780496603,
      "flos": 28035137255040.0,
      "grad_norm": 1.756565120371662,
      "language_loss": 0.69491482,
      "learning_rate": 9.236676592036628e-07,
      "loss": 0.71672118,
      "num_input_tokens_seen": 123524250,
      "step": 5742,
      "time_per_iteration": 2.5537710189819336
    },
    {
      "auxiliary_loss_clip": 0.01137743,
      "auxiliary_loss_mlp": 0.01024637,
      "balance_loss_clip": 1.0504117,
      "balance_loss_mlp": 1.01746619,
      "epoch": 0.6905549209402994,
      "flos": 23624266008960.0,
      "grad_norm": 1.7368069349114752,
      "language_loss": 0.73636407,
      "learning_rate": 9.230111933634228e-07,
      "loss": 0.75798786,
      "num_input_tokens_seen": 123545845,
      "step": 5743,
      "time_per_iteration": 3.376593828201294
    },
    {
      "auxiliary_loss_clip": 0.01157333,
      "auxiliary_loss_mlp": 0.01031381,
      "balance_loss_clip": 1.05115724,
      "balance_loss_mlp": 1.02401042,
      "epoch": 0.6906751638309385,
      "flos": 23114945111040.0,
      "grad_norm": 1.5491217838116822,
      "language_loss": 0.80557334,
      "learning_rate": 9.223548909015288e-07,
      "loss": 0.82746041,
      "num_input_tokens_seen": 123567535,
      "step": 5744,
      "time_per_iteration": 2.5351738929748535
    },
    {
      "auxiliary_loss_clip": 0.01104795,
      "auxiliary_loss_mlp": 0.0103055,
      "balance_loss_clip": 1.04376888,
      "balance_loss_mlp": 1.02379966,
      "epoch": 0.6907954067215776,
      "flos": 27305468375040.0,
      "grad_norm": 1.8054295318970235,
      "language_loss": 0.7180391,
      "learning_rate": 9.216987519175407e-07,
      "loss": 0.73939252,
      "num_input_tokens_seen": 123587710,
      "step": 5745,
      "time_per_iteration": 2.626535177230835
    },
    {
      "auxiliary_loss_clip": 0.01146995,
      "auxiliary_loss_mlp": 0.01024023,
      "balance_loss_clip": 1.04758334,
      "balance_loss_mlp": 1.01664126,
      "epoch": 0.6909156496122166,
      "flos": 21689399070720.0,
      "grad_norm": 1.5879151496259174,
      "language_loss": 0.68259799,
      "learning_rate": 9.210427765109942e-07,
      "loss": 0.70430815,
      "num_input_tokens_seen": 123607385,
      "step": 5746,
      "time_per_iteration": 2.490229606628418
    },
    {
      "auxiliary_loss_clip": 0.01138253,
      "auxiliary_loss_mlp": 0.01032623,
      "balance_loss_clip": 1.04463732,
      "balance_loss_mlp": 1.0250001,
      "epoch": 0.6910358925028558,
      "flos": 22561453463040.0,
      "grad_norm": 3.441334886836369,
      "language_loss": 0.81636655,
      "learning_rate": 9.20386964781402e-07,
      "loss": 0.83807528,
      "num_input_tokens_seen": 123625405,
      "step": 5747,
      "time_per_iteration": 2.528012752532959
    },
    {
      "auxiliary_loss_clip": 0.01136347,
      "auxiliary_loss_mlp": 0.01029206,
      "balance_loss_clip": 1.04535162,
      "balance_loss_mlp": 1.02157021,
      "epoch": 0.6911561353934949,
      "flos": 22054107813120.0,
      "grad_norm": 2.5925726725448857,
      "language_loss": 0.84636551,
      "learning_rate": 9.197313168282472e-07,
      "loss": 0.86802107,
      "num_input_tokens_seen": 123642850,
      "step": 5748,
      "time_per_iteration": 2.554379940032959
    },
    {
      "auxiliary_loss_clip": 0.01146882,
      "auxiliary_loss_mlp": 0.01029538,
      "balance_loss_clip": 1.04441667,
      "balance_loss_mlp": 1.02206016,
      "epoch": 0.6912763782841339,
      "flos": 24206557386240.0,
      "grad_norm": 3.3948231664646764,
      "language_loss": 0.71999443,
      "learning_rate": 9.190758327509935e-07,
      "loss": 0.74175858,
      "num_input_tokens_seen": 123661595,
      "step": 5749,
      "time_per_iteration": 2.510038137435913
    },
    {
      "auxiliary_loss_clip": 0.01019319,
      "auxiliary_loss_mlp": 0.00751307,
      "balance_loss_clip": 1.01276302,
      "balance_loss_mlp": 1.0001061,
      "epoch": 0.6913966211747731,
      "flos": 52329641091840.0,
      "grad_norm": 0.9268056157586627,
      "language_loss": 0.64469326,
      "learning_rate": 9.184205126490767e-07,
      "loss": 0.66239941,
      "num_input_tokens_seen": 123710490,
      "step": 5750,
      "time_per_iteration": 2.9966542720794678
    },
    {
      "auxiliary_loss_clip": 0.0103025,
      "auxiliary_loss_mlp": 0.00751179,
      "balance_loss_clip": 1.01048756,
      "balance_loss_mlp": 1.00002694,
      "epoch": 0.6915168640654121,
      "flos": 66741274851840.0,
      "grad_norm": 1.112088164373328,
      "language_loss": 0.5967145,
      "learning_rate": 9.177653566219075e-07,
      "loss": 0.61452878,
      "num_input_tokens_seen": 123765215,
      "step": 5751,
      "time_per_iteration": 3.0298500061035156
    },
    {
      "auxiliary_loss_clip": 0.01127851,
      "auxiliary_loss_mlp": 0.01027952,
      "balance_loss_clip": 1.04264474,
      "balance_loss_mlp": 1.02055252,
      "epoch": 0.6916371069560512,
      "flos": 18296523175680.0,
      "grad_norm": 2.086506029357863,
      "language_loss": 0.75620037,
      "learning_rate": 9.171103647688744e-07,
      "loss": 0.77775842,
      "num_input_tokens_seen": 123783955,
      "step": 5752,
      "time_per_iteration": 2.5580313205718994
    },
    {
      "auxiliary_loss_clip": 0.01073691,
      "auxiliary_loss_mlp": 0.01026766,
      "balance_loss_clip": 1.03857088,
      "balance_loss_mlp": 1.01995611,
      "epoch": 0.6917573498466904,
      "flos": 19645794685440.0,
      "grad_norm": 3.3424390289817736,
      "language_loss": 0.68924171,
      "learning_rate": 9.164555371893367e-07,
      "loss": 0.71024632,
      "num_input_tokens_seen": 123803885,
      "step": 5753,
      "time_per_iteration": 3.4237396717071533
    },
    {
      "auxiliary_loss_clip": 0.01151898,
      "auxiliary_loss_mlp": 0.00760017,
      "balance_loss_clip": 1.04760933,
      "balance_loss_mlp": 1.00022769,
      "epoch": 0.6918775927373294,
      "flos": 14210319985920.0,
      "grad_norm": 1.869484816953531,
      "language_loss": 0.74994606,
      "learning_rate": 9.158008739826333e-07,
      "loss": 0.7690652,
      "num_input_tokens_seen": 123821485,
      "step": 5754,
      "time_per_iteration": 2.4832136631011963
    },
    {
      "auxiliary_loss_clip": 0.01136767,
      "auxiliary_loss_mlp": 0.01034513,
      "balance_loss_clip": 1.04690266,
      "balance_loss_mlp": 1.02720213,
      "epoch": 0.6919978356279685,
      "flos": 23985455218560.0,
      "grad_norm": 1.548373761498636,
      "language_loss": 0.86625338,
      "learning_rate": 9.151463752480744e-07,
      "loss": 0.88796622,
      "num_input_tokens_seen": 123840215,
      "step": 5755,
      "time_per_iteration": 3.395021438598633
    },
    {
      "auxiliary_loss_clip": 0.0111494,
      "auxiliary_loss_mlp": 0.01027691,
      "balance_loss_clip": 1.04363871,
      "balance_loss_mlp": 1.02061582,
      "epoch": 0.6921180785186076,
      "flos": 23622937205760.0,
      "grad_norm": 1.3771447696298464,
      "language_loss": 0.80349791,
      "learning_rate": 9.144920410849493e-07,
      "loss": 0.82492423,
      "num_input_tokens_seen": 123861450,
      "step": 5756,
      "time_per_iteration": 3.3393185138702393
    },
    {
      "auxiliary_loss_clip": 0.01144161,
      "auxiliary_loss_mlp": 0.01024398,
      "balance_loss_clip": 1.0472964,
      "balance_loss_mlp": 1.01710534,
      "epoch": 0.6922383214092467,
      "flos": 21142623265920.0,
      "grad_norm": 1.913948526839435,
      "language_loss": 0.80599356,
      "learning_rate": 9.138378715925176e-07,
      "loss": 0.82767922,
      "num_input_tokens_seen": 123880545,
      "step": 5757,
      "time_per_iteration": 2.530811309814453
    },
    {
      "auxiliary_loss_clip": 0.01132139,
      "auxiliary_loss_mlp": 0.01027845,
      "balance_loss_clip": 1.04373765,
      "balance_loss_mlp": 1.02029324,
      "epoch": 0.6923585642998857,
      "flos": 21470667200640.0,
      "grad_norm": 1.6404207756645195,
      "language_loss": 0.8086257,
      "learning_rate": 9.131838668700167e-07,
      "loss": 0.83022559,
      "num_input_tokens_seen": 123900615,
      "step": 5758,
      "time_per_iteration": 2.534411907196045
    },
    {
      "auxiliary_loss_clip": 0.01121952,
      "auxiliary_loss_mlp": 0.0102529,
      "balance_loss_clip": 1.04271126,
      "balance_loss_mlp": 1.0177319,
      "epoch": 0.6924788071905249,
      "flos": 21105204272640.0,
      "grad_norm": 1.716142850470053,
      "language_loss": 0.86380702,
      "learning_rate": 9.125300270166598e-07,
      "loss": 0.88527942,
      "num_input_tokens_seen": 123921220,
      "step": 5759,
      "time_per_iteration": 2.59421968460083
    },
    {
      "auxiliary_loss_clip": 0.01128518,
      "auxiliary_loss_mlp": 0.01024648,
      "balance_loss_clip": 1.04391289,
      "balance_loss_mlp": 1.01754594,
      "epoch": 0.692599050081164,
      "flos": 26250018117120.0,
      "grad_norm": 1.6981209311535594,
      "language_loss": 0.8576951,
      "learning_rate": 9.118763521316324e-07,
      "loss": 0.87922674,
      "num_input_tokens_seen": 123941795,
      "step": 5760,
      "time_per_iteration": 2.6021220684051514
    },
    {
      "auxiliary_loss_clip": 0.01166478,
      "auxiliary_loss_mlp": 0.00760814,
      "balance_loss_clip": 1.0481956,
      "balance_loss_mlp": 1.00020564,
      "epoch": 0.692719292971803,
      "flos": 20885215426560.0,
      "grad_norm": 1.5422521662049735,
      "language_loss": 0.76197231,
      "learning_rate": 9.112228423140987e-07,
      "loss": 0.78124523,
      "num_input_tokens_seen": 123960715,
      "step": 5761,
      "time_per_iteration": 2.4581542015075684
    },
    {
      "auxiliary_loss_clip": 0.01142284,
      "auxiliary_loss_mlp": 0.01029374,
      "balance_loss_clip": 1.04612255,
      "balance_loss_mlp": 1.02168441,
      "epoch": 0.6928395358624422,
      "flos": 25921938268800.0,
      "grad_norm": 2.2946724632605657,
      "language_loss": 0.86477536,
      "learning_rate": 9.105694976631932e-07,
      "loss": 0.88649189,
      "num_input_tokens_seen": 123978625,
      "step": 5762,
      "time_per_iteration": 2.544137716293335
    },
    {
      "auxiliary_loss_clip": 0.01155256,
      "auxiliary_loss_mlp": 0.01022126,
      "balance_loss_clip": 1.05047154,
      "balance_loss_mlp": 1.01549816,
      "epoch": 0.6929597787530812,
      "flos": 23586559706880.0,
      "grad_norm": 2.0024764046055523,
      "language_loss": 0.72764099,
      "learning_rate": 9.099163182780283e-07,
      "loss": 0.7494148,
      "num_input_tokens_seen": 123996780,
      "step": 5763,
      "time_per_iteration": 2.513568639755249
    },
    {
      "auxiliary_loss_clip": 0.01135002,
      "auxiliary_loss_mlp": 0.01031447,
      "balance_loss_clip": 1.04698849,
      "balance_loss_mlp": 1.02450323,
      "epoch": 0.6930800216437203,
      "flos": 18255656476800.0,
      "grad_norm": 2.979453153673986,
      "language_loss": 0.48804629,
      "learning_rate": 9.092633042576916e-07,
      "loss": 0.50971079,
      "num_input_tokens_seen": 124014045,
      "step": 5764,
      "time_per_iteration": 2.489847421646118
    },
    {
      "auxiliary_loss_clip": 0.01140231,
      "auxiliary_loss_mlp": 0.01024743,
      "balance_loss_clip": 1.05029774,
      "balance_loss_mlp": 1.01783228,
      "epoch": 0.6932002645343595,
      "flos": 29168621809920.0,
      "grad_norm": 1.9542100702187768,
      "language_loss": 0.55829382,
      "learning_rate": 9.086104557012446e-07,
      "loss": 0.57994354,
      "num_input_tokens_seen": 124034615,
      "step": 5765,
      "time_per_iteration": 2.602405071258545
    },
    {
      "auxiliary_loss_clip": 0.01143328,
      "auxiliary_loss_mlp": 0.01024036,
      "balance_loss_clip": 1.04605198,
      "balance_loss_mlp": 1.01699615,
      "epoch": 0.6933205074249985,
      "flos": 23842746483840.0,
      "grad_norm": 1.8121072065876174,
      "language_loss": 0.65813488,
      "learning_rate": 9.079577727077239e-07,
      "loss": 0.67980856,
      "num_input_tokens_seen": 124053445,
      "step": 5766,
      "time_per_iteration": 2.505826711654663
    },
    {
      "auxiliary_loss_clip": 0.01153524,
      "auxiliary_loss_mlp": 0.01028309,
      "balance_loss_clip": 1.04785573,
      "balance_loss_mlp": 1.02075088,
      "epoch": 0.6934407503156376,
      "flos": 24166696268160.0,
      "grad_norm": 2.044699950799127,
      "language_loss": 0.71895641,
      "learning_rate": 9.073052553761404e-07,
      "loss": 0.74077475,
      "num_input_tokens_seen": 124072810,
      "step": 5767,
      "time_per_iteration": 2.5104804039001465
    },
    {
      "auxiliary_loss_clip": 0.01113938,
      "auxiliary_loss_mlp": 0.01022284,
      "balance_loss_clip": 1.04355717,
      "balance_loss_mlp": 1.01460052,
      "epoch": 0.6935609932062767,
      "flos": 20631327120000.0,
      "grad_norm": 1.5904178488206369,
      "language_loss": 0.78221267,
      "learning_rate": 9.066529038054805e-07,
      "loss": 0.80357492,
      "num_input_tokens_seen": 124092875,
      "step": 5768,
      "time_per_iteration": 2.6197214126586914
    },
    {
      "auxiliary_loss_clip": 0.01137075,
      "auxiliary_loss_mlp": 0.01025585,
      "balance_loss_clip": 1.04582739,
      "balance_loss_mlp": 1.01918089,
      "epoch": 0.6936812360969158,
      "flos": 18254184019200.0,
      "grad_norm": 1.7028796858108963,
      "language_loss": 0.73843378,
      "learning_rate": 9.060007180947071e-07,
      "loss": 0.76006043,
      "num_input_tokens_seen": 124110930,
      "step": 5769,
      "time_per_iteration": 3.3316965103149414
    },
    {
      "auxiliary_loss_clip": 0.01106749,
      "auxiliary_loss_mlp": 0.01026449,
      "balance_loss_clip": 1.03931403,
      "balance_loss_mlp": 1.01872444,
      "epoch": 0.6938014789875548,
      "flos": 31317336368640.0,
      "grad_norm": 2.569058229895464,
      "language_loss": 0.73034328,
      "learning_rate": 9.053486983427534e-07,
      "loss": 0.75167525,
      "num_input_tokens_seen": 124132180,
      "step": 5770,
      "time_per_iteration": 2.667313575744629
    },
    {
      "auxiliary_loss_clip": 0.01144825,
      "auxiliary_loss_mlp": 0.01027993,
      "balance_loss_clip": 1.04714537,
      "balance_loss_mlp": 1.02103436,
      "epoch": 0.6939217218781939,
      "flos": 17528429721600.0,
      "grad_norm": 1.788468537288641,
      "language_loss": 0.7060408,
      "learning_rate": 9.046968446485326e-07,
      "loss": 0.72776896,
      "num_input_tokens_seen": 124150585,
      "step": 5771,
      "time_per_iteration": 2.5142362117767334
    },
    {
      "auxiliary_loss_clip": 0.01156985,
      "auxiliary_loss_mlp": 0.01029736,
      "balance_loss_clip": 1.04905438,
      "balance_loss_mlp": 1.02153981,
      "epoch": 0.6940419647688331,
      "flos": 18551776199040.0,
      "grad_norm": 2.3075660714802786,
      "language_loss": 0.70356917,
      "learning_rate": 9.040451571109295e-07,
      "loss": 0.72543627,
      "num_input_tokens_seen": 124166205,
      "step": 5772,
      "time_per_iteration": 2.4545812606811523
    },
    {
      "auxiliary_loss_clip": 0.01021994,
      "auxiliary_loss_mlp": 0.01001371,
      "balance_loss_clip": 1.00844491,
      "balance_loss_mlp": 1.00028038,
      "epoch": 0.6941622076594721,
      "flos": 66926286829440.0,
      "grad_norm": 0.8276997743815718,
      "language_loss": 0.60385895,
      "learning_rate": 9.033936358288042e-07,
      "loss": 0.62409258,
      "num_input_tokens_seen": 124219940,
      "step": 5773,
      "time_per_iteration": 3.0159482955932617
    },
    {
      "auxiliary_loss_clip": 0.01168923,
      "auxiliary_loss_mlp": 0.01025496,
      "balance_loss_clip": 1.05043828,
      "balance_loss_mlp": 1.01879609,
      "epoch": 0.6942824505501112,
      "flos": 26578062051840.0,
      "grad_norm": 1.5751878367666334,
      "language_loss": 0.82283241,
      "learning_rate": 9.027422809009937e-07,
      "loss": 0.84477663,
      "num_input_tokens_seen": 124239885,
      "step": 5774,
      "time_per_iteration": 2.5000133514404297
    },
    {
      "auxiliary_loss_clip": 0.01152694,
      "auxiliary_loss_mlp": 0.01030818,
      "balance_loss_clip": 1.04500747,
      "balance_loss_mlp": 1.02361774,
      "epoch": 0.6944026934407503,
      "flos": 21248308056960.0,
      "grad_norm": 1.561970681331322,
      "language_loss": 0.83204174,
      "learning_rate": 9.020910924263054e-07,
      "loss": 0.85387689,
      "num_input_tokens_seen": 124258410,
      "step": 5775,
      "time_per_iteration": 2.4969499111175537
    },
    {
      "auxiliary_loss_clip": 0.01020285,
      "auxiliary_loss_mlp": 0.0100137,
      "balance_loss_clip": 1.00840592,
      "balance_loss_mlp": 1.00023746,
      "epoch": 0.6945229363313894,
      "flos": 70677191537280.0,
      "grad_norm": 0.8218927528050084,
      "language_loss": 0.58166867,
      "learning_rate": 9.014400705035261e-07,
      "loss": 0.6018852,
      "num_input_tokens_seen": 124315315,
      "step": 5776,
      "time_per_iteration": 3.176245927810669
    },
    {
      "auxiliary_loss_clip": 0.0116735,
      "auxiliary_loss_mlp": 0.01026914,
      "balance_loss_clip": 1.05123854,
      "balance_loss_mlp": 1.01949942,
      "epoch": 0.6946431792220285,
      "flos": 18952934267520.0,
      "grad_norm": 1.870377877803489,
      "language_loss": 0.76448852,
      "learning_rate": 9.00789215231414e-07,
      "loss": 0.78643113,
      "num_input_tokens_seen": 124333710,
      "step": 5777,
      "time_per_iteration": 2.450627565383911
    },
    {
      "auxiliary_loss_clip": 0.01122008,
      "auxiliary_loss_mlp": 0.00761234,
      "balance_loss_clip": 1.04038525,
      "balance_loss_mlp": 1.000193,
      "epoch": 0.6947634221126676,
      "flos": 20338834671360.0,
      "grad_norm": 1.8085806150652846,
      "language_loss": 0.81715369,
      "learning_rate": 9.001385267087056e-07,
      "loss": 0.83598614,
      "num_input_tokens_seen": 124352855,
      "step": 5778,
      "time_per_iteration": 2.5647757053375244
    },
    {
      "auxiliary_loss_clip": 0.01157353,
      "auxiliary_loss_mlp": 0.01027376,
      "balance_loss_clip": 1.05036998,
      "balance_loss_mlp": 1.02036059,
      "epoch": 0.6948836650033067,
      "flos": 21833723917440.0,
      "grad_norm": 1.4730761752565158,
      "language_loss": 0.70301723,
      "learning_rate": 8.994880050341072e-07,
      "loss": 0.72486454,
      "num_input_tokens_seen": 124372960,
      "step": 5779,
      "time_per_iteration": 3.2798702716827393
    },
    {
      "auxiliary_loss_clip": 0.01132889,
      "auxiliary_loss_mlp": 0.01036425,
      "balance_loss_clip": 1.04597723,
      "balance_loss_mlp": 1.02902269,
      "epoch": 0.6950039078939457,
      "flos": 23657519024640.0,
      "grad_norm": 1.7246428544315826,
      "language_loss": 0.77391571,
      "learning_rate": 8.988376503063026e-07,
      "loss": 0.79560882,
      "num_input_tokens_seen": 124394220,
      "step": 5780,
      "time_per_iteration": 2.579139232635498
    },
    {
      "auxiliary_loss_clip": 0.0111982,
      "auxiliary_loss_mlp": 0.01025592,
      "balance_loss_clip": 1.04590178,
      "balance_loss_mlp": 1.01790333,
      "epoch": 0.6951241507845849,
      "flos": 21792462168960.0,
      "grad_norm": 2.5855876339743507,
      "language_loss": 0.81502253,
      "learning_rate": 8.981874626239521e-07,
      "loss": 0.83647668,
      "num_input_tokens_seen": 124412795,
      "step": 5781,
      "time_per_iteration": 3.368726968765259
    },
    {
      "auxiliary_loss_clip": 0.01156242,
      "auxiliary_loss_mlp": 0.01035422,
      "balance_loss_clip": 1.05069363,
      "balance_loss_mlp": 1.02796507,
      "epoch": 0.695244393675224,
      "flos": 14647568244480.0,
      "grad_norm": 1.9070884902838607,
      "language_loss": 0.88525248,
      "learning_rate": 8.975374420856872e-07,
      "loss": 0.9071691,
      "num_input_tokens_seen": 124429690,
      "step": 5782,
      "time_per_iteration": 3.2339119911193848
    },
    {
      "auxiliary_loss_clip": 0.01115648,
      "auxiliary_loss_mlp": 0.01024172,
      "balance_loss_clip": 1.04168558,
      "balance_loss_mlp": 1.01667356,
      "epoch": 0.695364636565863,
      "flos": 16873203778560.0,
      "grad_norm": 2.2694039503162413,
      "language_loss": 0.72529227,
      "learning_rate": 8.968875887901157e-07,
      "loss": 0.74669051,
      "num_input_tokens_seen": 124447070,
      "step": 5783,
      "time_per_iteration": 2.5403735637664795
    },
    {
      "auxiliary_loss_clip": 0.01139984,
      "auxiliary_loss_mlp": 0.0103106,
      "balance_loss_clip": 1.04504251,
      "balance_loss_mlp": 1.02350807,
      "epoch": 0.6954848794565022,
      "flos": 19354523299200.0,
      "grad_norm": 1.796993001160796,
      "language_loss": 0.62148094,
      "learning_rate": 8.9623790283582e-07,
      "loss": 0.6431914,
      "num_input_tokens_seen": 124464950,
      "step": 5784,
      "time_per_iteration": 2.550201177597046
    },
    {
      "auxiliary_loss_clip": 0.0112852,
      "auxiliary_loss_mlp": 0.01033925,
      "balance_loss_clip": 1.04702401,
      "balance_loss_mlp": 1.02675462,
      "epoch": 0.6956051223471412,
      "flos": 18990209606400.0,
      "grad_norm": 1.9500221153847093,
      "language_loss": 0.76325321,
      "learning_rate": 8.955883843213561e-07,
      "loss": 0.78487772,
      "num_input_tokens_seen": 124483965,
      "step": 5785,
      "time_per_iteration": 2.5740325450897217
    },
    {
      "auxiliary_loss_clip": 0.01162492,
      "auxiliary_loss_mlp": 0.01030506,
      "balance_loss_clip": 1.0501318,
      "balance_loss_mlp": 1.02271605,
      "epoch": 0.6957253652377803,
      "flos": 16107229226880.0,
      "grad_norm": 1.7977836921335948,
      "language_loss": 0.86733103,
      "learning_rate": 8.949390333452569e-07,
      "loss": 0.88926101,
      "num_input_tokens_seen": 124501910,
      "step": 5786,
      "time_per_iteration": 2.4971866607666016
    },
    {
      "auxiliary_loss_clip": 0.01167303,
      "auxiliary_loss_mlp": 0.0103211,
      "balance_loss_clip": 1.05040681,
      "balance_loss_mlp": 1.02503538,
      "epoch": 0.6958456081284194,
      "flos": 29388646569600.0,
      "grad_norm": 2.06581412377045,
      "language_loss": 0.67747438,
      "learning_rate": 8.942898500060279e-07,
      "loss": 0.69946849,
      "num_input_tokens_seen": 124521625,
      "step": 5787,
      "time_per_iteration": 2.522145986557007
    },
    {
      "auxiliary_loss_clip": 0.011167,
      "auxiliary_loss_mlp": 0.01034258,
      "balance_loss_clip": 1.04581547,
      "balance_loss_mlp": 1.02693212,
      "epoch": 0.6959658510190585,
      "flos": 25154850395520.0,
      "grad_norm": 4.081670175532175,
      "language_loss": 0.71256196,
      "learning_rate": 8.936408344021493e-07,
      "loss": 0.73407149,
      "num_input_tokens_seen": 124538540,
      "step": 5788,
      "time_per_iteration": 2.629964590072632
    },
    {
      "auxiliary_loss_clip": 0.01145917,
      "auxiliary_loss_mlp": 0.01032924,
      "balance_loss_clip": 1.0502218,
      "balance_loss_mlp": 1.02543736,
      "epoch": 0.6960860939096976,
      "flos": 42814388759040.0,
      "grad_norm": 2.254324620653938,
      "language_loss": 0.71200734,
      "learning_rate": 8.929919866320765e-07,
      "loss": 0.73379576,
      "num_input_tokens_seen": 124559355,
      "step": 5789,
      "time_per_iteration": 2.724989891052246
    },
    {
      "auxiliary_loss_clip": 0.01131454,
      "auxiliary_loss_mlp": 0.00761693,
      "balance_loss_clip": 1.04522133,
      "balance_loss_mlp": 1.00017595,
      "epoch": 0.6962063368003367,
      "flos": 17566566986880.0,
      "grad_norm": 1.872096031518265,
      "language_loss": 0.81174505,
      "learning_rate": 8.923433067942385e-07,
      "loss": 0.83067656,
      "num_input_tokens_seen": 124577920,
      "step": 5790,
      "time_per_iteration": 2.5689969062805176
    },
    {
      "auxiliary_loss_clip": 0.01131778,
      "auxiliary_loss_mlp": 0.01029979,
      "balance_loss_clip": 1.0484556,
      "balance_loss_mlp": 1.02249539,
      "epoch": 0.6963265796909758,
      "flos": 21251648021760.0,
      "grad_norm": 1.983868673448697,
      "language_loss": 0.6856277,
      "learning_rate": 8.916947949870417e-07,
      "loss": 0.70724523,
      "num_input_tokens_seen": 124597585,
      "step": 5791,
      "time_per_iteration": 2.555908441543579
    },
    {
      "auxiliary_loss_clip": 0.01048802,
      "auxiliary_loss_mlp": 0.01000939,
      "balance_loss_clip": 1.00962245,
      "balance_loss_mlp": 0.9998126,
      "epoch": 0.6964468225816148,
      "flos": 68828295801600.0,
      "grad_norm": 0.7393180438891565,
      "language_loss": 0.58109879,
      "learning_rate": 8.910464513088615e-07,
      "loss": 0.60159624,
      "num_input_tokens_seen": 124661625,
      "step": 5792,
      "time_per_iteration": 3.1899287700653076
    },
    {
      "auxiliary_loss_clip": 0.01134755,
      "auxiliary_loss_mlp": 0.01030898,
      "balance_loss_clip": 1.04666054,
      "balance_loss_mlp": 1.0230931,
      "epoch": 0.696567065472254,
      "flos": 18950887192320.0,
      "grad_norm": 1.8140604490714414,
      "language_loss": 0.78513747,
      "learning_rate": 8.903982758580542e-07,
      "loss": 0.80679399,
      "num_input_tokens_seen": 124680565,
      "step": 5793,
      "time_per_iteration": 2.524005174636841
    },
    {
      "auxiliary_loss_clip": 0.01132425,
      "auxiliary_loss_mlp": 0.01025292,
      "balance_loss_clip": 1.04396737,
      "balance_loss_mlp": 1.01803756,
      "epoch": 0.696687308362893,
      "flos": 22856675345280.0,
      "grad_norm": 1.9294533179211195,
      "language_loss": 0.80509031,
      "learning_rate": 8.897502687329457e-07,
      "loss": 0.82666743,
      "num_input_tokens_seen": 124700365,
      "step": 5794,
      "time_per_iteration": 2.539825677871704
    },
    {
      "auxiliary_loss_clip": 0.01120303,
      "auxiliary_loss_mlp": 0.01030551,
      "balance_loss_clip": 1.0431751,
      "balance_loss_mlp": 1.02339578,
      "epoch": 0.6968075512535321,
      "flos": 24972926987520.0,
      "grad_norm": 1.9036001341227715,
      "language_loss": 0.799164,
      "learning_rate": 8.891024300318382e-07,
      "loss": 0.82067257,
      "num_input_tokens_seen": 124718935,
      "step": 5795,
      "time_per_iteration": 3.347975015640259
    },
    {
      "auxiliary_loss_clip": 0.01117112,
      "auxiliary_loss_mlp": 0.01026731,
      "balance_loss_clip": 1.04245162,
      "balance_loss_mlp": 1.02012956,
      "epoch": 0.6969277941441713,
      "flos": 21030438113280.0,
      "grad_norm": 1.4160816358124628,
      "language_loss": 0.75874621,
      "learning_rate": 8.884547598530103e-07,
      "loss": 0.78018463,
      "num_input_tokens_seen": 124739505,
      "step": 5796,
      "time_per_iteration": 2.771071195602417
    },
    {
      "auxiliary_loss_clip": 0.01071724,
      "auxiliary_loss_mlp": 0.01026817,
      "balance_loss_clip": 1.03796029,
      "balance_loss_mlp": 1.019593,
      "epoch": 0.6970480370348103,
      "flos": 21579404647680.0,
      "grad_norm": 1.8229623382749671,
      "language_loss": 0.75096893,
      "learning_rate": 8.8780725829471e-07,
      "loss": 0.77195442,
      "num_input_tokens_seen": 124757410,
      "step": 5797,
      "time_per_iteration": 2.6394848823547363
    },
    {
      "auxiliary_loss_clip": 0.01167765,
      "auxiliary_loss_mlp": 0.01033042,
      "balance_loss_clip": 1.04950523,
      "balance_loss_mlp": 1.02552891,
      "epoch": 0.6971682799254494,
      "flos": 22419175691520.0,
      "grad_norm": 1.758212336818437,
      "language_loss": 0.78020328,
      "learning_rate": 8.87159925455165e-07,
      "loss": 0.80221128,
      "num_input_tokens_seen": 124777240,
      "step": 5798,
      "time_per_iteration": 2.461414337158203
    },
    {
      "auxiliary_loss_clip": 0.01121866,
      "auxiliary_loss_mlp": 0.01025459,
      "balance_loss_clip": 1.04454195,
      "balance_loss_mlp": 1.01845586,
      "epoch": 0.6972885228160886,
      "flos": 20005834659840.0,
      "grad_norm": 1.8353313739091295,
      "language_loss": 0.73191297,
      "learning_rate": 8.865127614325738e-07,
      "loss": 0.7533862,
      "num_input_tokens_seen": 124795670,
      "step": 5799,
      "time_per_iteration": 2.5503647327423096
    },
    {
      "auxiliary_loss_clip": 0.01131247,
      "auxiliary_loss_mlp": 0.01032613,
      "balance_loss_clip": 1.04298735,
      "balance_loss_mlp": 1.02458465,
      "epoch": 0.6974087657067276,
      "flos": 37853437656960.0,
      "grad_norm": 1.7197523792300766,
      "language_loss": 0.66769058,
      "learning_rate": 8.85865766325113e-07,
      "loss": 0.68932915,
      "num_input_tokens_seen": 124819600,
      "step": 5800,
      "time_per_iteration": 2.675478219985962
    },
    {
      "auxiliary_loss_clip": 0.01135178,
      "auxiliary_loss_mlp": 0.01028236,
      "balance_loss_clip": 1.04574418,
      "balance_loss_mlp": 1.02106285,
      "epoch": 0.6975290085973667,
      "flos": 29489267543040.0,
      "grad_norm": 2.5630787160092727,
      "language_loss": 0.71849018,
      "learning_rate": 8.852189402309287e-07,
      "loss": 0.74012429,
      "num_input_tokens_seen": 124838785,
      "step": 5801,
      "time_per_iteration": 2.579350709915161
    },
    {
      "auxiliary_loss_clip": 0.01154822,
      "auxiliary_loss_mlp": 0.01026283,
      "balance_loss_clip": 1.05144083,
      "balance_loss_mlp": 1.01944065,
      "epoch": 0.6976492514880057,
      "flos": 12895630295040.0,
      "grad_norm": 2.0031023016269565,
      "language_loss": 0.73839939,
      "learning_rate": 8.845722832481441e-07,
      "loss": 0.76021039,
      "num_input_tokens_seen": 124854215,
      "step": 5802,
      "time_per_iteration": 2.4543538093566895
    },
    {
      "auxiliary_loss_clip": 0.0115238,
      "auxiliary_loss_mlp": 0.01033203,
      "balance_loss_clip": 1.04875088,
      "balance_loss_mlp": 1.02539515,
      "epoch": 0.6977694943786449,
      "flos": 24352929308160.0,
      "grad_norm": 1.74198669929843,
      "language_loss": 0.77000058,
      "learning_rate": 8.83925795474858e-07,
      "loss": 0.79185641,
      "num_input_tokens_seen": 124874340,
      "step": 5803,
      "time_per_iteration": 2.5304338932037354
    },
    {
      "auxiliary_loss_clip": 0.01124046,
      "auxiliary_loss_mlp": 0.01032507,
      "balance_loss_clip": 1.04799771,
      "balance_loss_mlp": 1.02501428,
      "epoch": 0.6978897372692839,
      "flos": 29898470257920.0,
      "grad_norm": 2.1002700245293386,
      "language_loss": 0.58731067,
      "learning_rate": 8.832794770091414e-07,
      "loss": 0.60887623,
      "num_input_tokens_seen": 124895175,
      "step": 5804,
      "time_per_iteration": 2.6071298122406006
    },
    {
      "auxiliary_loss_clip": 0.01142207,
      "auxiliary_loss_mlp": 0.01026802,
      "balance_loss_clip": 1.04473281,
      "balance_loss_mlp": 1.0202843,
      "epoch": 0.698009980159923,
      "flos": 21761579450880.0,
      "grad_norm": 1.9165332500855048,
      "language_loss": 0.82098019,
      "learning_rate": 8.826333279490401e-07,
      "loss": 0.84267032,
      "num_input_tokens_seen": 124915810,
      "step": 5805,
      "time_per_iteration": 3.311262607574463
    },
    {
      "auxiliary_loss_clip": 0.01143534,
      "auxiliary_loss_mlp": 0.01026617,
      "balance_loss_clip": 1.04911828,
      "balance_loss_mlp": 1.01908875,
      "epoch": 0.6981302230505622,
      "flos": 19857164267520.0,
      "grad_norm": 1.980368275503191,
      "language_loss": 0.68050158,
      "learning_rate": 8.819873483925748e-07,
      "loss": 0.70220315,
      "num_input_tokens_seen": 124932930,
      "step": 5806,
      "time_per_iteration": 2.5157883167266846
    },
    {
      "auxiliary_loss_clip": 0.01131221,
      "auxiliary_loss_mlp": 0.007605,
      "balance_loss_clip": 1.04708123,
      "balance_loss_mlp": 1.00015974,
      "epoch": 0.6982504659412012,
      "flos": 22198648141440.0,
      "grad_norm": 1.8202533774014515,
      "language_loss": 0.74286604,
      "learning_rate": 8.81341538437739e-07,
      "loss": 0.76178324,
      "num_input_tokens_seen": 124951220,
      "step": 5807,
      "time_per_iteration": 3.402418613433838
    },
    {
      "auxiliary_loss_clip": 0.01143384,
      "auxiliary_loss_mlp": 0.01022692,
      "balance_loss_clip": 1.04508853,
      "balance_loss_mlp": 1.01540875,
      "epoch": 0.6983707088318403,
      "flos": 35588479708800.0,
      "grad_norm": 1.8554290740263364,
      "language_loss": 0.68085206,
      "learning_rate": 8.80695898182503e-07,
      "loss": 0.7025128,
      "num_input_tokens_seen": 124972200,
      "step": 5808,
      "time_per_iteration": 3.3935585021972656
    },
    {
      "auxiliary_loss_clip": 0.01040503,
      "auxiliary_loss_mlp": 0.01001898,
      "balance_loss_clip": 1.00893462,
      "balance_loss_mlp": 1.00084925,
      "epoch": 0.6984909517224794,
      "flos": 65440052760960.0,
      "grad_norm": 0.8251627906677085,
      "language_loss": 0.65138257,
      "learning_rate": 8.800504277248093e-07,
      "loss": 0.67180657,
      "num_input_tokens_seen": 125036950,
      "step": 5809,
      "time_per_iteration": 3.0882134437561035
    },
    {
      "auxiliary_loss_clip": 0.0112054,
      "auxiliary_loss_mlp": 0.00760453,
      "balance_loss_clip": 1.04870939,
      "balance_loss_mlp": 1.0001713,
      "epoch": 0.6986111946131185,
      "flos": 18546927863040.0,
      "grad_norm": 1.667063018018832,
      "language_loss": 0.74956727,
      "learning_rate": 8.794051271625753e-07,
      "loss": 0.76837724,
      "num_input_tokens_seen": 125054585,
      "step": 5810,
      "time_per_iteration": 2.525453567504883
    },
    {
      "auxiliary_loss_clip": 0.01139275,
      "auxiliary_loss_mlp": 0.0102751,
      "balance_loss_clip": 1.0454464,
      "balance_loss_mlp": 1.02027369,
      "epoch": 0.6987314375037575,
      "flos": 23039173370880.0,
      "grad_norm": 1.554309501488968,
      "language_loss": 0.82934266,
      "learning_rate": 8.787599965936925e-07,
      "loss": 0.8510105,
      "num_input_tokens_seen": 125075515,
      "step": 5811,
      "time_per_iteration": 2.540602445602417
    },
    {
      "auxiliary_loss_clip": 0.01122694,
      "auxiliary_loss_mlp": 0.0102432,
      "balance_loss_clip": 1.04706717,
      "balance_loss_mlp": 1.01714683,
      "epoch": 0.6988516803943967,
      "flos": 38400393029760.0,
      "grad_norm": 1.7781028879873573,
      "language_loss": 0.7207495,
      "learning_rate": 8.781150361160261e-07,
      "loss": 0.74221957,
      "num_input_tokens_seen": 125097425,
      "step": 5812,
      "time_per_iteration": 2.697218656539917
    },
    {
      "auxiliary_loss_clip": 0.01125493,
      "auxiliary_loss_mlp": 0.01030814,
      "balance_loss_clip": 1.0470953,
      "balance_loss_mlp": 1.02355731,
      "epoch": 0.6989719232850358,
      "flos": 24096993926400.0,
      "grad_norm": 1.6998201816230187,
      "language_loss": 0.73635674,
      "learning_rate": 8.774702458274181e-07,
      "loss": 0.75791985,
      "num_input_tokens_seen": 125117830,
      "step": 5813,
      "time_per_iteration": 2.562938928604126
    },
    {
      "auxiliary_loss_clip": 0.01148776,
      "auxiliary_loss_mlp": 0.01023648,
      "balance_loss_clip": 1.04590285,
      "balance_loss_mlp": 1.01650739,
      "epoch": 0.6990921661756748,
      "flos": 14866838818560.0,
      "grad_norm": 2.256991388508272,
      "language_loss": 0.70674235,
      "learning_rate": 8.768256258256799e-07,
      "loss": 0.72846657,
      "num_input_tokens_seen": 125134455,
      "step": 5814,
      "time_per_iteration": 2.4377193450927734
    },
    {
      "auxiliary_loss_clip": 0.01155159,
      "auxiliary_loss_mlp": 0.01025919,
      "balance_loss_clip": 1.04804659,
      "balance_loss_mlp": 1.01885283,
      "epoch": 0.699212409066314,
      "flos": 20193719725440.0,
      "grad_norm": 1.709806744181285,
      "language_loss": 0.73955011,
      "learning_rate": 8.76181176208602e-07,
      "loss": 0.76136094,
      "num_input_tokens_seen": 125152555,
      "step": 5815,
      "time_per_iteration": 2.482174873352051
    },
    {
      "auxiliary_loss_clip": 0.01097516,
      "auxiliary_loss_mlp": 0.01029185,
      "balance_loss_clip": 1.03858399,
      "balance_loss_mlp": 1.02115631,
      "epoch": 0.699332651956953,
      "flos": 19427888828160.0,
      "grad_norm": 1.723583608420724,
      "language_loss": 0.73329335,
      "learning_rate": 8.755368970739461e-07,
      "loss": 0.75456035,
      "num_input_tokens_seen": 125171915,
      "step": 5816,
      "time_per_iteration": 2.6492061614990234
    },
    {
      "auxiliary_loss_clip": 0.01128843,
      "auxiliary_loss_mlp": 0.01024183,
      "balance_loss_clip": 1.0421592,
      "balance_loss_mlp": 1.01643991,
      "epoch": 0.6994528948475921,
      "flos": 16143714466560.0,
      "grad_norm": 2.209335058612384,
      "language_loss": 0.61344159,
      "learning_rate": 8.748927885194479e-07,
      "loss": 0.63497186,
      "num_input_tokens_seen": 125190220,
      "step": 5817,
      "time_per_iteration": 2.536794424057007
    },
    {
      "auxiliary_loss_clip": 0.01022299,
      "auxiliary_loss_mlp": 0.01001556,
      "balance_loss_clip": 1.01156878,
      "balance_loss_mlp": 1.00033462,
      "epoch": 0.6995731377382313,
      "flos": 64952420699520.0,
      "grad_norm": 0.800800386397483,
      "language_loss": 0.57423663,
      "learning_rate": 8.742488506428209e-07,
      "loss": 0.59447515,
      "num_input_tokens_seen": 125249310,
      "step": 5818,
      "time_per_iteration": 3.100766658782959
    },
    {
      "auxiliary_loss_clip": 0.01140886,
      "auxiliary_loss_mlp": 0.00760849,
      "balance_loss_clip": 1.04391372,
      "balance_loss_mlp": 1.00013852,
      "epoch": 0.6996933806288703,
      "flos": 24900136076160.0,
      "grad_norm": 2.318750953148414,
      "language_loss": 0.7815032,
      "learning_rate": 8.736050835417466e-07,
      "loss": 0.80052054,
      "num_input_tokens_seen": 125269350,
      "step": 5819,
      "time_per_iteration": 2.5597891807556152
    },
    {
      "auxiliary_loss_clip": 0.01158275,
      "auxiliary_loss_mlp": 0.01028124,
      "balance_loss_clip": 1.05068779,
      "balance_loss_mlp": 1.02080488,
      "epoch": 0.6998136235195094,
      "flos": 20777806782720.0,
      "grad_norm": 7.416729308829997,
      "language_loss": 0.61136651,
      "learning_rate": 8.729614873138862e-07,
      "loss": 0.63323045,
      "num_input_tokens_seen": 125286985,
      "step": 5820,
      "time_per_iteration": 2.4939725399017334
    },
    {
      "auxiliary_loss_clip": 0.01117358,
      "auxiliary_loss_mlp": 0.01026722,
      "balance_loss_clip": 1.04673553,
      "balance_loss_mlp": 1.01908374,
      "epoch": 0.6999338664101485,
      "flos": 23733470332800.0,
      "grad_norm": 1.9889710053702145,
      "language_loss": 0.77558291,
      "learning_rate": 8.723180620568716e-07,
      "loss": 0.79702377,
      "num_input_tokens_seen": 125306240,
      "step": 5821,
      "time_per_iteration": 2.5891356468200684
    },
    {
      "auxiliary_loss_clip": 0.01140057,
      "auxiliary_loss_mlp": 0.01027717,
      "balance_loss_clip": 1.04658163,
      "balance_loss_mlp": 1.02085042,
      "epoch": 0.7000541093007876,
      "flos": 19864598382720.0,
      "grad_norm": 1.9410457932721552,
      "language_loss": 0.85160106,
      "learning_rate": 8.716748078683116e-07,
      "loss": 0.8732788,
      "num_input_tokens_seen": 125323015,
      "step": 5822,
      "time_per_iteration": 3.299949884414673
    },
    {
      "auxiliary_loss_clip": 0.01073295,
      "auxiliary_loss_mlp": 0.01029801,
      "balance_loss_clip": 1.03860855,
      "balance_loss_mlp": 1.0219537,
      "epoch": 0.7001743521914267,
      "flos": 29679056029440.0,
      "grad_norm": 2.0103081119463493,
      "language_loss": 0.68484998,
      "learning_rate": 8.710317248457855e-07,
      "loss": 0.70588088,
      "num_input_tokens_seen": 125342630,
      "step": 5823,
      "time_per_iteration": 2.7237963676452637
    },
    {
      "auxiliary_loss_clip": 0.0113444,
      "auxiliary_loss_mlp": 0.01028602,
      "balance_loss_clip": 1.0484606,
      "balance_loss_mlp": 1.02102923,
      "epoch": 0.7002945950820658,
      "flos": 27489762080640.0,
      "grad_norm": 1.7158994952690103,
      "language_loss": 0.72314167,
      "learning_rate": 8.703888130868482e-07,
      "loss": 0.74477214,
      "num_input_tokens_seen": 125364480,
      "step": 5824,
      "time_per_iteration": 2.824151039123535
    },
    {
      "auxiliary_loss_clip": 0.0112507,
      "auxiliary_loss_mlp": 0.01030219,
      "balance_loss_clip": 1.04604101,
      "balance_loss_mlp": 1.02290559,
      "epoch": 0.7004148379727049,
      "flos": 22158463800960.0,
      "grad_norm": 10.389204859002279,
      "language_loss": 0.82144964,
      "learning_rate": 8.697460726890307e-07,
      "loss": 0.84300256,
      "num_input_tokens_seen": 125381625,
      "step": 5825,
      "time_per_iteration": 2.5360069274902344
    },
    {
      "auxiliary_loss_clip": 0.01121246,
      "auxiliary_loss_mlp": 0.00760797,
      "balance_loss_clip": 1.04128885,
      "balance_loss_mlp": 1.00016761,
      "epoch": 0.7005350808633439,
      "flos": 19423758764160.0,
      "grad_norm": 2.015459502124115,
      "language_loss": 0.90569198,
      "learning_rate": 8.691035037498354e-07,
      "loss": 0.92451239,
      "num_input_tokens_seen": 125397615,
      "step": 5826,
      "time_per_iteration": 2.5637917518615723
    },
    {
      "auxiliary_loss_clip": 0.01135136,
      "auxiliary_loss_mlp": 0.01027709,
      "balance_loss_clip": 1.04379833,
      "balance_loss_mlp": 1.02063417,
      "epoch": 0.7006553237539831,
      "flos": 23476708938240.0,
      "grad_norm": 1.6077474272061392,
      "language_loss": 0.72206962,
      "learning_rate": 8.684611063667391e-07,
      "loss": 0.74369806,
      "num_input_tokens_seen": 125418080,
      "step": 5827,
      "time_per_iteration": 2.5230302810668945
    },
    {
      "auxiliary_loss_clip": 0.01150448,
      "auxiliary_loss_mlp": 0.01025443,
      "balance_loss_clip": 1.04536366,
      "balance_loss_mlp": 1.01829934,
      "epoch": 0.7007755666446221,
      "flos": 31212872640000.0,
      "grad_norm": 4.347869231692518,
      "language_loss": 0.76887155,
      "learning_rate": 8.678188806371935e-07,
      "loss": 0.7906304,
      "num_input_tokens_seen": 125440115,
      "step": 5828,
      "time_per_iteration": 2.5747458934783936
    },
    {
      "auxiliary_loss_clip": 0.0115092,
      "auxiliary_loss_mlp": 0.01030015,
      "balance_loss_clip": 1.04647946,
      "balance_loss_mlp": 1.02363169,
      "epoch": 0.7008958095352612,
      "flos": 18149899858560.0,
      "grad_norm": 1.7303015009828693,
      "language_loss": 0.85509479,
      "learning_rate": 8.671768266586228e-07,
      "loss": 0.87690413,
      "num_input_tokens_seen": 125458240,
      "step": 5829,
      "time_per_iteration": 2.4542412757873535
    },
    {
      "auxiliary_loss_clip": 0.01122875,
      "auxiliary_loss_mlp": 0.01027682,
      "balance_loss_clip": 1.04507101,
      "balance_loss_mlp": 1.02032995,
      "epoch": 0.7010160524259004,
      "flos": 27452307173760.0,
      "grad_norm": 2.1809614796688837,
      "language_loss": 0.77901095,
      "learning_rate": 8.665349445284275e-07,
      "loss": 0.80051649,
      "num_input_tokens_seen": 125477980,
      "step": 5830,
      "time_per_iteration": 2.6131842136383057
    },
    {
      "auxiliary_loss_clip": 0.01124221,
      "auxiliary_loss_mlp": 0.0102378,
      "balance_loss_clip": 1.0454011,
      "balance_loss_mlp": 1.01685703,
      "epoch": 0.7011362953165394,
      "flos": 23842064125440.0,
      "grad_norm": 1.5914114058583368,
      "language_loss": 0.80896455,
      "learning_rate": 8.658932343439799e-07,
      "loss": 0.83044457,
      "num_input_tokens_seen": 125497765,
      "step": 5831,
      "time_per_iteration": 3.3624064922332764
    },
    {
      "auxiliary_loss_clip": 0.01168234,
      "auxiliary_loss_mlp": 0.01025367,
      "balance_loss_clip": 1.05207503,
      "balance_loss_mlp": 1.01828051,
      "epoch": 0.7012565382071785,
      "flos": 24823430582400.0,
      "grad_norm": 2.7038116477894283,
      "language_loss": 0.77791321,
      "learning_rate": 8.65251696202627e-07,
      "loss": 0.79984921,
      "num_input_tokens_seen": 125514145,
      "step": 5832,
      "time_per_iteration": 2.467630624771118
    },
    {
      "auxiliary_loss_clip": 0.01123632,
      "auxiliary_loss_mlp": 0.01029424,
      "balance_loss_clip": 1.04619431,
      "balance_loss_mlp": 1.02175915,
      "epoch": 0.7013767810978175,
      "flos": 21397445326080.0,
      "grad_norm": 2.05243756724098,
      "language_loss": 0.87693346,
      "learning_rate": 8.646103302016896e-07,
      "loss": 0.89846408,
      "num_input_tokens_seen": 125533115,
      "step": 5833,
      "time_per_iteration": 4.374024152755737
    },
    {
      "auxiliary_loss_clip": 0.01120336,
      "auxiliary_loss_mlp": 0.0102603,
      "balance_loss_clip": 1.04256976,
      "balance_loss_mlp": 1.01791215,
      "epoch": 0.7014970239884567,
      "flos": 16687150306560.0,
      "grad_norm": 1.6737092885690472,
      "language_loss": 0.88233256,
      "learning_rate": 8.639691364384614e-07,
      "loss": 0.9037962,
      "num_input_tokens_seen": 125550740,
      "step": 5834,
      "time_per_iteration": 2.577583074569702
    },
    {
      "auxiliary_loss_clip": 0.01144367,
      "auxiliary_loss_mlp": 0.01024903,
      "balance_loss_clip": 1.04934692,
      "balance_loss_mlp": 1.01707411,
      "epoch": 0.7016172668790958,
      "flos": 12568268718720.0,
      "grad_norm": 1.7329353642345469,
      "language_loss": 0.72947764,
      "learning_rate": 8.633281150102136e-07,
      "loss": 0.75117028,
      "num_input_tokens_seen": 125567590,
      "step": 5835,
      "time_per_iteration": 2.4972727298736572
    },
    {
      "auxiliary_loss_clip": 0.01139762,
      "auxiliary_loss_mlp": 0.01029328,
      "balance_loss_clip": 1.04738498,
      "balance_loss_mlp": 1.02206516,
      "epoch": 0.7017375097697348,
      "flos": 17452729808640.0,
      "grad_norm": 2.2297452069803056,
      "language_loss": 0.67720175,
      "learning_rate": 8.626872660141855e-07,
      "loss": 0.69889259,
      "num_input_tokens_seen": 125585500,
      "step": 5836,
      "time_per_iteration": 2.480466842651367
    },
    {
      "auxiliary_loss_clip": 0.01114482,
      "auxiliary_loss_mlp": 0.01030514,
      "balance_loss_clip": 1.04612553,
      "balance_loss_mlp": 1.02345085,
      "epoch": 0.701857752660374,
      "flos": 18513028402560.0,
      "grad_norm": 1.805058169306628,
      "language_loss": 0.74666572,
      "learning_rate": 8.620465895475957e-07,
      "loss": 0.76811576,
      "num_input_tokens_seen": 125603720,
      "step": 5837,
      "time_per_iteration": 2.5641074180603027
    },
    {
      "auxiliary_loss_clip": 0.01106556,
      "auxiliary_loss_mlp": 0.01026182,
      "balance_loss_clip": 1.04415405,
      "balance_loss_mlp": 1.0195179,
      "epoch": 0.701977995551013,
      "flos": 24425971614720.0,
      "grad_norm": 1.4771786835200609,
      "language_loss": 0.74928373,
      "learning_rate": 8.614060857076333e-07,
      "loss": 0.77061111,
      "num_input_tokens_seen": 125624390,
      "step": 5838,
      "time_per_iteration": 2.6109323501586914
    },
    {
      "auxiliary_loss_clip": 0.01136844,
      "auxiliary_loss_mlp": 0.01035368,
      "balance_loss_clip": 1.04506087,
      "balance_loss_mlp": 1.02828693,
      "epoch": 0.7020982384416521,
      "flos": 23002759958400.0,
      "grad_norm": 1.782012910396407,
      "language_loss": 0.74713439,
      "learning_rate": 8.60765754591462e-07,
      "loss": 0.76885653,
      "num_input_tokens_seen": 125644085,
      "step": 5839,
      "time_per_iteration": 2.534482479095459
    },
    {
      "auxiliary_loss_clip": 0.01164315,
      "auxiliary_loss_mlp": 0.0102657,
      "balance_loss_clip": 1.04736912,
      "balance_loss_mlp": 1.01944113,
      "epoch": 0.7022184813322913,
      "flos": 20449080489600.0,
      "grad_norm": 1.7553078608548847,
      "language_loss": 0.72872043,
      "learning_rate": 8.601255962962211e-07,
      "loss": 0.75062931,
      "num_input_tokens_seen": 125663095,
      "step": 5840,
      "time_per_iteration": 2.4771130084991455
    },
    {
      "auxiliary_loss_clip": 0.01161641,
      "auxiliary_loss_mlp": 0.01032266,
      "balance_loss_clip": 1.05065227,
      "balance_loss_mlp": 1.02395749,
      "epoch": 0.7023387242229303,
      "flos": 19790514581760.0,
      "grad_norm": 2.361359719373076,
      "language_loss": 0.71627361,
      "learning_rate": 8.594856109190194e-07,
      "loss": 0.7382127,
      "num_input_tokens_seen": 125680125,
      "step": 5841,
      "time_per_iteration": 2.4574127197265625
    },
    {
      "auxiliary_loss_clip": 0.01166325,
      "auxiliary_loss_mlp": 0.01022315,
      "balance_loss_clip": 1.04860735,
      "balance_loss_mlp": 1.01502562,
      "epoch": 0.7024589671135694,
      "flos": 33259278286080.0,
      "grad_norm": 1.5633934692384597,
      "language_loss": 0.69064808,
      "learning_rate": 8.588457985569446e-07,
      "loss": 0.71253455,
      "num_input_tokens_seen": 125703035,
      "step": 5842,
      "time_per_iteration": 2.599175214767456
    },
    {
      "auxiliary_loss_clip": 0.01168815,
      "auxiliary_loss_mlp": 0.01028384,
      "balance_loss_clip": 1.05070829,
      "balance_loss_mlp": 1.02090371,
      "epoch": 0.7025792100042085,
      "flos": 19098982967040.0,
      "grad_norm": 6.7565518560093745,
      "language_loss": 0.71670163,
      "learning_rate": 8.582061593070542e-07,
      "loss": 0.73867363,
      "num_input_tokens_seen": 125723765,
      "step": 5843,
      "time_per_iteration": 2.4649415016174316
    },
    {
      "auxiliary_loss_clip": 0.01168946,
      "auxiliary_loss_mlp": 0.00761028,
      "balance_loss_clip": 1.0499804,
      "balance_loss_mlp": 1.00020158,
      "epoch": 0.7026994528948476,
      "flos": 18952611045120.0,
      "grad_norm": 2.069856854353654,
      "language_loss": 0.76855522,
      "learning_rate": 8.57566693266383e-07,
      "loss": 0.78785491,
      "num_input_tokens_seen": 125741455,
      "step": 5844,
      "time_per_iteration": 2.464463472366333
    },
    {
      "auxiliary_loss_clip": 0.01142302,
      "auxiliary_loss_mlp": 0.0076084,
      "balance_loss_clip": 1.04408979,
      "balance_loss_mlp": 1.00016999,
      "epoch": 0.7028196957854866,
      "flos": 19536662188800.0,
      "grad_norm": 2.2888845349366074,
      "language_loss": 0.6908474,
      "learning_rate": 8.569274005319354e-07,
      "loss": 0.7098788,
      "num_input_tokens_seen": 125759855,
      "step": 5845,
      "time_per_iteration": 2.528874635696411
    },
    {
      "auxiliary_loss_clip": 0.0114973,
      "auxiliary_loss_mlp": 0.01028907,
      "balance_loss_clip": 1.04759622,
      "balance_loss_mlp": 1.02206099,
      "epoch": 0.7029399386761258,
      "flos": 20845318394880.0,
      "grad_norm": 1.6928417620050253,
      "language_loss": 0.79827845,
      "learning_rate": 8.562882812006913e-07,
      "loss": 0.82006478,
      "num_input_tokens_seen": 125777345,
      "step": 5846,
      "time_per_iteration": 2.4733903408050537
    },
    {
      "auxiliary_loss_clip": 0.01165792,
      "auxiliary_loss_mlp": 0.01033097,
      "balance_loss_clip": 1.04988468,
      "balance_loss_mlp": 1.02576888,
      "epoch": 0.7030601815667649,
      "flos": 22055005653120.0,
      "grad_norm": 2.008758616515949,
      "language_loss": 0.77360368,
      "learning_rate": 8.556493353696066e-07,
      "loss": 0.79559261,
      "num_input_tokens_seen": 125796345,
      "step": 5847,
      "time_per_iteration": 3.288814067840576
    },
    {
      "auxiliary_loss_clip": 0.01158155,
      "auxiliary_loss_mlp": 0.00761201,
      "balance_loss_clip": 1.04999495,
      "balance_loss_mlp": 1.00017869,
      "epoch": 0.7031804244574039,
      "flos": 27198742089600.0,
      "grad_norm": 2.3253242169129553,
      "language_loss": 0.67533851,
      "learning_rate": 8.550105631356077e-07,
      "loss": 0.69453204,
      "num_input_tokens_seen": 125816070,
      "step": 5848,
      "time_per_iteration": 2.54215669631958
    },
    {
      "auxiliary_loss_clip": 0.01119916,
      "auxiliary_loss_mlp": 0.01023138,
      "balance_loss_clip": 1.04292631,
      "balance_loss_mlp": 1.01610148,
      "epoch": 0.7033006673480431,
      "flos": 22379853277440.0,
      "grad_norm": 2.1438639449993637,
      "language_loss": 0.77168387,
      "learning_rate": 8.543719645955961e-07,
      "loss": 0.79311436,
      "num_input_tokens_seen": 125834400,
      "step": 5849,
      "time_per_iteration": 2.5561394691467285
    },
    {
      "auxiliary_loss_clip": 0.01139824,
      "auxiliary_loss_mlp": 0.01026472,
      "balance_loss_clip": 1.0449003,
      "balance_loss_mlp": 1.01942372,
      "epoch": 0.7034209102386821,
      "flos": 24715986024960.0,
      "grad_norm": 1.5726345943522109,
      "language_loss": 0.74317092,
      "learning_rate": 8.537335398464467e-07,
      "loss": 0.76483387,
      "num_input_tokens_seen": 125854720,
      "step": 5850,
      "time_per_iteration": 2.551536798477173
    },
    {
      "auxiliary_loss_clip": 0.01136793,
      "auxiliary_loss_mlp": 0.01028489,
      "balance_loss_clip": 1.04284787,
      "balance_loss_mlp": 1.02099681,
      "epoch": 0.7035411531293212,
      "flos": 22556174163840.0,
      "grad_norm": 2.5988442581358258,
      "language_loss": 0.85414803,
      "learning_rate": 8.53095288985007e-07,
      "loss": 0.87580085,
      "num_input_tokens_seen": 125868455,
      "step": 5851,
      "time_per_iteration": 2.4996259212493896
    },
    {
      "auxiliary_loss_clip": 0.01166573,
      "auxiliary_loss_mlp": 0.01023143,
      "balance_loss_clip": 1.05114532,
      "balance_loss_mlp": 1.01661921,
      "epoch": 0.7036613960199604,
      "flos": 22674967418880.0,
      "grad_norm": 1.6379293516023261,
      "language_loss": 0.82240051,
      "learning_rate": 8.524572121081009e-07,
      "loss": 0.84429771,
      "num_input_tokens_seen": 125888555,
      "step": 5852,
      "time_per_iteration": 2.465022087097168
    },
    {
      "auxiliary_loss_clip": 0.01155864,
      "auxiliary_loss_mlp": 0.01035076,
      "balance_loss_clip": 1.04683709,
      "balance_loss_mlp": 1.02757215,
      "epoch": 0.7037816389105994,
      "flos": 22492146170880.0,
      "grad_norm": 2.3772022839489404,
      "language_loss": 0.62222338,
      "learning_rate": 8.518193093125232e-07,
      "loss": 0.64413273,
      "num_input_tokens_seen": 125907610,
      "step": 5853,
      "time_per_iteration": 2.501202344894409
    },
    {
      "auxiliary_loss_clip": 0.01144477,
      "auxiliary_loss_mlp": 0.01027456,
      "balance_loss_clip": 1.04724956,
      "balance_loss_mlp": 1.02070284,
      "epoch": 0.7039018818012385,
      "flos": 27087490690560.0,
      "grad_norm": 1.643712436434686,
      "language_loss": 0.80690074,
      "learning_rate": 8.511815806950436e-07,
      "loss": 0.8286202,
      "num_input_tokens_seen": 125928640,
      "step": 5854,
      "time_per_iteration": 2.5687015056610107
    },
    {
      "auxiliary_loss_clip": 0.01148179,
      "auxiliary_loss_mlp": 0.01026715,
      "balance_loss_clip": 1.0444175,
      "balance_loss_mlp": 1.01989675,
      "epoch": 0.7040221246918776,
      "flos": 17749819198080.0,
      "grad_norm": 1.8883968287257795,
      "language_loss": 0.77893436,
      "learning_rate": 8.505440263524044e-07,
      "loss": 0.80068332,
      "num_input_tokens_seen": 125947485,
      "step": 5855,
      "time_per_iteration": 2.4825944900512695
    },
    {
      "auxiliary_loss_clip": 0.01154425,
      "auxiliary_loss_mlp": 0.01031728,
      "balance_loss_clip": 1.04669535,
      "balance_loss_mlp": 1.02342534,
      "epoch": 0.7041423675825167,
      "flos": 16279851012480.0,
      "grad_norm": 2.8303404423686622,
      "language_loss": 0.87788528,
      "learning_rate": 8.49906646381322e-07,
      "loss": 0.89974678,
      "num_input_tokens_seen": 125960320,
      "step": 5856,
      "time_per_iteration": 3.223134756088257
    },
    {
      "auxiliary_loss_clip": 0.01125785,
      "auxiliary_loss_mlp": 0.01027957,
      "balance_loss_clip": 1.04697669,
      "balance_loss_mlp": 1.02110219,
      "epoch": 0.7042626104731557,
      "flos": 25483181639040.0,
      "grad_norm": 1.7500516844237126,
      "language_loss": 0.71863431,
      "learning_rate": 8.492694408784884e-07,
      "loss": 0.74017167,
      "num_input_tokens_seen": 125980575,
      "step": 5857,
      "time_per_iteration": 2.5636167526245117
    },
    {
      "auxiliary_loss_clip": 0.01158313,
      "auxiliary_loss_mlp": 0.01023894,
      "balance_loss_clip": 1.04999137,
      "balance_loss_mlp": 1.01675928,
      "epoch": 0.7043828533637949,
      "flos": 17857622891520.0,
      "grad_norm": 2.643907696542298,
      "language_loss": 0.62652957,
      "learning_rate": 8.486324099405642e-07,
      "loss": 0.64835167,
      "num_input_tokens_seen": 125997420,
      "step": 5858,
      "time_per_iteration": 2.4577834606170654
    },
    {
      "auxiliary_loss_clip": 0.01152786,
      "auxiliary_loss_mlp": 0.01029047,
      "balance_loss_clip": 1.04696536,
      "balance_loss_mlp": 1.02206147,
      "epoch": 0.704503096254434,
      "flos": 29494259533440.0,
      "grad_norm": 1.6086773927101337,
      "language_loss": 0.74761856,
      "learning_rate": 8.479955536641887e-07,
      "loss": 0.76943684,
      "num_input_tokens_seen": 126018915,
      "step": 5859,
      "time_per_iteration": 4.106623888015747
    },
    {
      "auxiliary_loss_clip": 0.01130372,
      "auxiliary_loss_mlp": 0.01029775,
      "balance_loss_clip": 1.04088163,
      "balance_loss_mlp": 1.02271461,
      "epoch": 0.704623339145073,
      "flos": 30920739327360.0,
      "grad_norm": 3.3525152385614043,
      "language_loss": 0.66147757,
      "learning_rate": 8.473588721459716e-07,
      "loss": 0.683079,
      "num_input_tokens_seen": 126038825,
      "step": 5860,
      "time_per_iteration": 2.5856990814208984
    },
    {
      "auxiliary_loss_clip": 0.01157522,
      "auxiliary_loss_mlp": 0.01031802,
      "balance_loss_clip": 1.05072761,
      "balance_loss_mlp": 1.02351677,
      "epoch": 0.7047435820357122,
      "flos": 23914747296000.0,
      "grad_norm": 3.0905451455387887,
      "language_loss": 0.70779622,
      "learning_rate": 8.467223654824967e-07,
      "loss": 0.72968948,
      "num_input_tokens_seen": 126058280,
      "step": 5861,
      "time_per_iteration": 2.5104033946990967
    },
    {
      "auxiliary_loss_clip": 0.0114849,
      "auxiliary_loss_mlp": 0.01029175,
      "balance_loss_clip": 1.04786158,
      "balance_loss_mlp": 1.02195072,
      "epoch": 0.7048638249263512,
      "flos": 46494010926720.0,
      "grad_norm": 1.815347003049604,
      "language_loss": 0.62156832,
      "learning_rate": 8.460860337703233e-07,
      "loss": 0.64334488,
      "num_input_tokens_seen": 126078885,
      "step": 5862,
      "time_per_iteration": 2.75278639793396
    },
    {
      "auxiliary_loss_clip": 0.01112428,
      "auxiliary_loss_mlp": 0.01027618,
      "balance_loss_clip": 1.04225755,
      "balance_loss_mlp": 1.01964319,
      "epoch": 0.7049840678169903,
      "flos": 21689219502720.0,
      "grad_norm": 2.1291851969369704,
      "language_loss": 0.70745528,
      "learning_rate": 8.454498771059797e-07,
      "loss": 0.72885573,
      "num_input_tokens_seen": 126098260,
      "step": 5863,
      "time_per_iteration": 2.6473634243011475
    },
    {
      "auxiliary_loss_clip": 0.01104552,
      "auxiliary_loss_mlp": 0.01032503,
      "balance_loss_clip": 1.04328847,
      "balance_loss_mlp": 1.0252161,
      "epoch": 0.7051043107076294,
      "flos": 18405081054720.0,
      "grad_norm": 2.2936549333823515,
      "language_loss": 0.83522642,
      "learning_rate": 8.448138955859725e-07,
      "loss": 0.85659707,
      "num_input_tokens_seen": 126114845,
      "step": 5864,
      "time_per_iteration": 2.664006471633911
    },
    {
      "auxiliary_loss_clip": 0.01139335,
      "auxiliary_loss_mlp": 0.01027594,
      "balance_loss_clip": 1.04485941,
      "balance_loss_mlp": 1.01991653,
      "epoch": 0.7052245535982685,
      "flos": 19319043640320.0,
      "grad_norm": 1.8896071052256271,
      "language_loss": 0.89798778,
      "learning_rate": 8.44178089306778e-07,
      "loss": 0.91965705,
      "num_input_tokens_seen": 126132780,
      "step": 5865,
      "time_per_iteration": 2.59967041015625
    },
    {
      "auxiliary_loss_clip": 0.01166187,
      "auxiliary_loss_mlp": 0.01026144,
      "balance_loss_clip": 1.04959488,
      "balance_loss_mlp": 1.0196712,
      "epoch": 0.7053447964889076,
      "flos": 19062138591360.0,
      "grad_norm": 1.803492143934694,
      "language_loss": 0.77018666,
      "learning_rate": 8.4354245836485e-07,
      "loss": 0.79210991,
      "num_input_tokens_seen": 126151225,
      "step": 5866,
      "time_per_iteration": 2.5396835803985596
    },
    {
      "auxiliary_loss_clip": 0.01127413,
      "auxiliary_loss_mlp": 0.0102902,
      "balance_loss_clip": 1.04627538,
      "balance_loss_mlp": 1.02160549,
      "epoch": 0.7054650393795466,
      "flos": 27379228953600.0,
      "grad_norm": 1.564754910459291,
      "language_loss": 0.72827452,
      "learning_rate": 8.429070028566108e-07,
      "loss": 0.74983883,
      "num_input_tokens_seen": 126172535,
      "step": 5867,
      "time_per_iteration": 2.6511738300323486
    },
    {
      "auxiliary_loss_clip": 0.01150754,
      "auxiliary_loss_mlp": 0.01028313,
      "balance_loss_clip": 1.04626513,
      "balance_loss_mlp": 1.0210973,
      "epoch": 0.7055852822701858,
      "flos": 16102201322880.0,
      "grad_norm": 1.9212349373759396,
      "language_loss": 0.74893123,
      "learning_rate": 8.422717228784586e-07,
      "loss": 0.77072191,
      "num_input_tokens_seen": 126189410,
      "step": 5868,
      "time_per_iteration": 2.5666069984436035
    },
    {
      "auxiliary_loss_clip": 0.01110105,
      "auxiliary_loss_mlp": 0.0102984,
      "balance_loss_clip": 1.04621458,
      "balance_loss_mlp": 1.02295864,
      "epoch": 0.7057055251608249,
      "flos": 11692299744000.0,
      "grad_norm": 1.9270579328052857,
      "language_loss": 0.69024694,
      "learning_rate": 8.416366185267663e-07,
      "loss": 0.71164644,
      "num_input_tokens_seen": 126206910,
      "step": 5869,
      "time_per_iteration": 2.6389591693878174
    },
    {
      "auxiliary_loss_clip": 0.01150769,
      "auxiliary_loss_mlp": 0.010239,
      "balance_loss_clip": 1.04458475,
      "balance_loss_mlp": 1.01701581,
      "epoch": 0.7058257680514639,
      "flos": 22711560399360.0,
      "grad_norm": 1.6130552783751424,
      "language_loss": 0.77723861,
      "learning_rate": 8.410016898978778e-07,
      "loss": 0.79898524,
      "num_input_tokens_seen": 126224385,
      "step": 5870,
      "time_per_iteration": 2.477135181427002
    },
    {
      "auxiliary_loss_clip": 0.01105404,
      "auxiliary_loss_mlp": 0.01025944,
      "balance_loss_clip": 1.0442102,
      "balance_loss_mlp": 1.01938462,
      "epoch": 0.7059460109421031,
      "flos": 17529543043200.0,
      "grad_norm": 2.0077918683130234,
      "language_loss": 0.78756577,
      "learning_rate": 8.403669370881115e-07,
      "loss": 0.80887926,
      "num_input_tokens_seen": 126243120,
      "step": 5871,
      "time_per_iteration": 2.633226156234741
    },
    {
      "auxiliary_loss_clip": 0.01167069,
      "auxiliary_loss_mlp": 0.01027259,
      "balance_loss_clip": 1.05010128,
      "balance_loss_mlp": 1.02012467,
      "epoch": 0.7060662538327421,
      "flos": 23544687427200.0,
      "grad_norm": 3.6625208808778695,
      "language_loss": 0.7836284,
      "learning_rate": 8.397323601937587e-07,
      "loss": 0.80557168,
      "num_input_tokens_seen": 126263020,
      "step": 5872,
      "time_per_iteration": 2.4632811546325684
    },
    {
      "auxiliary_loss_clip": 0.01122548,
      "auxiliary_loss_mlp": 0.01031255,
      "balance_loss_clip": 1.04441595,
      "balance_loss_mlp": 1.02435589,
      "epoch": 0.7061864967233812,
      "flos": 30260736875520.0,
      "grad_norm": 1.825884418619177,
      "language_loss": 0.77134728,
      "learning_rate": 8.390979593110838e-07,
      "loss": 0.7928853,
      "num_input_tokens_seen": 126285150,
      "step": 5873,
      "time_per_iteration": 2.6225132942199707
    },
    {
      "auxiliary_loss_clip": 0.01142449,
      "auxiliary_loss_mlp": 0.0102426,
      "balance_loss_clip": 1.04782617,
      "balance_loss_mlp": 1.01674366,
      "epoch": 0.7063067396140204,
      "flos": 20701460424960.0,
      "grad_norm": 1.586588707491305,
      "language_loss": 0.81548047,
      "learning_rate": 8.384637345363262e-07,
      "loss": 0.83714747,
      "num_input_tokens_seen": 126304340,
      "step": 5874,
      "time_per_iteration": 3.2509214878082275
    },
    {
      "auxiliary_loss_clip": 0.01130059,
      "auxiliary_loss_mlp": 0.01029183,
      "balance_loss_clip": 1.04152393,
      "balance_loss_mlp": 1.02175045,
      "epoch": 0.7064269825046594,
      "flos": 32266168081920.0,
      "grad_norm": 1.6568057335174635,
      "language_loss": 0.76717126,
      "learning_rate": 8.378296859656964e-07,
      "loss": 0.78876364,
      "num_input_tokens_seen": 126325495,
      "step": 5875,
      "time_per_iteration": 2.597907543182373
    },
    {
      "auxiliary_loss_clip": 0.01140407,
      "auxiliary_loss_mlp": 0.01024456,
      "balance_loss_clip": 1.04671383,
      "balance_loss_mlp": 1.01776838,
      "epoch": 0.7065472253952985,
      "flos": 30227124723840.0,
      "grad_norm": 2.224292461697898,
      "language_loss": 0.68294233,
      "learning_rate": 8.371958136953792e-07,
      "loss": 0.70459104,
      "num_input_tokens_seen": 126345525,
      "step": 5876,
      "time_per_iteration": 2.608757495880127
    },
    {
      "auxiliary_loss_clip": 0.0113325,
      "auxiliary_loss_mlp": 0.01028676,
      "balance_loss_clip": 1.04679561,
      "balance_loss_mlp": 1.02060556,
      "epoch": 0.7066674682859376,
      "flos": 16216720859520.0,
      "grad_norm": 2.6403830089911264,
      "language_loss": 0.66480088,
      "learning_rate": 8.365621178215326e-07,
      "loss": 0.6864202,
      "num_input_tokens_seen": 126361995,
      "step": 5877,
      "time_per_iteration": 2.525420904159546
    },
    {
      "auxiliary_loss_clip": 0.01145499,
      "auxiliary_loss_mlp": 0.01027356,
      "balance_loss_clip": 1.04472184,
      "balance_loss_mlp": 1.02000713,
      "epoch": 0.7067877111765767,
      "flos": 14830461319680.0,
      "grad_norm": 2.1651071930942702,
      "language_loss": 0.75574213,
      "learning_rate": 8.359285984402871e-07,
      "loss": 0.77747065,
      "num_input_tokens_seen": 126379260,
      "step": 5878,
      "time_per_iteration": 2.4760303497314453
    },
    {
      "auxiliary_loss_clip": 0.01134715,
      "auxiliary_loss_mlp": 0.01024874,
      "balance_loss_clip": 1.04662478,
      "balance_loss_mlp": 1.01799273,
      "epoch": 0.7069079540672157,
      "flos": 25440196037760.0,
      "grad_norm": 2.007078143413203,
      "language_loss": 0.74045736,
      "learning_rate": 8.352952556477489e-07,
      "loss": 0.76205325,
      "num_input_tokens_seen": 126397170,
      "step": 5879,
      "time_per_iteration": 2.5716190338134766
    },
    {
      "auxiliary_loss_clip": 0.01152091,
      "auxiliary_loss_mlp": 0.01027904,
      "balance_loss_clip": 1.04863739,
      "balance_loss_mlp": 1.02076387,
      "epoch": 0.7070281969578549,
      "flos": 24607751368320.0,
      "grad_norm": 1.6882053455512651,
      "language_loss": 0.76613593,
      "learning_rate": 8.34662089539993e-07,
      "loss": 0.78793585,
      "num_input_tokens_seen": 126416680,
      "step": 5880,
      "time_per_iteration": 2.523428201675415
    },
    {
      "auxiliary_loss_clip": 0.01165288,
      "auxiliary_loss_mlp": 0.01026084,
      "balance_loss_clip": 1.04916835,
      "balance_loss_mlp": 1.01946783,
      "epoch": 0.707148439848494,
      "flos": 26724469887360.0,
      "grad_norm": 1.8739469475091177,
      "language_loss": 0.79157043,
      "learning_rate": 8.340291002130722e-07,
      "loss": 0.81348419,
      "num_input_tokens_seen": 126435870,
      "step": 5881,
      "time_per_iteration": 2.5144615173339844
    },
    {
      "auxiliary_loss_clip": 0.01169134,
      "auxiliary_loss_mlp": 0.0102818,
      "balance_loss_clip": 1.05018497,
      "balance_loss_mlp": 1.02085495,
      "epoch": 0.707268682739133,
      "flos": 15085750256640.0,
      "grad_norm": 2.5417359573233664,
      "language_loss": 0.79184103,
      "learning_rate": 8.3339628776301e-07,
      "loss": 0.81381416,
      "num_input_tokens_seen": 126454010,
      "step": 5882,
      "time_per_iteration": 3.236354351043701
    },
    {
      "auxiliary_loss_clip": 0.01165931,
      "auxiliary_loss_mlp": 0.01023512,
      "balance_loss_clip": 1.05010927,
      "balance_loss_mlp": 1.01686597,
      "epoch": 0.7073889256297722,
      "flos": 34313148345600.0,
      "grad_norm": 1.7687816121496613,
      "language_loss": 0.57387298,
      "learning_rate": 8.327636522858033e-07,
      "loss": 0.59576744,
      "num_input_tokens_seen": 126473615,
      "step": 5883,
      "time_per_iteration": 2.5978281497955322
    },
    {
      "auxiliary_loss_clip": 0.01111164,
      "auxiliary_loss_mlp": 0.01027926,
      "balance_loss_clip": 1.04627419,
      "balance_loss_mlp": 1.02102673,
      "epoch": 0.7075091685204112,
      "flos": 20083940784000.0,
      "grad_norm": 1.9478741770793402,
      "language_loss": 0.76864421,
      "learning_rate": 8.321311938774225e-07,
      "loss": 0.79003513,
      "num_input_tokens_seen": 126492705,
      "step": 5884,
      "time_per_iteration": 3.3583285808563232
    },
    {
      "auxiliary_loss_clip": 0.01168751,
      "auxiliary_loss_mlp": 0.01029582,
      "balance_loss_clip": 1.04873252,
      "balance_loss_mlp": 1.02176762,
      "epoch": 0.7076294114110503,
      "flos": 20777124424320.0,
      "grad_norm": 1.9705229719053095,
      "language_loss": 0.79183394,
      "learning_rate": 8.314989126338104e-07,
      "loss": 0.81381726,
      "num_input_tokens_seen": 126512715,
      "step": 5885,
      "time_per_iteration": 3.249850273132324
    },
    {
      "auxiliary_loss_clip": 0.01155191,
      "auxiliary_loss_mlp": 0.01027284,
      "balance_loss_clip": 1.04792666,
      "balance_loss_mlp": 1.01994956,
      "epoch": 0.7077496543016895,
      "flos": 17967689141760.0,
      "grad_norm": 1.7257720075279035,
      "language_loss": 0.84774017,
      "learning_rate": 8.308668086508847e-07,
      "loss": 0.86956489,
      "num_input_tokens_seen": 126530795,
      "step": 5886,
      "time_per_iteration": 2.473177671432495
    },
    {
      "auxiliary_loss_clip": 0.01126009,
      "auxiliary_loss_mlp": 0.01030258,
      "balance_loss_clip": 1.04122162,
      "balance_loss_mlp": 1.02250338,
      "epoch": 0.7078698971923285,
      "flos": 45478098564480.0,
      "grad_norm": 1.8752824600779856,
      "language_loss": 0.73798913,
      "learning_rate": 8.302348820245342e-07,
      "loss": 0.75955176,
      "num_input_tokens_seen": 126553360,
      "step": 5887,
      "time_per_iteration": 2.7934060096740723
    },
    {
      "auxiliary_loss_clip": 0.01120533,
      "auxiliary_loss_mlp": 0.01035,
      "balance_loss_clip": 1.04377353,
      "balance_loss_mlp": 1.02703106,
      "epoch": 0.7079901400829676,
      "flos": 26943704547840.0,
      "grad_norm": 4.232421285298609,
      "language_loss": 0.69523096,
      "learning_rate": 8.296031328506232e-07,
      "loss": 0.71678632,
      "num_input_tokens_seen": 126573110,
      "step": 5888,
      "time_per_iteration": 2.593430280685425
    },
    {
      "auxiliary_loss_clip": 0.01140416,
      "auxiliary_loss_mlp": 0.01030484,
      "balance_loss_clip": 1.04712737,
      "balance_loss_mlp": 1.02300096,
      "epoch": 0.7081103829736067,
      "flos": 24423206267520.0,
      "grad_norm": 1.7566459584076268,
      "language_loss": 0.75501245,
      "learning_rate": 8.289715612249857e-07,
      "loss": 0.77672148,
      "num_input_tokens_seen": 126593725,
      "step": 5889,
      "time_per_iteration": 2.573807954788208
    },
    {
      "auxiliary_loss_clip": 0.0113584,
      "auxiliary_loss_mlp": 0.01026536,
      "balance_loss_clip": 1.04551697,
      "balance_loss_mlp": 1.01886451,
      "epoch": 0.7082306258642458,
      "flos": 18543300589440.0,
      "grad_norm": 2.5673152059384448,
      "language_loss": 0.77500725,
      "learning_rate": 8.283401672434305e-07,
      "loss": 0.79663104,
      "num_input_tokens_seen": 126608950,
      "step": 5890,
      "time_per_iteration": 2.4797518253326416
    },
    {
      "auxiliary_loss_clip": 0.01135789,
      "auxiliary_loss_mlp": 0.01027615,
      "balance_loss_clip": 1.04676378,
      "balance_loss_mlp": 1.02087116,
      "epoch": 0.7083508687548848,
      "flos": 23477534951040.0,
      "grad_norm": 4.034952064423403,
      "language_loss": 0.70197719,
      "learning_rate": 8.277089510017412e-07,
      "loss": 0.72361124,
      "num_input_tokens_seen": 126629755,
      "step": 5891,
      "time_per_iteration": 2.557799816131592
    },
    {
      "auxiliary_loss_clip": 0.0113713,
      "auxiliary_loss_mlp": 0.0102521,
      "balance_loss_clip": 1.04795945,
      "balance_loss_mlp": 1.01832604,
      "epoch": 0.708471111645524,
      "flos": 22419463000320.0,
      "grad_norm": 1.7810892218867753,
      "language_loss": 0.81972742,
      "learning_rate": 8.270779125956719e-07,
      "loss": 0.84135079,
      "num_input_tokens_seen": 126650135,
      "step": 5892,
      "time_per_iteration": 2.522712469100952
    },
    {
      "auxiliary_loss_clip": 0.01103442,
      "auxiliary_loss_mlp": 0.01022167,
      "balance_loss_clip": 1.0419507,
      "balance_loss_mlp": 1.01548517,
      "epoch": 0.7085913545361631,
      "flos": 20922885815040.0,
      "grad_norm": 1.9280816115009511,
      "language_loss": 0.80005538,
      "learning_rate": 8.264470521209505e-07,
      "loss": 0.82131147,
      "num_input_tokens_seen": 126668500,
      "step": 5893,
      "time_per_iteration": 2.6111981868743896
    },
    {
      "auxiliary_loss_clip": 0.01143725,
      "auxiliary_loss_mlp": 0.0103226,
      "balance_loss_clip": 1.04499078,
      "balance_loss_mlp": 1.02443743,
      "epoch": 0.7087115974268021,
      "flos": 15012384727680.0,
      "grad_norm": 2.279620653419183,
      "language_loss": 0.76413298,
      "learning_rate": 8.258163696732785e-07,
      "loss": 0.78589284,
      "num_input_tokens_seen": 126686090,
      "step": 5894,
      "time_per_iteration": 2.471235990524292
    },
    {
      "auxiliary_loss_clip": 0.01149051,
      "auxiliary_loss_mlp": 0.0102864,
      "balance_loss_clip": 1.04768801,
      "balance_loss_mlp": 1.02136815,
      "epoch": 0.7088318403174413,
      "flos": 21539040739200.0,
      "grad_norm": 1.7940745003323173,
      "language_loss": 0.76772308,
      "learning_rate": 8.251858653483288e-07,
      "loss": 0.7895,
      "num_input_tokens_seen": 126704255,
      "step": 5895,
      "time_per_iteration": 2.5018742084503174
    },
    {
      "auxiliary_loss_clip": 0.01155287,
      "auxiliary_loss_mlp": 0.01030401,
      "balance_loss_clip": 1.05022836,
      "balance_loss_mlp": 1.02349854,
      "epoch": 0.7089520832080803,
      "flos": 15516785462400.0,
      "grad_norm": 3.7582560147478357,
      "language_loss": 0.85741544,
      "learning_rate": 8.245555392417501e-07,
      "loss": 0.87927222,
      "num_input_tokens_seen": 126718910,
      "step": 5896,
      "time_per_iteration": 2.4600350856781006
    },
    {
      "auxiliary_loss_clip": 0.01096327,
      "auxiliary_loss_mlp": 0.01031702,
      "balance_loss_clip": 1.0390799,
      "balance_loss_mlp": 1.02484775,
      "epoch": 0.7090723260987194,
      "flos": 20412667077120.0,
      "grad_norm": 1.6591802783538785,
      "language_loss": 0.78639156,
      "learning_rate": 8.239253914491613e-07,
      "loss": 0.80767184,
      "num_input_tokens_seen": 126737235,
      "step": 5897,
      "time_per_iteration": 2.6094937324523926
    },
    {
      "auxiliary_loss_clip": 0.01120189,
      "auxiliary_loss_mlp": 0.01028182,
      "balance_loss_clip": 1.04524326,
      "balance_loss_mlp": 1.02089548,
      "epoch": 0.7091925689893585,
      "flos": 25668337271040.0,
      "grad_norm": 1.9579554016668974,
      "language_loss": 0.75238913,
      "learning_rate": 8.232954220661556e-07,
      "loss": 0.77387285,
      "num_input_tokens_seen": 126759970,
      "step": 5898,
      "time_per_iteration": 2.6289472579956055
    },
    {
      "auxiliary_loss_clip": 0.01165934,
      "auxiliary_loss_mlp": 0.01026979,
      "balance_loss_clip": 1.05119538,
      "balance_loss_mlp": 1.01986575,
      "epoch": 0.7093128118799976,
      "flos": 24206629213440.0,
      "grad_norm": 2.4908415612928754,
      "language_loss": 0.70219332,
      "learning_rate": 8.226656311882989e-07,
      "loss": 0.72412246,
      "num_input_tokens_seen": 126779280,
      "step": 5899,
      "time_per_iteration": 2.488739490509033
    },
    {
      "auxiliary_loss_clip": 0.01152205,
      "auxiliary_loss_mlp": 0.01025088,
      "balance_loss_clip": 1.05134666,
      "balance_loss_mlp": 1.01845086,
      "epoch": 0.7094330547706367,
      "flos": 16646786398080.0,
      "grad_norm": 2.0266106489030857,
      "language_loss": 0.76892614,
      "learning_rate": 8.22036018911129e-07,
      "loss": 0.79069912,
      "num_input_tokens_seen": 126797310,
      "step": 5900,
      "time_per_iteration": 3.2956314086914062
    },
    {
      "auxiliary_loss_clip": 0.0116876,
      "auxiliary_loss_mlp": 0.01028709,
      "balance_loss_clip": 1.04858816,
      "balance_loss_mlp": 1.0211302,
      "epoch": 0.7095532976612757,
      "flos": 16283370545280.0,
      "grad_norm": 1.9254889812378222,
      "language_loss": 0.80742991,
      "learning_rate": 8.214065853301599e-07,
      "loss": 0.82940459,
      "num_input_tokens_seen": 126812840,
      "step": 5901,
      "time_per_iteration": 2.437049627304077
    },
    {
      "auxiliary_loss_clip": 0.01050315,
      "auxiliary_loss_mlp": 0.01001844,
      "balance_loss_clip": 1.01036167,
      "balance_loss_mlp": 1.00080061,
      "epoch": 0.7096735405519149,
      "flos": 70722080559360.0,
      "grad_norm": 0.810224270062796,
      "language_loss": 0.58287442,
      "learning_rate": 8.207773305408734e-07,
      "loss": 0.603396,
      "num_input_tokens_seen": 126880060,
      "step": 5902,
      "time_per_iteration": 3.2224206924438477
    },
    {
      "auxiliary_loss_clip": 0.01118983,
      "auxiliary_loss_mlp": 0.01030667,
      "balance_loss_clip": 1.04260814,
      "balance_loss_mlp": 1.02298975,
      "epoch": 0.709793783442554,
      "flos": 23621500661760.0,
      "grad_norm": 2.073064510696055,
      "language_loss": 0.79268181,
      "learning_rate": 8.201482546387288e-07,
      "loss": 0.81417823,
      "num_input_tokens_seen": 126899535,
      "step": 5903,
      "time_per_iteration": 2.622485876083374
    },
    {
      "auxiliary_loss_clip": 0.01152732,
      "auxiliary_loss_mlp": 0.01028256,
      "balance_loss_clip": 1.04889321,
      "balance_loss_mlp": 1.02157712,
      "epoch": 0.709914026333193,
      "flos": 25993472204160.0,
      "grad_norm": 1.7217694993614585,
      "language_loss": 0.92229736,
      "learning_rate": 8.195193577191553e-07,
      "loss": 0.94410729,
      "num_input_tokens_seen": 126921365,
      "step": 5904,
      "time_per_iteration": 2.557504177093506
    },
    {
      "auxiliary_loss_clip": 0.01141713,
      "auxiliary_loss_mlp": 0.00760709,
      "balance_loss_clip": 1.04554749,
      "balance_loss_mlp": 1.00022292,
      "epoch": 0.7100342692238322,
      "flos": 24861531934080.0,
      "grad_norm": 1.588097479331853,
      "language_loss": 0.84546775,
      "learning_rate": 8.188906398775579e-07,
      "loss": 0.86449194,
      "num_input_tokens_seen": 126941910,
      "step": 5905,
      "time_per_iteration": 2.5589025020599365
    },
    {
      "auxiliary_loss_clip": 0.0116653,
      "auxiliary_loss_mlp": 0.00761443,
      "balance_loss_clip": 1.04805875,
      "balance_loss_mlp": 1.0002284,
      "epoch": 0.7101545121144712,
      "flos": 24932203943040.0,
      "grad_norm": 1.9123917841672793,
      "language_loss": 0.68748665,
      "learning_rate": 8.18262101209311e-07,
      "loss": 0.70676637,
      "num_input_tokens_seen": 126961120,
      "step": 5906,
      "time_per_iteration": 2.535945177078247
    },
    {
      "auxiliary_loss_clip": 0.01151664,
      "auxiliary_loss_mlp": 0.0102945,
      "balance_loss_clip": 1.04690719,
      "balance_loss_mlp": 1.0222795,
      "epoch": 0.7102747550051103,
      "flos": 23768842250880.0,
      "grad_norm": 1.7374613624801922,
      "language_loss": 0.70201194,
      "learning_rate": 8.176337418097626e-07,
      "loss": 0.72382319,
      "num_input_tokens_seen": 126981590,
      "step": 5907,
      "time_per_iteration": 2.4988150596618652
    },
    {
      "auxiliary_loss_clip": 0.01153484,
      "auxiliary_loss_mlp": 0.00761152,
      "balance_loss_clip": 1.04916489,
      "balance_loss_mlp": 1.00019264,
      "epoch": 0.7103949978957494,
      "flos": 15303907509120.0,
      "grad_norm": 2.043163575821353,
      "language_loss": 0.79307842,
      "learning_rate": 8.170055617742364e-07,
      "loss": 0.81222486,
      "num_input_tokens_seen": 126998870,
      "step": 5908,
      "time_per_iteration": 3.2770326137542725
    },
    {
      "auxiliary_loss_clip": 0.01135218,
      "auxiliary_loss_mlp": 0.0103012,
      "balance_loss_clip": 1.04659891,
      "balance_loss_mlp": 1.02271724,
      "epoch": 0.7105152407863885,
      "flos": 22638805401600.0,
      "grad_norm": 1.8422044169648748,
      "language_loss": 0.70486486,
      "learning_rate": 8.163775611980252e-07,
      "loss": 0.72651827,
      "num_input_tokens_seen": 127017980,
      "step": 5909,
      "time_per_iteration": 2.550135374069214
    },
    {
      "auxiliary_loss_clip": 0.01141051,
      "auxiliary_loss_mlp": 0.01024339,
      "balance_loss_clip": 1.04752088,
      "balance_loss_mlp": 1.01705575,
      "epoch": 0.7106354836770276,
      "flos": 17238594879360.0,
      "grad_norm": 1.6568353374910332,
      "language_loss": 0.78700662,
      "learning_rate": 8.157497401763982e-07,
      "loss": 0.80866051,
      "num_input_tokens_seen": 127035645,
      "step": 5910,
      "time_per_iteration": 3.281853199005127
    },
    {
      "auxiliary_loss_clip": 0.01151433,
      "auxiliary_loss_mlp": 0.01029137,
      "balance_loss_clip": 1.0483067,
      "balance_loss_mlp": 1.02224684,
      "epoch": 0.7107557265676667,
      "flos": 20193647898240.0,
      "grad_norm": 1.9339660613300778,
      "language_loss": 0.77723074,
      "learning_rate": 8.151220988045935e-07,
      "loss": 0.79903644,
      "num_input_tokens_seen": 127054900,
      "step": 5911,
      "time_per_iteration": 3.2386882305145264
    },
    {
      "auxiliary_loss_clip": 0.01149101,
      "auxiliary_loss_mlp": 0.01026731,
      "balance_loss_clip": 1.04536915,
      "balance_loss_mlp": 1.01990056,
      "epoch": 0.7108759694583058,
      "flos": 21507080613120.0,
      "grad_norm": 1.7023936830519582,
      "language_loss": 0.82699609,
      "learning_rate": 8.144946371778234e-07,
      "loss": 0.84875441,
      "num_input_tokens_seen": 127075010,
      "step": 5912,
      "time_per_iteration": 2.495692253112793
    },
    {
      "auxiliary_loss_clip": 0.01136961,
      "auxiliary_loss_mlp": 0.00761111,
      "balance_loss_clip": 1.04496598,
      "balance_loss_mlp": 1.00021064,
      "epoch": 0.7109962123489448,
      "flos": 24061909317120.0,
      "grad_norm": 1.5881360834084037,
      "language_loss": 0.78092092,
      "learning_rate": 8.138673553912751e-07,
      "loss": 0.79990166,
      "num_input_tokens_seen": 127095570,
      "step": 5913,
      "time_per_iteration": 2.5643937587738037
    },
    {
      "auxiliary_loss_clip": 0.01108065,
      "auxiliary_loss_mlp": 0.01028318,
      "balance_loss_clip": 1.0439713,
      "balance_loss_mlp": 1.02111208,
      "epoch": 0.711116455239584,
      "flos": 30480474326400.0,
      "grad_norm": 2.4338272594919914,
      "language_loss": 0.56729263,
      "learning_rate": 8.132402535401059e-07,
      "loss": 0.58865643,
      "num_input_tokens_seen": 127116825,
      "step": 5914,
      "time_per_iteration": 2.657670021057129
    },
    {
      "auxiliary_loss_clip": 0.01146857,
      "auxiliary_loss_mlp": 0.01021624,
      "balance_loss_clip": 1.04690993,
      "balance_loss_mlp": 1.01503456,
      "epoch": 0.711236698130223,
      "flos": 25045610158080.0,
      "grad_norm": 1.732933656880314,
      "language_loss": 0.74288392,
      "learning_rate": 8.126133317194465e-07,
      "loss": 0.76456869,
      "num_input_tokens_seen": 127137015,
      "step": 5915,
      "time_per_iteration": 2.5464138984680176
    },
    {
      "auxiliary_loss_clip": 0.01098627,
      "auxiliary_loss_mlp": 0.01029175,
      "balance_loss_clip": 1.03845572,
      "balance_loss_mlp": 1.0221417,
      "epoch": 0.7113569410208621,
      "flos": 24206701040640.0,
      "grad_norm": 1.7189237815626226,
      "language_loss": 0.74213123,
      "learning_rate": 8.11986590024401e-07,
      "loss": 0.7634092,
      "num_input_tokens_seen": 127156755,
      "step": 5916,
      "time_per_iteration": 2.6436192989349365
    },
    {
      "auxiliary_loss_clip": 0.01137097,
      "auxiliary_loss_mlp": 0.01031449,
      "balance_loss_clip": 1.04483759,
      "balance_loss_mlp": 1.02417159,
      "epoch": 0.7114771839115013,
      "flos": 35439306526080.0,
      "grad_norm": 1.57880264053548,
      "language_loss": 0.68848467,
      "learning_rate": 8.113600285500442e-07,
      "loss": 0.71017015,
      "num_input_tokens_seen": 127176965,
      "step": 5917,
      "time_per_iteration": 2.6642587184906006
    },
    {
      "auxiliary_loss_clip": 0.01165992,
      "auxiliary_loss_mlp": 0.01023922,
      "balance_loss_clip": 1.04810941,
      "balance_loss_mlp": 1.01706731,
      "epoch": 0.7115974268021403,
      "flos": 21099458096640.0,
      "grad_norm": 1.7439420593869805,
      "language_loss": 0.7443608,
      "learning_rate": 8.107336473914268e-07,
      "loss": 0.76625991,
      "num_input_tokens_seen": 127195595,
      "step": 5918,
      "time_per_iteration": 2.473623514175415
    },
    {
      "auxiliary_loss_clip": 0.01038293,
      "auxiliary_loss_mlp": 0.01002865,
      "balance_loss_clip": 1.0107851,
      "balance_loss_mlp": 1.00183392,
      "epoch": 0.7117176696927794,
      "flos": 56752866616320.0,
      "grad_norm": 0.7638180467868849,
      "language_loss": 0.55711603,
      "learning_rate": 8.101074466435694e-07,
      "loss": 0.57752764,
      "num_input_tokens_seen": 127255070,
      "step": 5919,
      "time_per_iteration": 3.082484483718872
    },
    {
      "auxiliary_loss_clip": 0.01147301,
      "auxiliary_loss_mlp": 0.0103454,
      "balance_loss_clip": 1.04678619,
      "balance_loss_mlp": 1.02748871,
      "epoch": 0.7118379125834186,
      "flos": 15925269905280.0,
      "grad_norm": 1.6869747409131655,
      "language_loss": 0.67543042,
      "learning_rate": 8.094814264014662e-07,
      "loss": 0.69724888,
      "num_input_tokens_seen": 127273825,
      "step": 5920,
      "time_per_iteration": 2.516599178314209
    },
    {
      "auxiliary_loss_clip": 0.0116952,
      "auxiliary_loss_mlp": 0.01032581,
      "balance_loss_clip": 1.04948604,
      "balance_loss_mlp": 1.02481723,
      "epoch": 0.7119581554740576,
      "flos": 20193360589440.0,
      "grad_norm": 2.0154238168501584,
      "language_loss": 0.81293356,
      "learning_rate": 8.088555867600844e-07,
      "loss": 0.83495456,
      "num_input_tokens_seen": 127289990,
      "step": 5921,
      "time_per_iteration": 2.4605178833007812
    },
    {
      "auxiliary_loss_clip": 0.01120271,
      "auxiliary_loss_mlp": 0.01027936,
      "balance_loss_clip": 1.04357076,
      "balance_loss_mlp": 1.02110195,
      "epoch": 0.7120783983646967,
      "flos": 34715383822080.0,
      "grad_norm": 1.723939092615646,
      "language_loss": 0.60584635,
      "learning_rate": 8.08229927814362e-07,
      "loss": 0.62732846,
      "num_input_tokens_seen": 127312880,
      "step": 5922,
      "time_per_iteration": 2.7043328285217285
    },
    {
      "auxiliary_loss_clip": 0.0112516,
      "auxiliary_loss_mlp": 0.01026741,
      "balance_loss_clip": 1.04441893,
      "balance_loss_mlp": 1.01988363,
      "epoch": 0.7121986412553358,
      "flos": 26359114700160.0,
      "grad_norm": 1.6811045328903538,
      "language_loss": 0.64586228,
      "learning_rate": 8.076044496592134e-07,
      "loss": 0.66738129,
      "num_input_tokens_seen": 127334730,
      "step": 5923,
      "time_per_iteration": 2.6040077209472656
    },
    {
      "auxiliary_loss_clip": 0.01140287,
      "auxiliary_loss_mlp": 0.01027507,
      "balance_loss_clip": 1.04900861,
      "balance_loss_mlp": 1.02018738,
      "epoch": 0.7123188841459749,
      "flos": 11145344371200.0,
      "grad_norm": 2.3999773515420952,
      "language_loss": 0.77970093,
      "learning_rate": 8.069791523895204e-07,
      "loss": 0.80137885,
      "num_input_tokens_seen": 127351180,
      "step": 5924,
      "time_per_iteration": 2.5028090476989746
    },
    {
      "auxiliary_loss_clip": 0.01113194,
      "auxiliary_loss_mlp": 0.01029583,
      "balance_loss_clip": 1.03968239,
      "balance_loss_mlp": 1.02234983,
      "epoch": 0.7124391270366139,
      "flos": 20811670329600.0,
      "grad_norm": 1.7376676489009597,
      "language_loss": 0.77334929,
      "learning_rate": 8.063540361001422e-07,
      "loss": 0.7947771,
      "num_input_tokens_seen": 127369750,
      "step": 5925,
      "time_per_iteration": 3.2933363914489746
    },
    {
      "auxiliary_loss_clip": 0.01119282,
      "auxiliary_loss_mlp": 0.01033282,
      "balance_loss_clip": 1.04349422,
      "balance_loss_mlp": 1.02559614,
      "epoch": 0.7125593699272531,
      "flos": 17603734584960.0,
      "grad_norm": 2.063504550278858,
      "language_loss": 0.79362738,
      "learning_rate": 8.057291008859069e-07,
      "loss": 0.815153,
      "num_input_tokens_seen": 127387910,
      "step": 5926,
      "time_per_iteration": 2.5502047538757324
    },
    {
      "auxiliary_loss_clip": 0.01145814,
      "auxiliary_loss_mlp": 0.01027031,
      "balance_loss_clip": 1.04355788,
      "balance_loss_mlp": 1.02010453,
      "epoch": 0.7126796128178922,
      "flos": 28654057526400.0,
      "grad_norm": 1.8990206572128372,
      "language_loss": 0.68124866,
      "learning_rate": 8.051043468416187e-07,
      "loss": 0.70297718,
      "num_input_tokens_seen": 127409160,
      "step": 5927,
      "time_per_iteration": 2.5870728492736816
    },
    {
      "auxiliary_loss_clip": 0.0116586,
      "auxiliary_loss_mlp": 0.01028979,
      "balance_loss_clip": 1.04990995,
      "balance_loss_mlp": 1.02182972,
      "epoch": 0.7127998557085312,
      "flos": 16034438315520.0,
      "grad_norm": 1.9006594967857438,
      "language_loss": 0.82457882,
      "learning_rate": 8.044797740620506e-07,
      "loss": 0.84652722,
      "num_input_tokens_seen": 127427765,
      "step": 5928,
      "time_per_iteration": 2.4612295627593994
    },
    {
      "auxiliary_loss_clip": 0.01105955,
      "auxiliary_loss_mlp": 0.01028286,
      "balance_loss_clip": 1.0436573,
      "balance_loss_mlp": 1.02137184,
      "epoch": 0.7129200985991703,
      "flos": 23403271582080.0,
      "grad_norm": 1.9216664242861543,
      "language_loss": 0.78547251,
      "learning_rate": 8.038553826419494e-07,
      "loss": 0.80681497,
      "num_input_tokens_seen": 127446475,
      "step": 5929,
      "time_per_iteration": 2.6117544174194336
    },
    {
      "auxiliary_loss_clip": 0.01164476,
      "auxiliary_loss_mlp": 0.01029951,
      "balance_loss_clip": 1.04661107,
      "balance_loss_mlp": 1.0231626,
      "epoch": 0.7130403414898094,
      "flos": 21397445326080.0,
      "grad_norm": 1.6108016464088828,
      "language_loss": 0.81147122,
      "learning_rate": 8.032311726760364e-07,
      "loss": 0.83341545,
      "num_input_tokens_seen": 127467695,
      "step": 5930,
      "time_per_iteration": 2.4909331798553467
    },
    {
      "auxiliary_loss_clip": 0.01116278,
      "auxiliary_loss_mlp": 0.01028155,
      "balance_loss_clip": 1.0428648,
      "balance_loss_mlp": 1.02091324,
      "epoch": 0.7131605843804485,
      "flos": 74739045306240.0,
      "grad_norm": 1.87788207443658,
      "language_loss": 0.68661767,
      "learning_rate": 8.026071442590022e-07,
      "loss": 0.70806205,
      "num_input_tokens_seen": 127494590,
      "step": 5931,
      "time_per_iteration": 2.9639930725097656
    },
    {
      "auxiliary_loss_clip": 0.01153221,
      "auxiliary_loss_mlp": 0.01028275,
      "balance_loss_clip": 1.05070615,
      "balance_loss_mlp": 1.02150989,
      "epoch": 0.7132808272710875,
      "flos": 18368739469440.0,
      "grad_norm": 1.8416012499941172,
      "language_loss": 0.80782521,
      "learning_rate": 8.019832974855134e-07,
      "loss": 0.82964015,
      "num_input_tokens_seen": 127512550,
      "step": 5932,
      "time_per_iteration": 2.4698448181152344
    },
    {
      "auxiliary_loss_clip": 0.0112397,
      "auxiliary_loss_mlp": 0.01031207,
      "balance_loss_clip": 1.04510272,
      "balance_loss_mlp": 1.02385831,
      "epoch": 0.7134010701617267,
      "flos": 23253380127360.0,
      "grad_norm": 2.151551917725363,
      "language_loss": 0.82066029,
      "learning_rate": 8.013596324502052e-07,
      "loss": 0.84221208,
      "num_input_tokens_seen": 127531015,
      "step": 5933,
      "time_per_iteration": 2.5817081928253174
    },
    {
      "auxiliary_loss_clip": 0.01148767,
      "auxiliary_loss_mlp": 0.01027861,
      "balance_loss_clip": 1.04798806,
      "balance_loss_mlp": 1.0214746,
      "epoch": 0.7135213130523658,
      "flos": 23653137565440.0,
      "grad_norm": 1.882250517348598,
      "language_loss": 0.78458548,
      "learning_rate": 8.007361492476872e-07,
      "loss": 0.80635178,
      "num_input_tokens_seen": 127550340,
      "step": 5934,
      "time_per_iteration": 3.307887554168701
    },
    {
      "auxiliary_loss_clip": 0.01131457,
      "auxiliary_loss_mlp": 0.01030951,
      "balance_loss_clip": 1.04405177,
      "balance_loss_mlp": 1.02403104,
      "epoch": 0.7136415559430048,
      "flos": 24790644443520.0,
      "grad_norm": 1.4774560854308052,
      "language_loss": 0.79130387,
      "learning_rate": 8.001128479725426e-07,
      "loss": 0.81292796,
      "num_input_tokens_seen": 127572245,
      "step": 5935,
      "time_per_iteration": 2.589768171310425
    },
    {
      "auxiliary_loss_clip": 0.01101787,
      "auxiliary_loss_mlp": 0.01024259,
      "balance_loss_clip": 1.04019928,
      "balance_loss_mlp": 1.01767313,
      "epoch": 0.713761798833644,
      "flos": 18296954138880.0,
      "grad_norm": 1.5378703662022588,
      "language_loss": 0.81004906,
      "learning_rate": 7.994897287193248e-07,
      "loss": 0.83130956,
      "num_input_tokens_seen": 127591625,
      "step": 5936,
      "time_per_iteration": 3.4010837078094482
    },
    {
      "auxiliary_loss_clip": 0.01156084,
      "auxiliary_loss_mlp": 0.01024332,
      "balance_loss_clip": 1.04831707,
      "balance_loss_mlp": 1.01734352,
      "epoch": 0.713882041724283,
      "flos": 15558262692480.0,
      "grad_norm": 2.2941994844673266,
      "language_loss": 0.83325589,
      "learning_rate": 7.988667915825605e-07,
      "loss": 0.85505998,
      "num_input_tokens_seen": 127608690,
      "step": 5937,
      "time_per_iteration": 3.272545337677002
    },
    {
      "auxiliary_loss_clip": 0.0114038,
      "auxiliary_loss_mlp": 0.01029276,
      "balance_loss_clip": 1.04767585,
      "balance_loss_mlp": 1.02181995,
      "epoch": 0.7140022846149221,
      "flos": 24061011477120.0,
      "grad_norm": 1.9754028582420127,
      "language_loss": 0.75904524,
      "learning_rate": 7.982440366567491e-07,
      "loss": 0.78074181,
      "num_input_tokens_seen": 127627180,
      "step": 5938,
      "time_per_iteration": 2.5348894596099854
    },
    {
      "auxiliary_loss_clip": 0.01145227,
      "auxiliary_loss_mlp": 0.01023071,
      "balance_loss_clip": 1.04463315,
      "balance_loss_mlp": 1.0165503,
      "epoch": 0.7141225275055613,
      "flos": 27891710248320.0,
      "grad_norm": 1.6306928679278025,
      "language_loss": 0.75448823,
      "learning_rate": 7.97621464036361e-07,
      "loss": 0.77617121,
      "num_input_tokens_seen": 127648940,
      "step": 5939,
      "time_per_iteration": 2.5549774169921875
    },
    {
      "auxiliary_loss_clip": 0.0115485,
      "auxiliary_loss_mlp": 0.01027339,
      "balance_loss_clip": 1.04658544,
      "balance_loss_mlp": 1.02046096,
      "epoch": 0.7142427703962003,
      "flos": 19682603147520.0,
      "grad_norm": 1.5000406976633862,
      "language_loss": 0.68013501,
      "learning_rate": 7.969990738158417e-07,
      "loss": 0.70195687,
      "num_input_tokens_seen": 127667350,
      "step": 5940,
      "time_per_iteration": 2.5013041496276855
    },
    {
      "auxiliary_loss_clip": 0.01154024,
      "auxiliary_loss_mlp": 0.01027212,
      "balance_loss_clip": 1.04889512,
      "balance_loss_mlp": 1.01928496,
      "epoch": 0.7143630132868394,
      "flos": 21032377447680.0,
      "grad_norm": 1.9148934217850202,
      "language_loss": 0.84998155,
      "learning_rate": 7.963768660896062e-07,
      "loss": 0.87179393,
      "num_input_tokens_seen": 127685760,
      "step": 5941,
      "time_per_iteration": 2.498347043991089
    },
    {
      "auxiliary_loss_clip": 0.0115637,
      "auxiliary_loss_mlp": 0.01028059,
      "balance_loss_clip": 1.0469079,
      "balance_loss_mlp": 1.02088308,
      "epoch": 0.7144832561774785,
      "flos": 24129923719680.0,
      "grad_norm": 1.8310598865926073,
      "language_loss": 0.82487202,
      "learning_rate": 7.957548409520432e-07,
      "loss": 0.84671628,
      "num_input_tokens_seen": 127704985,
      "step": 5942,
      "time_per_iteration": 2.6304848194122314
    },
    {
      "auxiliary_loss_clip": 0.01125136,
      "auxiliary_loss_mlp": 0.01025997,
      "balance_loss_clip": 1.04417872,
      "balance_loss_mlp": 1.01885891,
      "epoch": 0.7146034990681176,
      "flos": 16325817442560.0,
      "grad_norm": 1.8495694538886032,
      "language_loss": 0.839849,
      "learning_rate": 7.951329984975135e-07,
      "loss": 0.86136031,
      "num_input_tokens_seen": 127721925,
      "step": 5943,
      "time_per_iteration": 2.5400516986846924
    },
    {
      "auxiliary_loss_clip": 0.01029281,
      "auxiliary_loss_mlp": 0.01002755,
      "balance_loss_clip": 1.01126611,
      "balance_loss_mlp": 1.00172997,
      "epoch": 0.7147237419587567,
      "flos": 69627164232960.0,
      "grad_norm": 0.7098325189050411,
      "language_loss": 0.54323924,
      "learning_rate": 7.94511338820349e-07,
      "loss": 0.56355965,
      "num_input_tokens_seen": 127784230,
      "step": 5944,
      "time_per_iteration": 3.15289306640625
    },
    {
      "auxiliary_loss_clip": 0.01137449,
      "auxiliary_loss_mlp": 0.00761356,
      "balance_loss_clip": 1.04642975,
      "balance_loss_mlp": 1.00017262,
      "epoch": 0.7148439848493958,
      "flos": 22266806198400.0,
      "grad_norm": 1.97482872786692,
      "language_loss": 0.78072196,
      "learning_rate": 7.938898620148575e-07,
      "loss": 0.79970998,
      "num_input_tokens_seen": 127801990,
      "step": 5945,
      "time_per_iteration": 2.5303831100463867
    },
    {
      "auxiliary_loss_clip": 0.01139795,
      "auxiliary_loss_mlp": 0.0102701,
      "balance_loss_clip": 1.04668951,
      "balance_loss_mlp": 1.01984906,
      "epoch": 0.7149642277400349,
      "flos": 17931383470080.0,
      "grad_norm": 2.115003691328379,
      "language_loss": 0.70841235,
      "learning_rate": 7.932685681753135e-07,
      "loss": 0.73008037,
      "num_input_tokens_seen": 127819270,
      "step": 5946,
      "time_per_iteration": 2.4923274517059326
    },
    {
      "auxiliary_loss_clip": 0.01163364,
      "auxiliary_loss_mlp": 0.01028036,
      "balance_loss_clip": 1.04920411,
      "balance_loss_mlp": 1.02137518,
      "epoch": 0.7150844706306739,
      "flos": 31681937370240.0,
      "grad_norm": 2.0422377501032947,
      "language_loss": 0.62757593,
      "learning_rate": 7.92647457395969e-07,
      "loss": 0.64948994,
      "num_input_tokens_seen": 127841095,
      "step": 5947,
      "time_per_iteration": 2.5573573112487793
    },
    {
      "auxiliary_loss_clip": 0.01100034,
      "auxiliary_loss_mlp": 0.01027153,
      "balance_loss_clip": 1.03776455,
      "balance_loss_mlp": 1.01941645,
      "epoch": 0.7152047135213131,
      "flos": 10926217451520.0,
      "grad_norm": 2.3360823123161483,
      "language_loss": 0.73830962,
      "learning_rate": 7.920265297710444e-07,
      "loss": 0.75958157,
      "num_input_tokens_seen": 127858485,
      "step": 5948,
      "time_per_iteration": 2.5832557678222656
    },
    {
      "auxiliary_loss_clip": 0.0115484,
      "auxiliary_loss_mlp": 0.01020854,
      "balance_loss_clip": 1.04858661,
      "balance_loss_mlp": 1.01392758,
      "epoch": 0.7153249564119522,
      "flos": 20995640812800.0,
      "grad_norm": 1.7613992082882095,
      "language_loss": 0.73099625,
      "learning_rate": 7.914057853947363e-07,
      "loss": 0.7527532,
      "num_input_tokens_seen": 127877665,
      "step": 5949,
      "time_per_iteration": 2.4927306175231934
    },
    {
      "auxiliary_loss_clip": 0.01121981,
      "auxiliary_loss_mlp": 0.01035484,
      "balance_loss_clip": 1.04368639,
      "balance_loss_mlp": 1.02838516,
      "epoch": 0.7154451993025912,
      "flos": 24243114453120.0,
      "grad_norm": 1.7709549146715946,
      "language_loss": 0.62599266,
      "learning_rate": 7.907852243612089e-07,
      "loss": 0.64756733,
      "num_input_tokens_seen": 127898070,
      "step": 5950,
      "time_per_iteration": 2.5997438430786133
    },
    {
      "auxiliary_loss_clip": 0.01133954,
      "auxiliary_loss_mlp": 0.01023595,
      "balance_loss_clip": 1.04398155,
      "balance_loss_mlp": 1.01676702,
      "epoch": 0.7155654421932304,
      "flos": 23330947547520.0,
      "grad_norm": 1.7528237280657342,
      "language_loss": 0.72195303,
      "learning_rate": 7.901648467646009e-07,
      "loss": 0.74352849,
      "num_input_tokens_seen": 127917010,
      "step": 5951,
      "time_per_iteration": 3.32944393157959
    },
    {
      "auxiliary_loss_clip": 0.01169616,
      "auxiliary_loss_mlp": 0.01030894,
      "balance_loss_clip": 1.05052423,
      "balance_loss_mlp": 1.02303767,
      "epoch": 0.7156856850838694,
      "flos": 22711883621760.0,
      "grad_norm": 1.5475573810004368,
      "language_loss": 0.72447568,
      "learning_rate": 7.895446526990244e-07,
      "loss": 0.74648082,
      "num_input_tokens_seen": 127937025,
      "step": 5952,
      "time_per_iteration": 2.484856367111206
    },
    {
      "auxiliary_loss_clip": 0.0112103,
      "auxiliary_loss_mlp": 0.01027096,
      "balance_loss_clip": 1.04514003,
      "balance_loss_mlp": 1.019642,
      "epoch": 0.7158059279745085,
      "flos": 19865424395520.0,
      "grad_norm": 1.5989460613176696,
      "language_loss": 0.75720429,
      "learning_rate": 7.889246422585609e-07,
      "loss": 0.77868551,
      "num_input_tokens_seen": 127956410,
      "step": 5953,
      "time_per_iteration": 2.5970053672790527
    },
    {
      "auxiliary_loss_clip": 0.01167155,
      "auxiliary_loss_mlp": 0.01027885,
      "balance_loss_clip": 1.04919517,
      "balance_loss_mlp": 1.02087235,
      "epoch": 0.7159261708651476,
      "flos": 24134772055680.0,
      "grad_norm": 2.166469838879143,
      "language_loss": 0.73461264,
      "learning_rate": 7.883048155372675e-07,
      "loss": 0.75656295,
      "num_input_tokens_seen": 127974925,
      "step": 5954,
      "time_per_iteration": 2.477022886276245
    },
    {
      "auxiliary_loss_clip": 0.01144354,
      "auxiliary_loss_mlp": 0.01033002,
      "balance_loss_clip": 1.04769862,
      "balance_loss_mlp": 1.02625775,
      "epoch": 0.7160464137557867,
      "flos": 16983198201600.0,
      "grad_norm": 2.0631318537498884,
      "language_loss": 0.71939278,
      "learning_rate": 7.876851726291698e-07,
      "loss": 0.74116635,
      "num_input_tokens_seen": 127993225,
      "step": 5955,
      "time_per_iteration": 2.5116283893585205
    },
    {
      "auxiliary_loss_clip": 0.01127623,
      "auxiliary_loss_mlp": 0.01029785,
      "balance_loss_clip": 1.04342711,
      "balance_loss_mlp": 1.02300525,
      "epoch": 0.7161666566464258,
      "flos": 25228251838080.0,
      "grad_norm": 2.019206919823149,
      "language_loss": 0.7818051,
      "learning_rate": 7.870657136282666e-07,
      "loss": 0.80337918,
      "num_input_tokens_seen": 128012085,
      "step": 5956,
      "time_per_iteration": 2.611811399459839
    },
    {
      "auxiliary_loss_clip": 0.01145871,
      "auxiliary_loss_mlp": 0.01027309,
      "balance_loss_clip": 1.04491138,
      "balance_loss_mlp": 1.02070785,
      "epoch": 0.7162868995370649,
      "flos": 26468390851200.0,
      "grad_norm": 1.4915529432765657,
      "language_loss": 0.81651485,
      "learning_rate": 7.86446438628531e-07,
      "loss": 0.83824676,
      "num_input_tokens_seen": 128033155,
      "step": 5957,
      "time_per_iteration": 2.5431430339813232
    },
    {
      "auxiliary_loss_clip": 0.010592,
      "auxiliary_loss_mlp": 0.01001543,
      "balance_loss_clip": 1.0109427,
      "balance_loss_mlp": 1.00049388,
      "epoch": 0.716407142427704,
      "flos": 69998912040960.0,
      "grad_norm": 0.7711958125089252,
      "language_loss": 0.56909561,
      "learning_rate": 7.858273477239059e-07,
      "loss": 0.58970296,
      "num_input_tokens_seen": 128101575,
      "step": 5958,
      "time_per_iteration": 3.100001573562622
    },
    {
      "auxiliary_loss_clip": 0.01096805,
      "auxiliary_loss_mlp": 0.01028381,
      "balance_loss_clip": 1.0400064,
      "balance_loss_mlp": 1.02065003,
      "epoch": 0.716527385318343,
      "flos": 20740459616640.0,
      "grad_norm": 1.5987328237543916,
      "language_loss": 0.71456599,
      "learning_rate": 7.852084410083067e-07,
      "loss": 0.73581785,
      "num_input_tokens_seen": 128120395,
      "step": 5959,
      "time_per_iteration": 2.6132917404174805
    },
    {
      "auxiliary_loss_clip": 0.01132234,
      "auxiliary_loss_mlp": 0.01023943,
      "balance_loss_clip": 1.04473448,
      "balance_loss_mlp": 1.01753569,
      "epoch": 0.7166476282089821,
      "flos": 25371966153600.0,
      "grad_norm": 1.5475383480146827,
      "language_loss": 0.63806546,
      "learning_rate": 7.84589718575621e-07,
      "loss": 0.65962732,
      "num_input_tokens_seen": 128140840,
      "step": 5960,
      "time_per_iteration": 3.3493032455444336
    },
    {
      "auxiliary_loss_clip": 0.01136965,
      "auxiliary_loss_mlp": 0.01025521,
      "balance_loss_clip": 1.04041028,
      "balance_loss_mlp": 1.0183568,
      "epoch": 0.7167678710996213,
      "flos": 24133730561280.0,
      "grad_norm": 2.0998701163098366,
      "language_loss": 0.68744814,
      "learning_rate": 7.83971180519708e-07,
      "loss": 0.70907307,
      "num_input_tokens_seen": 128159695,
      "step": 5961,
      "time_per_iteration": 2.54312801361084
    },
    {
      "auxiliary_loss_clip": 0.01168481,
      "auxiliary_loss_mlp": 0.01027123,
      "balance_loss_clip": 1.04997444,
      "balance_loss_mlp": 1.01968455,
      "epoch": 0.7168881139902603,
      "flos": 30226586019840.0,
      "grad_norm": 1.9233664046993462,
      "language_loss": 0.75718504,
      "learning_rate": 7.833528269344008e-07,
      "loss": 0.77914107,
      "num_input_tokens_seen": 128179600,
      "step": 5962,
      "time_per_iteration": 3.3426408767700195
    },
    {
      "auxiliary_loss_clip": 0.01121338,
      "auxiliary_loss_mlp": 0.01030428,
      "balance_loss_clip": 1.04475904,
      "balance_loss_mlp": 1.02331746,
      "epoch": 0.7170083568808994,
      "flos": 14606414236800.0,
      "grad_norm": 1.920163904192582,
      "language_loss": 0.77427518,
      "learning_rate": 7.827346579135023e-07,
      "loss": 0.79579282,
      "num_input_tokens_seen": 128196940,
      "step": 5963,
      "time_per_iteration": 2.5482499599456787
    },
    {
      "auxiliary_loss_clip": 0.01135901,
      "auxiliary_loss_mlp": 0.01027895,
      "balance_loss_clip": 1.04419482,
      "balance_loss_mlp": 1.02091813,
      "epoch": 0.7171285997715385,
      "flos": 23331091201920.0,
      "grad_norm": 1.7767441886995217,
      "language_loss": 0.83184975,
      "learning_rate": 7.821166735507885e-07,
      "loss": 0.85348767,
      "num_input_tokens_seen": 128215970,
      "step": 5964,
      "time_per_iteration": 3.2305281162261963
    },
    {
      "auxiliary_loss_clip": 0.01166009,
      "auxiliary_loss_mlp": 0.01027391,
      "balance_loss_clip": 1.04971135,
      "balance_loss_mlp": 1.02120066,
      "epoch": 0.7172488426621776,
      "flos": 16543543731840.0,
      "grad_norm": 2.1807081698456594,
      "language_loss": 0.68563569,
      "learning_rate": 7.81498873940007e-07,
      "loss": 0.70756966,
      "num_input_tokens_seen": 128233185,
      "step": 5965,
      "time_per_iteration": 2.446850061416626
    },
    {
      "auxiliary_loss_clip": 0.01158793,
      "auxiliary_loss_mlp": 0.01029435,
      "balance_loss_clip": 1.04665017,
      "balance_loss_mlp": 1.02211547,
      "epoch": 0.7173690855528166,
      "flos": 26541612725760.0,
      "grad_norm": 2.4254811945634285,
      "language_loss": 0.77381849,
      "learning_rate": 7.808812591748768e-07,
      "loss": 0.79570079,
      "num_input_tokens_seen": 128253565,
      "step": 5966,
      "time_per_iteration": 2.5213825702667236
    },
    {
      "auxiliary_loss_clip": 0.01123133,
      "auxiliary_loss_mlp": 0.0102729,
      "balance_loss_clip": 1.04392886,
      "balance_loss_mlp": 1.01937711,
      "epoch": 0.7174893284434558,
      "flos": 22784099915520.0,
      "grad_norm": 1.9202687117584822,
      "language_loss": 0.64687777,
      "learning_rate": 7.802638293490915e-07,
      "loss": 0.66838205,
      "num_input_tokens_seen": 128273210,
      "step": 5967,
      "time_per_iteration": 2.5792033672332764
    },
    {
      "auxiliary_loss_clip": 0.01143251,
      "auxiliary_loss_mlp": 0.01029313,
      "balance_loss_clip": 1.04709029,
      "balance_loss_mlp": 1.02245855,
      "epoch": 0.7176095713340949,
      "flos": 23293564467840.0,
      "grad_norm": 1.5970908042288992,
      "language_loss": 0.76645488,
      "learning_rate": 7.796465845563123e-07,
      "loss": 0.78818053,
      "num_input_tokens_seen": 128292085,
      "step": 5968,
      "time_per_iteration": 2.5741922855377197
    },
    {
      "auxiliary_loss_clip": 0.01133917,
      "auxiliary_loss_mlp": 0.00760815,
      "balance_loss_clip": 1.04559457,
      "balance_loss_mlp": 1.0002172,
      "epoch": 0.7177298142247339,
      "flos": 25591631777280.0,
      "grad_norm": 1.954043488595539,
      "language_loss": 0.79639053,
      "learning_rate": 7.790295248901766e-07,
      "loss": 0.81533778,
      "num_input_tokens_seen": 128313215,
      "step": 5969,
      "time_per_iteration": 2.5781195163726807
    },
    {
      "auxiliary_loss_clip": 0.0115151,
      "auxiliary_loss_mlp": 0.01031501,
      "balance_loss_clip": 1.04627442,
      "balance_loss_mlp": 1.02453589,
      "epoch": 0.7178500571153731,
      "flos": 31652778504960.0,
      "grad_norm": 2.11669043422757,
      "language_loss": 0.62434196,
      "learning_rate": 7.784126504442902e-07,
      "loss": 0.64617217,
      "num_input_tokens_seen": 128336445,
      "step": 5970,
      "time_per_iteration": 2.596994638442993
    },
    {
      "auxiliary_loss_clip": 0.01118317,
      "auxiliary_loss_mlp": 0.01027869,
      "balance_loss_clip": 1.04361343,
      "balance_loss_mlp": 1.02071619,
      "epoch": 0.7179703000060121,
      "flos": 19427242383360.0,
      "grad_norm": 1.4660849362495478,
      "language_loss": 0.67999989,
      "learning_rate": 7.777959613122351e-07,
      "loss": 0.70146173,
      "num_input_tokens_seen": 128356270,
      "step": 5971,
      "time_per_iteration": 2.5574264526367188
    },
    {
      "auxiliary_loss_clip": 0.01133471,
      "auxiliary_loss_mlp": 0.01030171,
      "balance_loss_clip": 1.04718101,
      "balance_loss_mlp": 1.02324462,
      "epoch": 0.7180905428966512,
      "flos": 28839249072000.0,
      "grad_norm": 1.8824534603329977,
      "language_loss": 0.78103983,
      "learning_rate": 7.771794575875604e-07,
      "loss": 0.8026762,
      "num_input_tokens_seen": 128378140,
      "step": 5972,
      "time_per_iteration": 2.5651705265045166
    },
    {
      "auxiliary_loss_clip": 0.01150719,
      "auxiliary_loss_mlp": 0.01031112,
      "balance_loss_clip": 1.04836011,
      "balance_loss_mlp": 1.02349436,
      "epoch": 0.7182107857872904,
      "flos": 20047563285120.0,
      "grad_norm": 2.1714685551824986,
      "language_loss": 0.77323079,
      "learning_rate": 7.765631393637888e-07,
      "loss": 0.79504919,
      "num_input_tokens_seen": 128396335,
      "step": 5973,
      "time_per_iteration": 2.4847867488861084
    },
    {
      "auxiliary_loss_clip": 0.01148721,
      "auxiliary_loss_mlp": 0.01025342,
      "balance_loss_clip": 1.04578793,
      "balance_loss_mlp": 1.0178169,
      "epoch": 0.7183310286779294,
      "flos": 22747686503040.0,
      "grad_norm": 2.7080029078864825,
      "language_loss": 0.47919869,
      "learning_rate": 7.75947006734417e-07,
      "loss": 0.50093937,
      "num_input_tokens_seen": 128414115,
      "step": 5974,
      "time_per_iteration": 2.4924046993255615
    },
    {
      "auxiliary_loss_clip": 0.01164504,
      "auxiliary_loss_mlp": 0.01020579,
      "balance_loss_clip": 1.04609609,
      "balance_loss_mlp": 1.01358724,
      "epoch": 0.7184512715685685,
      "flos": 17158262112000.0,
      "grad_norm": 1.992307425901427,
      "language_loss": 0.82479584,
      "learning_rate": 7.753310597929101e-07,
      "loss": 0.84664667,
      "num_input_tokens_seen": 128430755,
      "step": 5975,
      "time_per_iteration": 2.4324145317077637
    },
    {
      "auxiliary_loss_clip": 0.01058256,
      "auxiliary_loss_mlp": 0.01001963,
      "balance_loss_clip": 1.01003563,
      "balance_loss_mlp": 1.00093794,
      "epoch": 0.7185715144592076,
      "flos": 65509611448320.0,
      "grad_norm": 0.8558992715316296,
      "language_loss": 0.55145454,
      "learning_rate": 7.747152986327095e-07,
      "loss": 0.57205677,
      "num_input_tokens_seen": 128491300,
      "step": 5976,
      "time_per_iteration": 2.9796600341796875
    },
    {
      "auxiliary_loss_clip": 0.01114681,
      "auxiliary_loss_mlp": 0.01027579,
      "balance_loss_clip": 1.04304028,
      "balance_loss_mlp": 1.02083206,
      "epoch": 0.7186917573498467,
      "flos": 16180522928640.0,
      "grad_norm": 1.7621209743810458,
      "language_loss": 0.67889071,
      "learning_rate": 7.740997233472228e-07,
      "loss": 0.70031333,
      "num_input_tokens_seen": 128508920,
      "step": 5977,
      "time_per_iteration": 3.2995247840881348
    },
    {
      "auxiliary_loss_clip": 0.011367,
      "auxiliary_loss_mlp": 0.01026052,
      "balance_loss_clip": 1.04439139,
      "balance_loss_mlp": 1.01951647,
      "epoch": 0.7188120002404857,
      "flos": 29242274647680.0,
      "grad_norm": 2.460371751591185,
      "language_loss": 0.70806354,
      "learning_rate": 7.734843340298329e-07,
      "loss": 0.72969103,
      "num_input_tokens_seen": 128528745,
      "step": 5978,
      "time_per_iteration": 2.587094306945801
    },
    {
      "auxiliary_loss_clip": 0.0114331,
      "auxiliary_loss_mlp": 0.01028658,
      "balance_loss_clip": 1.04464126,
      "balance_loss_mlp": 1.02108204,
      "epoch": 0.7189322431311249,
      "flos": 33401161008000.0,
      "grad_norm": 1.935060334699244,
      "language_loss": 0.75292885,
      "learning_rate": 7.72869130773895e-07,
      "loss": 0.77464861,
      "num_input_tokens_seen": 128549345,
      "step": 5979,
      "time_per_iteration": 2.6138110160827637
    },
    {
      "auxiliary_loss_clip": 0.01049899,
      "auxiliary_loss_mlp": 0.01001106,
      "balance_loss_clip": 1.01013923,
      "balance_loss_mlp": 1.00008106,
      "epoch": 0.719052486021764,
      "flos": 61351263792000.0,
      "grad_norm": 0.7890391625262524,
      "language_loss": 0.59398305,
      "learning_rate": 7.722541136727343e-07,
      "loss": 0.61449313,
      "num_input_tokens_seen": 128605360,
      "step": 5980,
      "time_per_iteration": 2.964895009994507
    },
    {
      "auxiliary_loss_clip": 0.01151414,
      "auxiliary_loss_mlp": 0.01023892,
      "balance_loss_clip": 1.04654908,
      "balance_loss_mlp": 1.01714456,
      "epoch": 0.719172728912403,
      "flos": 15596795007360.0,
      "grad_norm": 2.042866559041551,
      "language_loss": 0.80497921,
      "learning_rate": 7.716392828196483e-07,
      "loss": 0.82673228,
      "num_input_tokens_seen": 128623160,
      "step": 5981,
      "time_per_iteration": 2.4552013874053955
    },
    {
      "auxiliary_loss_clip": 0.01153344,
      "auxiliary_loss_mlp": 0.01036069,
      "balance_loss_clip": 1.04867101,
      "balance_loss_mlp": 1.02874327,
      "epoch": 0.7192929718030422,
      "flos": 15553162961280.0,
      "grad_norm": 2.7445882801800643,
      "language_loss": 0.77061582,
      "learning_rate": 7.710246383079064e-07,
      "loss": 0.79250991,
      "num_input_tokens_seen": 128638545,
      "step": 5982,
      "time_per_iteration": 2.453763246536255
    },
    {
      "auxiliary_loss_clip": 0.01136849,
      "auxiliary_loss_mlp": 0.0103055,
      "balance_loss_clip": 1.04074848,
      "balance_loss_mlp": 1.02352834,
      "epoch": 0.7194132146936812,
      "flos": 21862487733120.0,
      "grad_norm": 3.179183341636985,
      "language_loss": 0.92354667,
      "learning_rate": 7.704101802307492e-07,
      "loss": 0.94522077,
      "num_input_tokens_seen": 128650845,
      "step": 5983,
      "time_per_iteration": 2.4911720752716064
    },
    {
      "auxiliary_loss_clip": 0.01117308,
      "auxiliary_loss_mlp": 0.01033323,
      "balance_loss_clip": 1.04234815,
      "balance_loss_mlp": 1.02581882,
      "epoch": 0.7195334575843203,
      "flos": 27338900958720.0,
      "grad_norm": 1.9066371765817658,
      "language_loss": 0.87085879,
      "learning_rate": 7.697959086813912e-07,
      "loss": 0.89236516,
      "num_input_tokens_seen": 128667010,
      "step": 5984,
      "time_per_iteration": 2.6022090911865234
    },
    {
      "auxiliary_loss_clip": 0.01116817,
      "auxiliary_loss_mlp": 0.0102933,
      "balance_loss_clip": 1.04229879,
      "balance_loss_mlp": 1.02233291,
      "epoch": 0.7196537004749595,
      "flos": 18770615809920.0,
      "grad_norm": 1.5787767181035346,
      "language_loss": 0.80093139,
      "learning_rate": 7.691818237530145e-07,
      "loss": 0.82239282,
      "num_input_tokens_seen": 128685870,
      "step": 5985,
      "time_per_iteration": 2.540581464767456
    },
    {
      "auxiliary_loss_clip": 0.01123047,
      "auxiliary_loss_mlp": 0.0102596,
      "balance_loss_clip": 1.04407859,
      "balance_loss_mlp": 1.01884937,
      "epoch": 0.7197739433655985,
      "flos": 24531009960960.0,
      "grad_norm": 1.9547924409232116,
      "language_loss": 0.77168894,
      "learning_rate": 7.685679255387774e-07,
      "loss": 0.79317898,
      "num_input_tokens_seen": 128704185,
      "step": 5986,
      "time_per_iteration": 3.4047093391418457
    },
    {
      "auxiliary_loss_clip": 0.01137237,
      "auxiliary_loss_mlp": 0.01025666,
      "balance_loss_clip": 1.04645061,
      "balance_loss_mlp": 1.01857948,
      "epoch": 0.7198941862562376,
      "flos": 18040587793920.0,
      "grad_norm": 1.9472946920517729,
      "language_loss": 0.77162278,
      "learning_rate": 7.679542141318065e-07,
      "loss": 0.79325175,
      "num_input_tokens_seen": 128721290,
      "step": 5987,
      "time_per_iteration": 2.5110974311828613
    },
    {
      "auxiliary_loss_clip": 0.0112412,
      "auxiliary_loss_mlp": 0.01026425,
      "balance_loss_clip": 1.0413729,
      "balance_loss_mlp": 1.01898086,
      "epoch": 0.7200144291468767,
      "flos": 29022393542400.0,
      "grad_norm": 1.64610599489961,
      "language_loss": 0.75758183,
      "learning_rate": 7.673406896252013e-07,
      "loss": 0.77908719,
      "num_input_tokens_seen": 128742665,
      "step": 5988,
      "time_per_iteration": 3.2978878021240234
    },
    {
      "auxiliary_loss_clip": 0.01122255,
      "auxiliary_loss_mlp": 0.01033808,
      "balance_loss_clip": 1.04157925,
      "balance_loss_mlp": 1.02597046,
      "epoch": 0.7201346720375158,
      "flos": 25374264624000.0,
      "grad_norm": 1.544570902879685,
      "language_loss": 0.78397924,
      "learning_rate": 7.667273521120347e-07,
      "loss": 0.80553985,
      "num_input_tokens_seen": 128762225,
      "step": 5989,
      "time_per_iteration": 3.3561501502990723
    },
    {
      "auxiliary_loss_clip": 0.01127706,
      "auxiliary_loss_mlp": 0.01026405,
      "balance_loss_clip": 1.04457736,
      "balance_loss_mlp": 1.019467,
      "epoch": 0.7202549149281549,
      "flos": 14355614499840.0,
      "grad_norm": 1.906997415890972,
      "language_loss": 0.79654211,
      "learning_rate": 7.661142016853468e-07,
      "loss": 0.81808323,
      "num_input_tokens_seen": 128779585,
      "step": 5990,
      "time_per_iteration": 2.542182683944702
    },
    {
      "auxiliary_loss_clip": 0.01108869,
      "auxiliary_loss_mlp": 0.01027541,
      "balance_loss_clip": 1.04210103,
      "balance_loss_mlp": 1.0211277,
      "epoch": 0.7203751578187939,
      "flos": 23001682550400.0,
      "grad_norm": 1.803878803807844,
      "language_loss": 0.74676669,
      "learning_rate": 7.655012384381543e-07,
      "loss": 0.7681309,
      "num_input_tokens_seen": 128799070,
      "step": 5991,
      "time_per_iteration": 2.600414276123047
    },
    {
      "auxiliary_loss_clip": 0.01136658,
      "auxiliary_loss_mlp": 0.01023366,
      "balance_loss_clip": 1.04790568,
      "balance_loss_mlp": 1.01627362,
      "epoch": 0.7204954007094331,
      "flos": 23692424065920.0,
      "grad_norm": 1.654421799121267,
      "language_loss": 0.81728899,
      "learning_rate": 7.648884624634415e-07,
      "loss": 0.83888918,
      "num_input_tokens_seen": 128817620,
      "step": 5992,
      "time_per_iteration": 2.5252840518951416
    },
    {
      "auxiliary_loss_clip": 0.01153992,
      "auxiliary_loss_mlp": 0.01028995,
      "balance_loss_clip": 1.04908311,
      "balance_loss_mlp": 1.02142549,
      "epoch": 0.7206156436000721,
      "flos": 16253026531200.0,
      "grad_norm": 2.2749561700646144,
      "language_loss": 0.88694954,
      "learning_rate": 7.642758738541683e-07,
      "loss": 0.90877938,
      "num_input_tokens_seen": 128834200,
      "step": 5993,
      "time_per_iteration": 2.4685490131378174
    },
    {
      "auxiliary_loss_clip": 0.01048931,
      "auxiliary_loss_mlp": 0.01003398,
      "balance_loss_clip": 1.01074028,
      "balance_loss_mlp": 1.00237834,
      "epoch": 0.7207358864907112,
      "flos": 54377806504320.0,
      "grad_norm": 0.7589792289607421,
      "language_loss": 0.60787463,
      "learning_rate": 7.636634727032621e-07,
      "loss": 0.62839782,
      "num_input_tokens_seen": 128891305,
      "step": 5994,
      "time_per_iteration": 2.94522762298584
    },
    {
      "auxiliary_loss_clip": 0.01122915,
      "auxiliary_loss_mlp": 0.01028469,
      "balance_loss_clip": 1.0393281,
      "balance_loss_mlp": 1.02021432,
      "epoch": 0.7208561293813504,
      "flos": 19135540033920.0,
      "grad_norm": 2.0816263235062755,
      "language_loss": 0.79029691,
      "learning_rate": 7.630512591036231e-07,
      "loss": 0.81181073,
      "num_input_tokens_seen": 128910615,
      "step": 5995,
      "time_per_iteration": 2.5494720935821533
    },
    {
      "auxiliary_loss_clip": 0.01153538,
      "auxiliary_loss_mlp": 0.01025231,
      "balance_loss_clip": 1.04793477,
      "balance_loss_mlp": 1.01792967,
      "epoch": 0.7209763722719894,
      "flos": 17748526308480.0,
      "grad_norm": 2.1457538059649743,
      "language_loss": 0.64419222,
      "learning_rate": 7.624392331481255e-07,
      "loss": 0.66597986,
      "num_input_tokens_seen": 128928270,
      "step": 5996,
      "time_per_iteration": 2.4707024097442627
    },
    {
      "auxiliary_loss_clip": 0.01046765,
      "auxiliary_loss_mlp": 0.01004465,
      "balance_loss_clip": 1.00899899,
      "balance_loss_mlp": 1.00348747,
      "epoch": 0.7210966151626285,
      "flos": 66819488716800.0,
      "grad_norm": 0.7456020188067483,
      "language_loss": 0.51846313,
      "learning_rate": 7.618273949296115e-07,
      "loss": 0.53897536,
      "num_input_tokens_seen": 128987780,
      "step": 5997,
      "time_per_iteration": 3.03424072265625
    },
    {
      "auxiliary_loss_clip": 0.01133685,
      "auxiliary_loss_mlp": 0.01027222,
      "balance_loss_clip": 1.04328966,
      "balance_loss_mlp": 1.01956296,
      "epoch": 0.7212168580532676,
      "flos": 21141869080320.0,
      "grad_norm": 2.024964884046603,
      "language_loss": 0.68511081,
      "learning_rate": 7.612157445408987e-07,
      "loss": 0.70671982,
      "num_input_tokens_seen": 129005590,
      "step": 5998,
      "time_per_iteration": 2.53128981590271
    },
    {
      "auxiliary_loss_clip": 0.01142928,
      "auxiliary_loss_mlp": 0.01029394,
      "balance_loss_clip": 1.04928899,
      "balance_loss_mlp": 1.02144885,
      "epoch": 0.7213371009439067,
      "flos": 22345738335360.0,
      "grad_norm": 2.004500326225906,
      "language_loss": 0.74225789,
      "learning_rate": 7.606042820747716e-07,
      "loss": 0.7639811,
      "num_input_tokens_seen": 129021995,
      "step": 5999,
      "time_per_iteration": 2.5275938510894775
    },
    {
      "auxiliary_loss_clip": 0.01148283,
      "auxiliary_loss_mlp": 0.01026605,
      "balance_loss_clip": 1.05031347,
      "balance_loss_mlp": 1.01929426,
      "epoch": 0.7214573438345457,
      "flos": 18515901490560.0,
      "grad_norm": 1.7301633239825274,
      "language_loss": 0.8543936,
      "learning_rate": 7.599930076239889e-07,
      "loss": 0.87614244,
      "num_input_tokens_seen": 129039280,
      "step": 6000,
      "time_per_iteration": 2.509868621826172
    },
    {
      "auxiliary_loss_clip": 0.01117807,
      "auxiliary_loss_mlp": 0.0076106,
      "balance_loss_clip": 1.04697335,
      "balance_loss_mlp": 1.00023389,
      "epoch": 0.7215775867251849,
      "flos": 35736108606720.0,
      "grad_norm": 1.8180902292505796,
      "language_loss": 0.70284384,
      "learning_rate": 7.593819212812818e-07,
      "loss": 0.72163254,
      "num_input_tokens_seen": 129060860,
      "step": 6001,
      "time_per_iteration": 2.705749750137329
    },
    {
      "auxiliary_loss_clip": 0.01151,
      "auxiliary_loss_mlp": 0.01026555,
      "balance_loss_clip": 1.04693985,
      "balance_loss_mlp": 1.01961732,
      "epoch": 0.721697829615824,
      "flos": 20372410909440.0,
      "grad_norm": 1.849893755158785,
      "language_loss": 0.71806359,
      "learning_rate": 7.587710231393508e-07,
      "loss": 0.73983914,
      "num_input_tokens_seen": 129079215,
      "step": 6002,
      "time_per_iteration": 3.2011187076568604
    },
    {
      "auxiliary_loss_clip": 0.01072191,
      "auxiliary_loss_mlp": 0.0102419,
      "balance_loss_clip": 1.03705049,
      "balance_loss_mlp": 1.01697826,
      "epoch": 0.721818072506463,
      "flos": 20229809915520.0,
      "grad_norm": 2.1288458115800295,
      "language_loss": 0.83858395,
      "learning_rate": 7.581603132908685e-07,
      "loss": 0.85954785,
      "num_input_tokens_seen": 129097185,
      "step": 6003,
      "time_per_iteration": 2.632154941558838
    },
    {
      "auxiliary_loss_clip": 0.0111815,
      "auxiliary_loss_mlp": 0.01024852,
      "balance_loss_clip": 1.04318392,
      "balance_loss_mlp": 1.01737487,
      "epoch": 0.7219383153971022,
      "flos": 18186887888640.0,
      "grad_norm": 1.9200366139786254,
      "language_loss": 0.78292811,
      "learning_rate": 7.575497918284795e-07,
      "loss": 0.80435818,
      "num_input_tokens_seen": 129114730,
      "step": 6004,
      "time_per_iteration": 2.5318708419799805
    },
    {
      "auxiliary_loss_clip": 0.01169518,
      "auxiliary_loss_mlp": 0.01031485,
      "balance_loss_clip": 1.04848695,
      "balance_loss_mlp": 1.02420092,
      "epoch": 0.7220585582877412,
      "flos": 17342124854400.0,
      "grad_norm": 2.0176781157168002,
      "language_loss": 0.74242133,
      "learning_rate": 7.569394588447984e-07,
      "loss": 0.76443136,
      "num_input_tokens_seen": 129131745,
      "step": 6005,
      "time_per_iteration": 2.431175947189331
    },
    {
      "auxiliary_loss_clip": 0.01142515,
      "auxiliary_loss_mlp": 0.01028107,
      "balance_loss_clip": 1.04310751,
      "balance_loss_mlp": 1.02090406,
      "epoch": 0.7221788011783803,
      "flos": 16976338704000.0,
      "grad_norm": 2.287071371331218,
      "language_loss": 0.7841208,
      "learning_rate": 7.563293144324146e-07,
      "loss": 0.80582702,
      "num_input_tokens_seen": 129147295,
      "step": 6006,
      "time_per_iteration": 2.4634854793548584
    },
    {
      "auxiliary_loss_clip": 0.01166682,
      "auxiliary_loss_mlp": 0.01027733,
      "balance_loss_clip": 1.05065835,
      "balance_loss_mlp": 1.02130795,
      "epoch": 0.7222990440690195,
      "flos": 26286359702400.0,
      "grad_norm": 1.7570543867430142,
      "language_loss": 0.79853022,
      "learning_rate": 7.557193586838834e-07,
      "loss": 0.82047439,
      "num_input_tokens_seen": 129162660,
      "step": 6007,
      "time_per_iteration": 2.489490509033203
    },
    {
      "auxiliary_loss_clip": 0.01136348,
      "auxiliary_loss_mlp": 0.01024888,
      "balance_loss_clip": 1.04478419,
      "balance_loss_mlp": 1.0177834,
      "epoch": 0.7224192869596585,
      "flos": 17601687509760.0,
      "grad_norm": 2.099593664387454,
      "language_loss": 0.70667857,
      "learning_rate": 7.551095916917371e-07,
      "loss": 0.72829092,
      "num_input_tokens_seen": 129179990,
      "step": 6008,
      "time_per_iteration": 2.4936330318450928
    },
    {
      "auxiliary_loss_clip": 0.01132923,
      "auxiliary_loss_mlp": 0.01029077,
      "balance_loss_clip": 1.04550099,
      "balance_loss_mlp": 1.0207206,
      "epoch": 0.7225395298502976,
      "flos": 12932331016320.0,
      "grad_norm": 2.461661262484918,
      "language_loss": 0.66439593,
      "learning_rate": 7.545000135484758e-07,
      "loss": 0.68601596,
      "num_input_tokens_seen": 129197425,
      "step": 6009,
      "time_per_iteration": 2.5281214714050293
    },
    {
      "auxiliary_loss_clip": 0.01168894,
      "auxiliary_loss_mlp": 0.00761656,
      "balance_loss_clip": 1.05031466,
      "balance_loss_mlp": 1.00020063,
      "epoch": 0.7226597727409367,
      "flos": 29643899592960.0,
      "grad_norm": 1.9290035644865262,
      "language_loss": 0.62604046,
      "learning_rate": 7.538906243465714e-07,
      "loss": 0.64534593,
      "num_input_tokens_seen": 129217560,
      "step": 6010,
      "time_per_iteration": 2.5306835174560547
    },
    {
      "auxiliary_loss_clip": 0.01169003,
      "auxiliary_loss_mlp": 0.01025207,
      "balance_loss_clip": 1.0506382,
      "balance_loss_mlp": 1.01785755,
      "epoch": 0.7227800156315758,
      "flos": 13771635183360.0,
      "grad_norm": 2.034806767747672,
      "language_loss": 0.78797686,
      "learning_rate": 7.5328142417847e-07,
      "loss": 0.809919,
      "num_input_tokens_seen": 129234325,
      "step": 6011,
      "time_per_iteration": 2.4336512088775635
    },
    {
      "auxiliary_loss_clip": 0.01149093,
      "auxiliary_loss_mlp": 0.01024613,
      "balance_loss_clip": 1.04431927,
      "balance_loss_mlp": 1.01763034,
      "epoch": 0.7229002585222148,
      "flos": 20301882554880.0,
      "grad_norm": 2.0577349814152677,
      "language_loss": 0.69263887,
      "learning_rate": 7.526724131365838e-07,
      "loss": 0.71437597,
      "num_input_tokens_seen": 129255280,
      "step": 6012,
      "time_per_iteration": 3.3039486408233643
    },
    {
      "auxiliary_loss_clip": 0.01139371,
      "auxiliary_loss_mlp": 0.01027586,
      "balance_loss_clip": 1.04902732,
      "balance_loss_mlp": 1.02012396,
      "epoch": 0.723020501412854,
      "flos": 16581250033920.0,
      "grad_norm": 1.6119477145600274,
      "language_loss": 0.70297396,
      "learning_rate": 7.520635913133017e-07,
      "loss": 0.72464359,
      "num_input_tokens_seen": 129273910,
      "step": 6013,
      "time_per_iteration": 2.496561050415039
    },
    {
      "auxiliary_loss_clip": 0.0115807,
      "auxiliary_loss_mlp": 0.01027345,
      "balance_loss_clip": 1.04805458,
      "balance_loss_mlp": 1.01944709,
      "epoch": 0.7231407443034931,
      "flos": 28548300908160.0,
      "grad_norm": 1.7615499992057173,
      "language_loss": 0.82696795,
      "learning_rate": 7.514549588009798e-07,
      "loss": 0.84882212,
      "num_input_tokens_seen": 129294785,
      "step": 6014,
      "time_per_iteration": 3.344890594482422
    },
    {
      "auxiliary_loss_clip": 0.01142728,
      "auxiliary_loss_mlp": 0.01025604,
      "balance_loss_clip": 1.04637384,
      "balance_loss_mlp": 1.01786089,
      "epoch": 0.7232609871941321,
      "flos": 30008536508160.0,
      "grad_norm": 1.7954453465708753,
      "language_loss": 0.70570242,
      "learning_rate": 7.508465156919492e-07,
      "loss": 0.72738576,
      "num_input_tokens_seen": 129318295,
      "step": 6015,
      "time_per_iteration": 2.6120429039001465
    },
    {
      "auxiliary_loss_clip": 0.01139889,
      "auxiliary_loss_mlp": 0.01029244,
      "balance_loss_clip": 1.04601681,
      "balance_loss_mlp": 1.02192426,
      "epoch": 0.7233812300847713,
      "flos": 16654005031680.0,
      "grad_norm": 2.549608226830489,
      "language_loss": 0.61169297,
      "learning_rate": 7.502382620785083e-07,
      "loss": 0.63338429,
      "num_input_tokens_seen": 129334845,
      "step": 6016,
      "time_per_iteration": 3.4682233333587646
    },
    {
      "auxiliary_loss_clip": 0.01023723,
      "auxiliary_loss_mlp": 0.01001365,
      "balance_loss_clip": 1.01146555,
      "balance_loss_mlp": 1.00023305,
      "epoch": 0.7235014729754103,
      "flos": 67258784050560.0,
      "grad_norm": 0.8205871340081141,
      "language_loss": 0.6255067,
      "learning_rate": 7.496301980529289e-07,
      "loss": 0.64575756,
      "num_input_tokens_seen": 129398055,
      "step": 6017,
      "time_per_iteration": 3.219339370727539
    },
    {
      "auxiliary_loss_clip": 0.01166516,
      "auxiliary_loss_mlp": 0.01026965,
      "balance_loss_clip": 1.04797769,
      "balance_loss_mlp": 1.02004457,
      "epoch": 0.7236217158660494,
      "flos": 26943237671040.0,
      "grad_norm": 2.0803004880534615,
      "language_loss": 0.74327517,
      "learning_rate": 7.490223237074547e-07,
      "loss": 0.76521003,
      "num_input_tokens_seen": 129417765,
      "step": 6018,
      "time_per_iteration": 2.517594337463379
    },
    {
      "auxiliary_loss_clip": 0.01122731,
      "auxiliary_loss_mlp": 0.01029973,
      "balance_loss_clip": 1.04195142,
      "balance_loss_mlp": 1.02290392,
      "epoch": 0.7237419587566886,
      "flos": 29423372042880.0,
      "grad_norm": 1.7449495140480127,
      "language_loss": 0.66046876,
      "learning_rate": 7.484146391342989e-07,
      "loss": 0.68199581,
      "num_input_tokens_seen": 129437560,
      "step": 6019,
      "time_per_iteration": 2.610764265060425
    },
    {
      "auxiliary_loss_clip": 0.01131366,
      "auxiliary_loss_mlp": 0.01028335,
      "balance_loss_clip": 1.04410815,
      "balance_loss_mlp": 1.02138853,
      "epoch": 0.7238622016473276,
      "flos": 17821496787840.0,
      "grad_norm": 2.1947883400193002,
      "language_loss": 0.56805044,
      "learning_rate": 7.478071444256484e-07,
      "loss": 0.58964741,
      "num_input_tokens_seen": 129455320,
      "step": 6020,
      "time_per_iteration": 2.4936747550964355
    },
    {
      "auxiliary_loss_clip": 0.01130958,
      "auxiliary_loss_mlp": 0.01027316,
      "balance_loss_clip": 1.04434514,
      "balance_loss_mlp": 1.01992202,
      "epoch": 0.7239824445379667,
      "flos": 25739117020800.0,
      "grad_norm": 1.8123824945681781,
      "language_loss": 0.79572105,
      "learning_rate": 7.471998396736579e-07,
      "loss": 0.8173039,
      "num_input_tokens_seen": 129475700,
      "step": 6021,
      "time_per_iteration": 2.5847086906433105
    },
    {
      "auxiliary_loss_clip": 0.01126142,
      "auxiliary_loss_mlp": 0.01024487,
      "balance_loss_clip": 1.04520226,
      "balance_loss_mlp": 1.01736152,
      "epoch": 0.7241026874286057,
      "flos": 23148916398720.0,
      "grad_norm": 1.6412704474877562,
      "language_loss": 0.75915045,
      "learning_rate": 7.465927249704549e-07,
      "loss": 0.78065675,
      "num_input_tokens_seen": 129493585,
      "step": 6022,
      "time_per_iteration": 2.574049949645996
    },
    {
      "auxiliary_loss_clip": 0.01151842,
      "auxiliary_loss_mlp": 0.0102607,
      "balance_loss_clip": 1.046875,
      "balance_loss_mlp": 1.01904225,
      "epoch": 0.7242229303192449,
      "flos": 20266905686400.0,
      "grad_norm": 1.6504415999141175,
      "language_loss": 0.77165931,
      "learning_rate": 7.459858004081398e-07,
      "loss": 0.79343843,
      "num_input_tokens_seen": 129511555,
      "step": 6023,
      "time_per_iteration": 2.4821877479553223
    },
    {
      "auxiliary_loss_clip": 0.01020224,
      "auxiliary_loss_mlp": 0.01000878,
      "balance_loss_clip": 1.00808764,
      "balance_loss_mlp": 0.99973965,
      "epoch": 0.724343173209884,
      "flos": 62311659684480.0,
      "grad_norm": 0.6533093154779064,
      "language_loss": 0.58022368,
      "learning_rate": 7.453790660787815e-07,
      "loss": 0.60043466,
      "num_input_tokens_seen": 129579650,
      "step": 6024,
      "time_per_iteration": 3.2357540130615234
    },
    {
      "auxiliary_loss_clip": 0.01140657,
      "auxiliary_loss_mlp": 0.01026526,
      "balance_loss_clip": 1.04651582,
      "balance_loss_mlp": 1.01870251,
      "epoch": 0.724463416100523,
      "flos": 35006403813120.0,
      "grad_norm": 1.9671603326116,
      "language_loss": 0.63376701,
      "learning_rate": 7.447725220744214e-07,
      "loss": 0.65543884,
      "num_input_tokens_seen": 129601895,
      "step": 6025,
      "time_per_iteration": 2.630401134490967
    },
    {
      "auxiliary_loss_clip": 0.01168221,
      "auxiliary_loss_mlp": 0.01030065,
      "balance_loss_clip": 1.04930341,
      "balance_loss_mlp": 1.02239108,
      "epoch": 0.7245836589911622,
      "flos": 21871968923520.0,
      "grad_norm": 1.9229967721616303,
      "language_loss": 0.77247822,
      "learning_rate": 7.441661684870717e-07,
      "loss": 0.79446107,
      "num_input_tokens_seen": 129622150,
      "step": 6026,
      "time_per_iteration": 2.4945008754730225
    },
    {
      "auxiliary_loss_clip": 0.01168937,
      "auxiliary_loss_mlp": 0.01029937,
      "balance_loss_clip": 1.05044246,
      "balance_loss_mlp": 1.02271259,
      "epoch": 0.7247039018818012,
      "flos": 23006494972800.0,
      "grad_norm": 1.6799633139673031,
      "language_loss": 0.81964236,
      "learning_rate": 7.435600054087152e-07,
      "loss": 0.84163111,
      "num_input_tokens_seen": 129644315,
      "step": 6027,
      "time_per_iteration": 2.500795364379883
    },
    {
      "auxiliary_loss_clip": 0.01170098,
      "auxiliary_loss_mlp": 0.01027071,
      "balance_loss_clip": 1.05189323,
      "balance_loss_mlp": 1.01960564,
      "epoch": 0.7248241447724403,
      "flos": 31722588587520.0,
      "grad_norm": 2.164655833649386,
      "language_loss": 0.74499679,
      "learning_rate": 7.42954032931308e-07,
      "loss": 0.76696849,
      "num_input_tokens_seen": 129665355,
      "step": 6028,
      "time_per_iteration": 3.333434581756592
    },
    {
      "auxiliary_loss_clip": 0.01141854,
      "auxiliary_loss_mlp": 0.01026436,
      "balance_loss_clip": 1.04630029,
      "balance_loss_mlp": 1.01937294,
      "epoch": 0.7249443876630794,
      "flos": 34896984007680.0,
      "grad_norm": 2.976625337344399,
      "language_loss": 0.74740255,
      "learning_rate": 7.423482511467733e-07,
      "loss": 0.76908547,
      "num_input_tokens_seen": 129686125,
      "step": 6029,
      "time_per_iteration": 2.6452536582946777
    },
    {
      "auxiliary_loss_clip": 0.01085339,
      "auxiliary_loss_mlp": 0.01027052,
      "balance_loss_clip": 1.03978872,
      "balance_loss_mlp": 1.0198729,
      "epoch": 0.7250646305537185,
      "flos": 26359294268160.0,
      "grad_norm": 2.3323351708798485,
      "language_loss": 0.64472157,
      "learning_rate": 7.417426601470099e-07,
      "loss": 0.66584551,
      "num_input_tokens_seen": 129706485,
      "step": 6030,
      "time_per_iteration": 2.6629509925842285
    },
    {
      "auxiliary_loss_clip": 0.01154973,
      "auxiliary_loss_mlp": 0.01029582,
      "balance_loss_clip": 1.04701805,
      "balance_loss_mlp": 1.02167284,
      "epoch": 0.7251848734443576,
      "flos": 30081614728320.0,
      "grad_norm": 2.208859703485093,
      "language_loss": 0.78734636,
      "learning_rate": 7.411372600238841e-07,
      "loss": 0.80919194,
      "num_input_tokens_seen": 129727100,
      "step": 6031,
      "time_per_iteration": 2.5671136379241943
    },
    {
      "auxiliary_loss_clip": 0.01166115,
      "auxiliary_loss_mlp": 0.01026083,
      "balance_loss_clip": 1.04766846,
      "balance_loss_mlp": 1.0189184,
      "epoch": 0.7253051163349967,
      "flos": 17785262943360.0,
      "grad_norm": 2.0314062574551772,
      "language_loss": 0.7385596,
      "learning_rate": 7.405320508692346e-07,
      "loss": 0.76048154,
      "num_input_tokens_seen": 129745840,
      "step": 6032,
      "time_per_iteration": 2.4470086097717285
    },
    {
      "auxiliary_loss_clip": 0.01163912,
      "auxiliary_loss_mlp": 0.01028705,
      "balance_loss_clip": 1.04892159,
      "balance_loss_mlp": 1.02201152,
      "epoch": 0.7254253592256358,
      "flos": 12641346938880.0,
      "grad_norm": 1.8565080144584087,
      "language_loss": 0.7557106,
      "learning_rate": 7.399270327748727e-07,
      "loss": 0.77763677,
      "num_input_tokens_seen": 129763500,
      "step": 6033,
      "time_per_iteration": 2.449816942214966
    },
    {
      "auxiliary_loss_clip": 0.01123251,
      "auxiliary_loss_mlp": 0.00760186,
      "balance_loss_clip": 1.04069805,
      "balance_loss_mlp": 1.0002085,
      "epoch": 0.7255456021162748,
      "flos": 27199208966400.0,
      "grad_norm": 1.7901941083950736,
      "language_loss": 0.74011099,
      "learning_rate": 7.39322205832577e-07,
      "loss": 0.75894535,
      "num_input_tokens_seen": 129784390,
      "step": 6034,
      "time_per_iteration": 2.610888957977295
    },
    {
      "auxiliary_loss_clip": 0.01134157,
      "auxiliary_loss_mlp": 0.01024969,
      "balance_loss_clip": 1.04488277,
      "balance_loss_mlp": 1.01835012,
      "epoch": 0.725665845006914,
      "flos": 21288205088640.0,
      "grad_norm": 1.775278754911872,
      "language_loss": 0.80799842,
      "learning_rate": 7.387175701341009e-07,
      "loss": 0.82958966,
      "num_input_tokens_seen": 129803060,
      "step": 6035,
      "time_per_iteration": 2.517200231552124
    },
    {
      "auxiliary_loss_clip": 0.01151753,
      "auxiliary_loss_mlp": 0.01027477,
      "balance_loss_clip": 1.04591179,
      "balance_loss_mlp": 1.01983595,
      "epoch": 0.7257860878975531,
      "flos": 16033684129920.0,
      "grad_norm": 2.5984919137778824,
      "language_loss": 0.72092247,
      "learning_rate": 7.381131257711659e-07,
      "loss": 0.74271476,
      "num_input_tokens_seen": 129820165,
      "step": 6036,
      "time_per_iteration": 2.4815425872802734
    },
    {
      "auxiliary_loss_clip": 0.01134178,
      "auxiliary_loss_mlp": 0.01030831,
      "balance_loss_clip": 1.04675698,
      "balance_loss_mlp": 1.02391076,
      "epoch": 0.7259063307881921,
      "flos": 12129943052160.0,
      "grad_norm": 1.8171268548371624,
      "language_loss": 0.83217037,
      "learning_rate": 7.375088728354677e-07,
      "loss": 0.85382044,
      "num_input_tokens_seen": 129835195,
      "step": 6037,
      "time_per_iteration": 2.486875057220459
    },
    {
      "auxiliary_loss_clip": 0.01125631,
      "auxiliary_loss_mlp": 0.01023891,
      "balance_loss_clip": 1.04246855,
      "balance_loss_mlp": 1.01644921,
      "epoch": 0.7260265736788313,
      "flos": 30443845432320.0,
      "grad_norm": 2.283523733435723,
      "language_loss": 0.67796856,
      "learning_rate": 7.369048114186691e-07,
      "loss": 0.69946373,
      "num_input_tokens_seen": 129856240,
      "step": 6038,
      "time_per_iteration": 2.642566442489624
    },
    {
      "auxiliary_loss_clip": 0.01129788,
      "auxiliary_loss_mlp": 0.00760327,
      "balance_loss_clip": 1.04547513,
      "balance_loss_mlp": 1.00021636,
      "epoch": 0.7261468165694703,
      "flos": 21142264129920.0,
      "grad_norm": 1.8309073878649382,
      "language_loss": 0.83058393,
      "learning_rate": 7.363009416124055e-07,
      "loss": 0.84948504,
      "num_input_tokens_seen": 129875565,
      "step": 6039,
      "time_per_iteration": 3.3588573932647705
    },
    {
      "auxiliary_loss_clip": 0.01129129,
      "auxiliary_loss_mlp": 0.01030194,
      "balance_loss_clip": 1.04660368,
      "balance_loss_mlp": 1.02254975,
      "epoch": 0.7262670594601094,
      "flos": 22306308180480.0,
      "grad_norm": 2.2862288491978533,
      "language_loss": 0.63154906,
      "learning_rate": 7.356972635082852e-07,
      "loss": 0.65314233,
      "num_input_tokens_seen": 129894420,
      "step": 6040,
      "time_per_iteration": 3.2535743713378906
    },
    {
      "auxiliary_loss_clip": 0.01108854,
      "auxiliary_loss_mlp": 0.01028775,
      "balance_loss_clip": 1.04587424,
      "balance_loss_mlp": 1.0211519,
      "epoch": 0.7263873023507486,
      "flos": 25335049950720.0,
      "grad_norm": 1.7271910863739415,
      "language_loss": 0.75217485,
      "learning_rate": 7.35093777197884e-07,
      "loss": 0.77355123,
      "num_input_tokens_seen": 129914490,
      "step": 6041,
      "time_per_iteration": 2.630213737487793
    },
    {
      "auxiliary_loss_clip": 0.01139674,
      "auxiliary_loss_mlp": 0.01025202,
      "balance_loss_clip": 1.04908001,
      "balance_loss_mlp": 1.01832414,
      "epoch": 0.7265075452413876,
      "flos": 23878621192320.0,
      "grad_norm": 1.9623093678696535,
      "language_loss": 0.85724187,
      "learning_rate": 7.344904827727525e-07,
      "loss": 0.87889069,
      "num_input_tokens_seen": 129931670,
      "step": 6042,
      "time_per_iteration": 3.2667293548583984
    },
    {
      "auxiliary_loss_clip": 0.01125952,
      "auxiliary_loss_mlp": 0.01027588,
      "balance_loss_clip": 1.04347134,
      "balance_loss_mlp": 1.02037537,
      "epoch": 0.7266277881320267,
      "flos": 28724549967360.0,
      "grad_norm": 2.7224732507511247,
      "language_loss": 0.73279637,
      "learning_rate": 7.338873803244076e-07,
      "loss": 0.75433177,
      "num_input_tokens_seen": 129946905,
      "step": 6043,
      "time_per_iteration": 2.600013017654419
    },
    {
      "auxiliary_loss_clip": 0.01134971,
      "auxiliary_loss_mlp": 0.01022707,
      "balance_loss_clip": 1.04539371,
      "balance_loss_mlp": 1.01490486,
      "epoch": 0.7267480310226658,
      "flos": 24863507182080.0,
      "grad_norm": 1.9306921239607409,
      "language_loss": 0.80971575,
      "learning_rate": 7.332844699443401e-07,
      "loss": 0.83129251,
      "num_input_tokens_seen": 129965505,
      "step": 6044,
      "time_per_iteration": 2.557462453842163
    },
    {
      "auxiliary_loss_clip": 0.01103195,
      "auxiliary_loss_mlp": 0.0102899,
      "balance_loss_clip": 1.04051208,
      "balance_loss_mlp": 1.02252841,
      "epoch": 0.7268682739133049,
      "flos": 27198490694400.0,
      "grad_norm": 1.8687207169122715,
      "language_loss": 0.75291967,
      "learning_rate": 7.326817517240121e-07,
      "loss": 0.77424157,
      "num_input_tokens_seen": 129987210,
      "step": 6045,
      "time_per_iteration": 2.6452417373657227
    },
    {
      "auxiliary_loss_clip": 0.01152167,
      "auxiliary_loss_mlp": 0.00760747,
      "balance_loss_clip": 1.04616225,
      "balance_loss_mlp": 1.00023389,
      "epoch": 0.7269885168039439,
      "flos": 33508138688640.0,
      "grad_norm": 1.6679077356258816,
      "language_loss": 0.83194202,
      "learning_rate": 7.320792257548545e-07,
      "loss": 0.85107112,
      "num_input_tokens_seen": 130008385,
      "step": 6046,
      "time_per_iteration": 2.6381494998931885
    },
    {
      "auxiliary_loss_clip": 0.01145042,
      "auxiliary_loss_mlp": 0.01029803,
      "balance_loss_clip": 1.04747057,
      "balance_loss_mlp": 1.02217913,
      "epoch": 0.7271087596945831,
      "flos": 24313750548480.0,
      "grad_norm": 3.261507694159518,
      "language_loss": 0.76324701,
      "learning_rate": 7.314768921282704e-07,
      "loss": 0.7849955,
      "num_input_tokens_seen": 130029040,
      "step": 6047,
      "time_per_iteration": 2.551164150238037
    },
    {
      "auxiliary_loss_clip": 0.01154774,
      "auxiliary_loss_mlp": 0.01032669,
      "balance_loss_clip": 1.04811943,
      "balance_loss_mlp": 1.0253917,
      "epoch": 0.7272290025852222,
      "flos": 23805147922560.0,
      "grad_norm": 2.738044261923142,
      "language_loss": 0.72350335,
      "learning_rate": 7.30874750935633e-07,
      "loss": 0.74537778,
      "num_input_tokens_seen": 130048725,
      "step": 6048,
      "time_per_iteration": 2.520831346511841
    },
    {
      "auxiliary_loss_clip": 0.01127022,
      "auxiliary_loss_mlp": 0.01028004,
      "balance_loss_clip": 1.04828668,
      "balance_loss_mlp": 1.0211432,
      "epoch": 0.7273492454758612,
      "flos": 16720367408640.0,
      "grad_norm": 1.8919683898677593,
      "language_loss": 0.79001284,
      "learning_rate": 7.30272802268286e-07,
      "loss": 0.81156301,
      "num_input_tokens_seen": 130065720,
      "step": 6049,
      "time_per_iteration": 2.5397629737854004
    },
    {
      "auxiliary_loss_clip": 0.01072968,
      "auxiliary_loss_mlp": 0.01025283,
      "balance_loss_clip": 1.03816688,
      "balance_loss_mlp": 1.01865172,
      "epoch": 0.7274694883665004,
      "flos": 28031330413440.0,
      "grad_norm": 1.6692677653600763,
      "language_loss": 0.76507086,
      "learning_rate": 7.29671046217547e-07,
      "loss": 0.78605336,
      "num_input_tokens_seen": 130084830,
      "step": 6050,
      "time_per_iteration": 2.666482448577881
    },
    {
      "auxiliary_loss_clip": 0.01128033,
      "auxiliary_loss_mlp": 0.01028444,
      "balance_loss_clip": 1.04581451,
      "balance_loss_mlp": 1.02193475,
      "epoch": 0.7275897312571394,
      "flos": 30372706546560.0,
      "grad_norm": 1.6733152034477852,
      "language_loss": 0.81395847,
      "learning_rate": 7.290694828746988e-07,
      "loss": 0.83552325,
      "num_input_tokens_seen": 130104495,
      "step": 6051,
      "time_per_iteration": 2.6247212886810303
    },
    {
      "auxiliary_loss_clip": 0.01129499,
      "auxiliary_loss_mlp": 0.01028058,
      "balance_loss_clip": 1.04485178,
      "balance_loss_mlp": 1.02156413,
      "epoch": 0.7277099741477785,
      "flos": 19204775498880.0,
      "grad_norm": 2.1554092415129147,
      "language_loss": 0.85999918,
      "learning_rate": 7.284681123310004e-07,
      "loss": 0.88157475,
      "num_input_tokens_seen": 130123210,
      "step": 6052,
      "time_per_iteration": 2.5638842582702637
    },
    {
      "auxiliary_loss_clip": 0.01151485,
      "auxiliary_loss_mlp": 0.01029071,
      "balance_loss_clip": 1.04738545,
      "balance_loss_mlp": 1.02142107,
      "epoch": 0.7278302170384175,
      "flos": 20667884186880.0,
      "grad_norm": 1.551524885540593,
      "language_loss": 0.79448247,
      "learning_rate": 7.27866934677678e-07,
      "loss": 0.81628799,
      "num_input_tokens_seen": 130142880,
      "step": 6053,
      "time_per_iteration": 2.4973134994506836
    },
    {
      "auxiliary_loss_clip": 0.01109092,
      "auxiliary_loss_mlp": 0.01027679,
      "balance_loss_clip": 1.04529428,
      "balance_loss_mlp": 1.02077341,
      "epoch": 0.7279504599290567,
      "flos": 19093200877440.0,
      "grad_norm": 1.8067615048472734,
      "language_loss": 0.78274691,
      "learning_rate": 7.272659500059297e-07,
      "loss": 0.80411458,
      "num_input_tokens_seen": 130160220,
      "step": 6054,
      "time_per_iteration": 3.289480686187744
    },
    {
      "auxiliary_loss_clip": 0.01148155,
      "auxiliary_loss_mlp": 0.01030759,
      "balance_loss_clip": 1.04780746,
      "balance_loss_mlp": 1.02301002,
      "epoch": 0.7280707028196958,
      "flos": 19062174504960.0,
      "grad_norm": 1.9340363824057272,
      "language_loss": 0.79844505,
      "learning_rate": 7.266651584069264e-07,
      "loss": 0.82023418,
      "num_input_tokens_seen": 130177885,
      "step": 6055,
      "time_per_iteration": 2.4690845012664795
    },
    {
      "auxiliary_loss_clip": 0.01157032,
      "auxiliary_loss_mlp": 0.01030563,
      "balance_loss_clip": 1.05006242,
      "balance_loss_mlp": 1.02351141,
      "epoch": 0.7281909457103348,
      "flos": 37196308293120.0,
      "grad_norm": 1.602151876923802,
      "language_loss": 0.57119906,
      "learning_rate": 7.260645599718045e-07,
      "loss": 0.59307504,
      "num_input_tokens_seen": 130204240,
      "step": 6056,
      "time_per_iteration": 2.6359200477600098
    },
    {
      "auxiliary_loss_clip": 0.0114002,
      "auxiliary_loss_mlp": 0.01032977,
      "balance_loss_clip": 1.04550982,
      "balance_loss_mlp": 1.02484095,
      "epoch": 0.728311188600974,
      "flos": 20667094087680.0,
      "grad_norm": 1.9072519677912556,
      "language_loss": 0.67138922,
      "learning_rate": 7.254641547916767e-07,
      "loss": 0.69311917,
      "num_input_tokens_seen": 130221735,
      "step": 6057,
      "time_per_iteration": 2.5202393531799316
    },
    {
      "auxiliary_loss_clip": 0.01169513,
      "auxiliary_loss_mlp": 0.01022981,
      "balance_loss_clip": 1.05249345,
      "balance_loss_mlp": 1.01568866,
      "epoch": 0.728431431491613,
      "flos": 28840685616000.0,
      "grad_norm": 1.7216179796725373,
      "language_loss": 0.69285572,
      "learning_rate": 7.248639429576226e-07,
      "loss": 0.71478069,
      "num_input_tokens_seen": 130241190,
      "step": 6058,
      "time_per_iteration": 2.5236005783081055
    },
    {
      "auxiliary_loss_clip": 0.0115501,
      "auxiliary_loss_mlp": 0.01027595,
      "balance_loss_clip": 1.04877877,
      "balance_loss_mlp": 1.0206449,
      "epoch": 0.7285516743822521,
      "flos": 25991856092160.0,
      "grad_norm": 1.581711271931304,
      "language_loss": 0.71976328,
      "learning_rate": 7.242639245606959e-07,
      "loss": 0.74158931,
      "num_input_tokens_seen": 130260980,
      "step": 6059,
      "time_per_iteration": 2.5250959396362305
    },
    {
      "auxiliary_loss_clip": 0.01141606,
      "auxiliary_loss_mlp": 0.01032148,
      "balance_loss_clip": 1.043872,
      "balance_loss_mlp": 1.02400589,
      "epoch": 0.7286719172728913,
      "flos": 16399721675520.0,
      "grad_norm": 1.6761927280456423,
      "language_loss": 0.82019222,
      "learning_rate": 7.236640996919168e-07,
      "loss": 0.84192979,
      "num_input_tokens_seen": 130280025,
      "step": 6060,
      "time_per_iteration": 2.517779588699341
    },
    {
      "auxiliary_loss_clip": 0.0115598,
      "auxiliary_loss_mlp": 0.01026752,
      "balance_loss_clip": 1.04875445,
      "balance_loss_mlp": 1.01970971,
      "epoch": 0.7287921601635303,
      "flos": 22018161277440.0,
      "grad_norm": 1.569949506168332,
      "language_loss": 0.70726645,
      "learning_rate": 7.230644684422782e-07,
      "loss": 0.72909379,
      "num_input_tokens_seen": 130300255,
      "step": 6061,
      "time_per_iteration": 2.4982268810272217
    },
    {
      "auxiliary_loss_clip": 0.01124308,
      "auxiliary_loss_mlp": 0.01028027,
      "balance_loss_clip": 1.04603958,
      "balance_loss_mlp": 1.02074611,
      "epoch": 0.7289124030541694,
      "flos": 24600927784320.0,
      "grad_norm": 1.7420495394179374,
      "language_loss": 0.81803882,
      "learning_rate": 7.224650309027451e-07,
      "loss": 0.83956218,
      "num_input_tokens_seen": 130320005,
      "step": 6062,
      "time_per_iteration": 2.5880842208862305
    },
    {
      "auxiliary_loss_clip": 0.01156234,
      "auxiliary_loss_mlp": 0.01031002,
      "balance_loss_clip": 1.04856205,
      "balance_loss_mlp": 1.02418923,
      "epoch": 0.7290326459448085,
      "flos": 21393638484480.0,
      "grad_norm": 1.74189197124845,
      "language_loss": 0.68199974,
      "learning_rate": 7.218657871642506e-07,
      "loss": 0.70387214,
      "num_input_tokens_seen": 130338810,
      "step": 6063,
      "time_per_iteration": 2.4788947105407715
    },
    {
      "auxiliary_loss_clip": 0.01170357,
      "auxiliary_loss_mlp": 0.01029442,
      "balance_loss_clip": 1.05062675,
      "balance_loss_mlp": 1.02191937,
      "epoch": 0.7291528888354476,
      "flos": 18587686821120.0,
      "grad_norm": 2.1037413182366196,
      "language_loss": 0.62487286,
      "learning_rate": 7.212667373177012e-07,
      "loss": 0.64687085,
      "num_input_tokens_seen": 130353805,
      "step": 6064,
      "time_per_iteration": 2.438267946243286
    },
    {
      "auxiliary_loss_clip": 0.01126276,
      "auxiliary_loss_mlp": 0.01024794,
      "balance_loss_clip": 1.04621923,
      "balance_loss_mlp": 1.01846755,
      "epoch": 0.7292731317260867,
      "flos": 18951066760320.0,
      "grad_norm": 1.7392754319997927,
      "language_loss": 0.75077534,
      "learning_rate": 7.206678814539704e-07,
      "loss": 0.77228606,
      "num_input_tokens_seen": 130372105,
      "step": 6065,
      "time_per_iteration": 3.352139949798584
    },
    {
      "auxiliary_loss_clip": 0.01120325,
      "auxiliary_loss_mlp": 0.01030134,
      "balance_loss_clip": 1.04414856,
      "balance_loss_mlp": 1.0233953,
      "epoch": 0.7293933746167258,
      "flos": 21067569797760.0,
      "grad_norm": 1.485201661390943,
      "language_loss": 0.72779918,
      "learning_rate": 7.20069219663904e-07,
      "loss": 0.74930376,
      "num_input_tokens_seen": 130391990,
      "step": 6066,
      "time_per_iteration": 3.4020602703094482
    },
    {
      "auxiliary_loss_clip": 0.01153146,
      "auxiliary_loss_mlp": 0.01025279,
      "balance_loss_clip": 1.04468048,
      "balance_loss_mlp": 1.01769161,
      "epoch": 0.7295136175073649,
      "flos": 22453326547200.0,
      "grad_norm": 2.2006849493544616,
      "language_loss": 0.79642832,
      "learning_rate": 7.1947075203832e-07,
      "loss": 0.81821251,
      "num_input_tokens_seen": 130411970,
      "step": 6067,
      "time_per_iteration": 2.5171258449554443
    },
    {
      "auxiliary_loss_clip": 0.01059016,
      "auxiliary_loss_mlp": 0.01002697,
      "balance_loss_clip": 1.01084137,
      "balance_loss_mlp": 1.00165367,
      "epoch": 0.7296338603980039,
      "flos": 56125506648960.0,
      "grad_norm": 0.8656823009127745,
      "language_loss": 0.60120118,
      "learning_rate": 7.188724786680049e-07,
      "loss": 0.6218183,
      "num_input_tokens_seen": 130472440,
      "step": 6068,
      "time_per_iteration": 3.7370970249176025
    },
    {
      "auxiliary_loss_clip": 0.01135465,
      "auxiliary_loss_mlp": 0.01025322,
      "balance_loss_clip": 1.04350579,
      "balance_loss_mlp": 1.01813388,
      "epoch": 0.7297541032886431,
      "flos": 25228287751680.0,
      "grad_norm": 1.6512731070546949,
      "language_loss": 0.75458741,
      "learning_rate": 7.182743996437162e-07,
      "loss": 0.77619529,
      "num_input_tokens_seen": 130491975,
      "step": 6069,
      "time_per_iteration": 2.5837912559509277
    },
    {
      "auxiliary_loss_clip": 0.01134217,
      "auxiliary_loss_mlp": 0.01033008,
      "balance_loss_clip": 1.04493988,
      "balance_loss_mlp": 1.02552509,
      "epoch": 0.7298743461792822,
      "flos": 26467600752000.0,
      "grad_norm": 1.876076242062857,
      "language_loss": 0.68731463,
      "learning_rate": 7.176765150561819e-07,
      "loss": 0.70898688,
      "num_input_tokens_seen": 130510580,
      "step": 6070,
      "time_per_iteration": 2.613851308822632
    },
    {
      "auxiliary_loss_clip": 0.01167972,
      "auxiliary_loss_mlp": 0.01025095,
      "balance_loss_clip": 1.04835176,
      "balance_loss_mlp": 1.01792789,
      "epoch": 0.7299945890699212,
      "flos": 19569053278080.0,
      "grad_norm": 1.9003474517007866,
      "language_loss": 0.80088556,
      "learning_rate": 7.170788249961002e-07,
      "loss": 0.82281625,
      "num_input_tokens_seen": 130529090,
      "step": 6071,
      "time_per_iteration": 2.4640073776245117
    },
    {
      "auxiliary_loss_clip": 0.0116304,
      "auxiliary_loss_mlp": 0.01024543,
      "balance_loss_clip": 1.04634547,
      "balance_loss_mlp": 1.01751804,
      "epoch": 0.7301148319605604,
      "flos": 22928963466240.0,
      "grad_norm": 2.093803216761503,
      "language_loss": 0.88236481,
      "learning_rate": 7.164813295541418e-07,
      "loss": 0.90424061,
      "num_input_tokens_seen": 130548655,
      "step": 6072,
      "time_per_iteration": 2.501742124557495
    },
    {
      "auxiliary_loss_clip": 0.01139901,
      "auxiliary_loss_mlp": 0.01027813,
      "balance_loss_clip": 1.04554272,
      "balance_loss_mlp": 1.02060628,
      "epoch": 0.7302350748511994,
      "flos": 25369703596800.0,
      "grad_norm": 1.578016737955804,
      "language_loss": 0.70346665,
      "learning_rate": 7.15884028820944e-07,
      "loss": 0.72514385,
      "num_input_tokens_seen": 130567710,
      "step": 6073,
      "time_per_iteration": 2.5824508666992188
    },
    {
      "auxiliary_loss_clip": 0.01122664,
      "auxiliary_loss_mlp": 0.01027544,
      "balance_loss_clip": 1.04293001,
      "balance_loss_mlp": 1.02036715,
      "epoch": 0.7303553177418385,
      "flos": 27819170732160.0,
      "grad_norm": 2.2224399866999125,
      "language_loss": 0.60103613,
      "learning_rate": 7.152869228871185e-07,
      "loss": 0.62253821,
      "num_input_tokens_seen": 130590195,
      "step": 6074,
      "time_per_iteration": 2.6101465225219727
    },
    {
      "auxiliary_loss_clip": 0.01133903,
      "auxiliary_loss_mlp": 0.01026408,
      "balance_loss_clip": 1.04468322,
      "balance_loss_mlp": 1.01888573,
      "epoch": 0.7304755606324776,
      "flos": 24426510318720.0,
      "grad_norm": 2.3794897439644065,
      "language_loss": 0.72398841,
      "learning_rate": 7.146900118432457e-07,
      "loss": 0.74559152,
      "num_input_tokens_seen": 130609940,
      "step": 6075,
      "time_per_iteration": 2.5549445152282715
    },
    {
      "auxiliary_loss_clip": 0.01078753,
      "auxiliary_loss_mlp": 0.01026661,
      "balance_loss_clip": 1.035043,
      "balance_loss_mlp": 1.02003002,
      "epoch": 0.7305958035231167,
      "flos": 23840483927040.0,
      "grad_norm": 1.8074785753540203,
      "language_loss": 0.86017978,
      "learning_rate": 7.140932957798753e-07,
      "loss": 0.88123393,
      "num_input_tokens_seen": 130628380,
      "step": 6076,
      "time_per_iteration": 2.6902780532836914
    },
    {
      "auxiliary_loss_clip": 0.01140369,
      "auxiliary_loss_mlp": 0.01029781,
      "balance_loss_clip": 1.04320097,
      "balance_loss_mlp": 1.02279258,
      "epoch": 0.7307160464137558,
      "flos": 16726939597440.0,
      "grad_norm": 1.77651439416414,
      "language_loss": 0.71268487,
      "learning_rate": 7.134967747875309e-07,
      "loss": 0.73438644,
      "num_input_tokens_seen": 130646590,
      "step": 6077,
      "time_per_iteration": 2.523655414581299
    },
    {
      "auxiliary_loss_clip": 0.01149107,
      "auxiliary_loss_mlp": 0.0102838,
      "balance_loss_clip": 1.04644644,
      "balance_loss_mlp": 1.02120996,
      "epoch": 0.7308362893043949,
      "flos": 21798280172160.0,
      "grad_norm": 1.8614000522154395,
      "language_loss": 0.8179034,
      "learning_rate": 7.129004489567014e-07,
      "loss": 0.83967823,
      "num_input_tokens_seen": 130664070,
      "step": 6078,
      "time_per_iteration": 2.51029896736145
    },
    {
      "auxiliary_loss_clip": 0.01128551,
      "auxiliary_loss_mlp": 0.01025632,
      "balance_loss_clip": 1.04581022,
      "balance_loss_mlp": 1.01853871,
      "epoch": 0.730956532195034,
      "flos": 10707377840640.0,
      "grad_norm": 4.181653702548048,
      "language_loss": 0.77513433,
      "learning_rate": 7.123043183778512e-07,
      "loss": 0.79667616,
      "num_input_tokens_seen": 130681400,
      "step": 6079,
      "time_per_iteration": 2.5309746265411377
    },
    {
      "auxiliary_loss_clip": 0.01132328,
      "auxiliary_loss_mlp": 0.0102367,
      "balance_loss_clip": 1.04799151,
      "balance_loss_mlp": 1.01638007,
      "epoch": 0.731076775085673,
      "flos": 19791987039360.0,
      "grad_norm": 1.5873961004848585,
      "language_loss": 0.65257543,
      "learning_rate": 7.117083831414114e-07,
      "loss": 0.67413539,
      "num_input_tokens_seen": 130700675,
      "step": 6080,
      "time_per_iteration": 3.2726364135742188
    },
    {
      "auxiliary_loss_clip": 0.01162459,
      "auxiliary_loss_mlp": 0.01025575,
      "balance_loss_clip": 1.04686964,
      "balance_loss_mlp": 1.01833916,
      "epoch": 0.7311970179763122,
      "flos": 20447033414400.0,
      "grad_norm": 1.7651006069856285,
      "language_loss": 0.6995213,
      "learning_rate": 7.11112643337787e-07,
      "loss": 0.72140169,
      "num_input_tokens_seen": 130719720,
      "step": 6081,
      "time_per_iteration": 2.4463605880737305
    },
    {
      "auxiliary_loss_clip": 0.0113791,
      "auxiliary_loss_mlp": 0.01029256,
      "balance_loss_clip": 1.04699802,
      "balance_loss_mlp": 1.0214895,
      "epoch": 0.7313172608669513,
      "flos": 18513818501760.0,
      "grad_norm": 2.267652427622491,
      "language_loss": 0.76538515,
      "learning_rate": 7.10517099057349e-07,
      "loss": 0.7870568,
      "num_input_tokens_seen": 130736670,
      "step": 6082,
      "time_per_iteration": 2.505002975463867
    },
    {
      "auxiliary_loss_clip": 0.0113964,
      "auxiliary_loss_mlp": 0.01028629,
      "balance_loss_clip": 1.04692566,
      "balance_loss_mlp": 1.02145314,
      "epoch": 0.7314375037575903,
      "flos": 16180738410240.0,
      "grad_norm": 2.5891969681082214,
      "language_loss": 0.61765552,
      "learning_rate": 7.099217503904411e-07,
      "loss": 0.6393382,
      "num_input_tokens_seen": 130754525,
      "step": 6083,
      "time_per_iteration": 2.484445810317993
    },
    {
      "auxiliary_loss_clip": 0.01143485,
      "auxiliary_loss_mlp": 0.01023909,
      "balance_loss_clip": 1.04685545,
      "balance_loss_mlp": 1.01732898,
      "epoch": 0.7315577466482295,
      "flos": 17967940536960.0,
      "grad_norm": 1.9508200031578633,
      "language_loss": 0.89885527,
      "learning_rate": 7.093265974273788e-07,
      "loss": 0.92052925,
      "num_input_tokens_seen": 130772420,
      "step": 6084,
      "time_per_iteration": 2.5266106128692627
    },
    {
      "auxiliary_loss_clip": 0.01154151,
      "auxiliary_loss_mlp": 0.01022451,
      "balance_loss_clip": 1.04659629,
      "balance_loss_mlp": 1.01555145,
      "epoch": 0.7316779895388685,
      "flos": 18405440190720.0,
      "grad_norm": 1.7209563171198137,
      "language_loss": 0.71540642,
      "learning_rate": 7.087316402584447e-07,
      "loss": 0.73717242,
      "num_input_tokens_seen": 130791245,
      "step": 6085,
      "time_per_iteration": 2.458700656890869
    },
    {
      "auxiliary_loss_clip": 0.01167599,
      "auxiliary_loss_mlp": 0.01023496,
      "balance_loss_clip": 1.04874802,
      "balance_loss_mlp": 1.01665902,
      "epoch": 0.7317982324295076,
      "flos": 17928294900480.0,
      "grad_norm": 1.7456933791010956,
      "language_loss": 0.86258221,
      "learning_rate": 7.081368789738953e-07,
      "loss": 0.88449311,
      "num_input_tokens_seen": 130808445,
      "step": 6086,
      "time_per_iteration": 2.44570255279541
    },
    {
      "auxiliary_loss_clip": 0.01131242,
      "auxiliary_loss_mlp": 0.01026153,
      "balance_loss_clip": 1.04089141,
      "balance_loss_mlp": 1.01900065,
      "epoch": 0.7319184753201466,
      "flos": 27229840289280.0,
      "grad_norm": 1.840393467214699,
      "language_loss": 0.77695543,
      "learning_rate": 7.075423136639537e-07,
      "loss": 0.79852939,
      "num_input_tokens_seen": 130827700,
      "step": 6087,
      "time_per_iteration": 2.5537731647491455
    },
    {
      "auxiliary_loss_clip": 0.01118515,
      "auxiliary_loss_mlp": 0.01029132,
      "balance_loss_clip": 1.04332769,
      "balance_loss_mlp": 1.02183032,
      "epoch": 0.7320387182107858,
      "flos": 37448544574080.0,
      "grad_norm": 1.7627458311487707,
      "language_loss": 0.74613822,
      "learning_rate": 7.069479444188149e-07,
      "loss": 0.76761472,
      "num_input_tokens_seen": 130848290,
      "step": 6088,
      "time_per_iteration": 2.7087104320526123
    },
    {
      "auxiliary_loss_clip": 0.01131889,
      "auxiliary_loss_mlp": 0.01030418,
      "balance_loss_clip": 1.04540253,
      "balance_loss_mlp": 1.02313399,
      "epoch": 0.7321589611014249,
      "flos": 17859023521920.0,
      "grad_norm": 1.736095684632061,
      "language_loss": 0.82084078,
      "learning_rate": 7.063537713286453e-07,
      "loss": 0.84246385,
      "num_input_tokens_seen": 130865970,
      "step": 6089,
      "time_per_iteration": 2.5029029846191406
    },
    {
      "auxiliary_loss_clip": 0.0114518,
      "auxiliary_loss_mlp": 0.01024006,
      "balance_loss_clip": 1.04617226,
      "balance_loss_mlp": 1.01735973,
      "epoch": 0.7322792039920639,
      "flos": 26100593539200.0,
      "grad_norm": 2.254304303864522,
      "language_loss": 0.80627823,
      "learning_rate": 7.057597944835803e-07,
      "loss": 0.82797003,
      "num_input_tokens_seen": 130885245,
      "step": 6090,
      "time_per_iteration": 3.333479166030884
    },
    {
      "auxiliary_loss_clip": 0.01130771,
      "auxiliary_loss_mlp": 0.01027771,
      "balance_loss_clip": 1.04361403,
      "balance_loss_mlp": 1.02031755,
      "epoch": 0.7323994468827031,
      "flos": 25369093065600.0,
      "grad_norm": 1.6122531804171425,
      "language_loss": 0.7463448,
      "learning_rate": 7.051660139737253e-07,
      "loss": 0.76793021,
      "num_input_tokens_seen": 130903465,
      "step": 6091,
      "time_per_iteration": 2.620601177215576
    },
    {
      "auxiliary_loss_clip": 0.01152967,
      "auxiliary_loss_mlp": 0.00760879,
      "balance_loss_clip": 1.05138087,
      "balance_loss_mlp": 1.00018883,
      "epoch": 0.7325196897733421,
      "flos": 26907075653760.0,
      "grad_norm": 1.838032849785397,
      "language_loss": 0.76427937,
      "learning_rate": 7.045724298891565e-07,
      "loss": 0.78341782,
      "num_input_tokens_seen": 130922935,
      "step": 6092,
      "time_per_iteration": 2.5493738651275635
    },
    {
      "auxiliary_loss_clip": 0.01153626,
      "auxiliary_loss_mlp": 0.01025371,
      "balance_loss_clip": 1.04853892,
      "balance_loss_mlp": 1.01868916,
      "epoch": 0.7326399326639812,
      "flos": 25775781828480.0,
      "grad_norm": 1.8421568869812313,
      "language_loss": 0.69293666,
      "learning_rate": 7.039790423199192e-07,
      "loss": 0.71472657,
      "num_input_tokens_seen": 130942575,
      "step": 6093,
      "time_per_iteration": 3.296217679977417
    },
    {
      "auxiliary_loss_clip": 0.01142883,
      "auxiliary_loss_mlp": 0.01025588,
      "balance_loss_clip": 1.04747057,
      "balance_loss_mlp": 1.01784253,
      "epoch": 0.7327601755546204,
      "flos": 21032269706880.0,
      "grad_norm": 2.163429367163025,
      "language_loss": 0.77776885,
      "learning_rate": 7.033858513560322e-07,
      "loss": 0.79945362,
      "num_input_tokens_seen": 130958870,
      "step": 6094,
      "time_per_iteration": 3.4726510047912598
    },
    {
      "auxiliary_loss_clip": 0.0115489,
      "auxiliary_loss_mlp": 0.01026436,
      "balance_loss_clip": 1.04924572,
      "balance_loss_mlp": 1.01977491,
      "epoch": 0.7328804184452594,
      "flos": 16289224462080.0,
      "grad_norm": 2.102136994530873,
      "language_loss": 0.76402855,
      "learning_rate": 7.027928570874794e-07,
      "loss": 0.78584182,
      "num_input_tokens_seen": 130977060,
      "step": 6095,
      "time_per_iteration": 2.4772212505340576
    },
    {
      "auxiliary_loss_clip": 0.01164155,
      "auxiliary_loss_mlp": 0.01030154,
      "balance_loss_clip": 1.04807448,
      "balance_loss_mlp": 1.02312326,
      "epoch": 0.7330006613358985,
      "flos": 17858233422720.0,
      "grad_norm": 1.740348312831524,
      "language_loss": 0.85184395,
      "learning_rate": 7.022000596042194e-07,
      "loss": 0.87378705,
      "num_input_tokens_seen": 130994160,
      "step": 6096,
      "time_per_iteration": 2.4233646392822266
    },
    {
      "auxiliary_loss_clip": 0.01125351,
      "auxiliary_loss_mlp": 0.01023097,
      "balance_loss_clip": 1.04130638,
      "balance_loss_mlp": 1.01605725,
      "epoch": 0.7331209042265376,
      "flos": 22492074343680.0,
      "grad_norm": 1.9381121811645123,
      "language_loss": 0.82091022,
      "learning_rate": 7.016074589961784e-07,
      "loss": 0.84239471,
      "num_input_tokens_seen": 131012725,
      "step": 6097,
      "time_per_iteration": 2.577152729034424
    },
    {
      "auxiliary_loss_clip": 0.01134063,
      "auxiliary_loss_mlp": 0.0102794,
      "balance_loss_clip": 1.04571903,
      "balance_loss_mlp": 1.02103508,
      "epoch": 0.7332411471171767,
      "flos": 33072757937280.0,
      "grad_norm": 1.6834614939276833,
      "language_loss": 0.66902936,
      "learning_rate": 7.01015055353253e-07,
      "loss": 0.69064939,
      "num_input_tokens_seen": 131035150,
      "step": 6098,
      "time_per_iteration": 2.6287682056427
    },
    {
      "auxiliary_loss_clip": 0.01099402,
      "auxiliary_loss_mlp": 0.01029054,
      "balance_loss_clip": 1.0417732,
      "balance_loss_mlp": 1.02168417,
      "epoch": 0.7333613900078157,
      "flos": 22743017735040.0,
      "grad_norm": 1.6997175676599443,
      "language_loss": 0.77777505,
      "learning_rate": 7.004228487653123e-07,
      "loss": 0.79905963,
      "num_input_tokens_seen": 131055955,
      "step": 6099,
      "time_per_iteration": 2.6396737098693848
    },
    {
      "auxiliary_loss_clip": 0.01118303,
      "auxiliary_loss_mlp": 0.01028734,
      "balance_loss_clip": 1.0381602,
      "balance_loss_mlp": 1.02138114,
      "epoch": 0.7334816328984549,
      "flos": 22346133384960.0,
      "grad_norm": 1.9438831773745733,
      "language_loss": 0.78194797,
      "learning_rate": 6.998308393221906e-07,
      "loss": 0.8034184,
      "num_input_tokens_seen": 131074360,
      "step": 6100,
      "time_per_iteration": 2.553419828414917
    },
    {
      "auxiliary_loss_clip": 0.01128322,
      "auxiliary_loss_mlp": 0.01025269,
      "balance_loss_clip": 1.04496217,
      "balance_loss_mlp": 1.01821148,
      "epoch": 0.733601875789094,
      "flos": 20736149984640.0,
      "grad_norm": 2.64116979001144,
      "language_loss": 0.71212769,
      "learning_rate": 6.992390271136977e-07,
      "loss": 0.73366356,
      "num_input_tokens_seen": 131090070,
      "step": 6101,
      "time_per_iteration": 2.5550544261932373
    },
    {
      "auxiliary_loss_clip": 0.01145346,
      "auxiliary_loss_mlp": 0.01025325,
      "balance_loss_clip": 1.04429901,
      "balance_loss_mlp": 1.01840472,
      "epoch": 0.733722118679733,
      "flos": 22564362464640.0,
      "grad_norm": 1.860957948995464,
      "language_loss": 0.85585046,
      "learning_rate": 6.986474122296094e-07,
      "loss": 0.87755716,
      "num_input_tokens_seen": 131109185,
      "step": 6102,
      "time_per_iteration": 2.5047428607940674
    },
    {
      "auxiliary_loss_clip": 0.01168182,
      "auxiliary_loss_mlp": 0.01029299,
      "balance_loss_clip": 1.04960012,
      "balance_loss_mlp": 1.02163076,
      "epoch": 0.7338423615703722,
      "flos": 20084192179200.0,
      "grad_norm": 1.9199562712764588,
      "language_loss": 0.72106981,
      "learning_rate": 6.980559947596751e-07,
      "loss": 0.74304467,
      "num_input_tokens_seen": 131127725,
      "step": 6103,
      "time_per_iteration": 2.457738161087036
    },
    {
      "auxiliary_loss_clip": 0.01113439,
      "auxiliary_loss_mlp": 0.01024777,
      "balance_loss_clip": 1.04402709,
      "balance_loss_mlp": 1.01740074,
      "epoch": 0.7339626044610112,
      "flos": 21687675217920.0,
      "grad_norm": 1.9895494919822285,
      "language_loss": 0.75913978,
      "learning_rate": 6.974647747936109e-07,
      "loss": 0.78052193,
      "num_input_tokens_seen": 131146110,
      "step": 6104,
      "time_per_iteration": 2.5988969802856445
    },
    {
      "auxiliary_loss_clip": 0.01164911,
      "auxiliary_loss_mlp": 0.00760503,
      "balance_loss_clip": 1.04791963,
      "balance_loss_mlp": 1.00024211,
      "epoch": 0.7340828473516503,
      "flos": 15268248282240.0,
      "grad_norm": 2.74359589120304,
      "language_loss": 0.82417929,
      "learning_rate": 6.968737524211039e-07,
      "loss": 0.8434335,
      "num_input_tokens_seen": 131162920,
      "step": 6105,
      "time_per_iteration": 2.4375219345092773
    },
    {
      "auxiliary_loss_clip": 0.01152913,
      "auxiliary_loss_mlp": 0.01026398,
      "balance_loss_clip": 1.04902148,
      "balance_loss_mlp": 1.01918542,
      "epoch": 0.7342030902422895,
      "flos": 22930112701440.0,
      "grad_norm": 2.013593794440417,
      "language_loss": 0.80205262,
      "learning_rate": 6.962829277318132e-07,
      "loss": 0.82384568,
      "num_input_tokens_seen": 131182515,
      "step": 6106,
      "time_per_iteration": 3.3168511390686035
    },
    {
      "auxiliary_loss_clip": 0.01152585,
      "auxiliary_loss_mlp": 0.01023129,
      "balance_loss_clip": 1.04878974,
      "balance_loss_mlp": 1.01645315,
      "epoch": 0.7343233331329285,
      "flos": 25847890381440.0,
      "grad_norm": 1.7690332967528735,
      "language_loss": 0.83631134,
      "learning_rate": 6.956923008153652e-07,
      "loss": 0.85806847,
      "num_input_tokens_seen": 131202280,
      "step": 6107,
      "time_per_iteration": 2.5339887142181396
    },
    {
      "auxiliary_loss_clip": 0.01153777,
      "auxiliary_loss_mlp": 0.01024858,
      "balance_loss_clip": 1.04611635,
      "balance_loss_mlp": 1.01816761,
      "epoch": 0.7344435760235676,
      "flos": 18478985287680.0,
      "grad_norm": 2.0704005920191753,
      "language_loss": 0.84147125,
      "learning_rate": 6.951018717613593e-07,
      "loss": 0.86325759,
      "num_input_tokens_seen": 131221295,
      "step": 6108,
      "time_per_iteration": 2.495081901550293
    },
    {
      "auxiliary_loss_clip": 0.01151879,
      "auxiliary_loss_mlp": 0.0103078,
      "balance_loss_clip": 1.04825377,
      "balance_loss_mlp": 1.02327561,
      "epoch": 0.7345638189142067,
      "flos": 17640040256640.0,
      "grad_norm": 1.7753675698479794,
      "language_loss": 0.78368527,
      "learning_rate": 6.945116406593614e-07,
      "loss": 0.80551189,
      "num_input_tokens_seen": 131240150,
      "step": 6109,
      "time_per_iteration": 2.472273588180542
    },
    {
      "auxiliary_loss_clip": 0.01111536,
      "auxiliary_loss_mlp": 0.01026256,
      "balance_loss_clip": 1.04466367,
      "balance_loss_mlp": 1.01930928,
      "epoch": 0.7346840618048458,
      "flos": 20260225756800.0,
      "grad_norm": 2.1000969550896516,
      "language_loss": 0.74290729,
      "learning_rate": 6.939216075989089e-07,
      "loss": 0.76428521,
      "num_input_tokens_seen": 131258080,
      "step": 6110,
      "time_per_iteration": 2.591343879699707
    },
    {
      "auxiliary_loss_clip": 0.011378,
      "auxiliary_loss_mlp": 0.01022161,
      "balance_loss_clip": 1.04563785,
      "balance_loss_mlp": 1.01499951,
      "epoch": 0.7348043046954849,
      "flos": 29023183641600.0,
      "grad_norm": 1.7423111879945414,
      "language_loss": 0.66055018,
      "learning_rate": 6.933317726695109e-07,
      "loss": 0.68214977,
      "num_input_tokens_seen": 131279310,
      "step": 6111,
      "time_per_iteration": 2.582185983657837
    },
    {
      "auxiliary_loss_clip": 0.01121567,
      "auxiliary_loss_mlp": 0.0103122,
      "balance_loss_clip": 1.04546785,
      "balance_loss_mlp": 1.02468419,
      "epoch": 0.734924547586124,
      "flos": 17931203902080.0,
      "grad_norm": 2.8508615732905347,
      "language_loss": 0.79471958,
      "learning_rate": 6.92742135960644e-07,
      "loss": 0.8162474,
      "num_input_tokens_seen": 131297010,
      "step": 6112,
      "time_per_iteration": 2.5187957286834717
    },
    {
      "auxiliary_loss_clip": 0.01051452,
      "auxiliary_loss_mlp": 0.01003704,
      "balance_loss_clip": 1.01223016,
      "balance_loss_mlp": 1.00255322,
      "epoch": 0.7350447904767631,
      "flos": 63588319850880.0,
      "grad_norm": 0.8122942830397575,
      "language_loss": 0.55747706,
      "learning_rate": 6.921526975617556e-07,
      "loss": 0.57802862,
      "num_input_tokens_seen": 131356470,
      "step": 6113,
      "time_per_iteration": 3.1154322624206543
    },
    {
      "auxiliary_loss_clip": 0.01138935,
      "auxiliary_loss_mlp": 0.01032733,
      "balance_loss_clip": 1.04534197,
      "balance_loss_mlp": 1.02522242,
      "epoch": 0.7351650333674021,
      "flos": 21580015178880.0,
      "grad_norm": 1.876554133146853,
      "language_loss": 0.753497,
      "learning_rate": 6.915634575622631e-07,
      "loss": 0.77521372,
      "num_input_tokens_seen": 131374985,
      "step": 6114,
      "time_per_iteration": 2.535419225692749
    },
    {
      "auxiliary_loss_clip": 0.01164288,
      "auxiliary_loss_mlp": 0.01027599,
      "balance_loss_clip": 1.0469507,
      "balance_loss_mlp": 1.0207206,
      "epoch": 0.7352852762580413,
      "flos": 18186349184640.0,
      "grad_norm": 1.7238567552660033,
      "language_loss": 0.70783466,
      "learning_rate": 6.909744160515532e-07,
      "loss": 0.72975355,
      "num_input_tokens_seen": 131393125,
      "step": 6115,
      "time_per_iteration": 2.4288628101348877
    },
    {
      "auxiliary_loss_clip": 0.01138295,
      "auxiliary_loss_mlp": 0.01025943,
      "balance_loss_clip": 1.0467844,
      "balance_loss_mlp": 1.01888919,
      "epoch": 0.7354055191486804,
      "flos": 38910073063680.0,
      "grad_norm": 1.9522917318525363,
      "language_loss": 0.69428176,
      "learning_rate": 6.903855731189849e-07,
      "loss": 0.71592414,
      "num_input_tokens_seen": 131415760,
      "step": 6116,
      "time_per_iteration": 2.687075138092041
    },
    {
      "auxiliary_loss_clip": 0.01146592,
      "auxiliary_loss_mlp": 0.01033281,
      "balance_loss_clip": 1.04791164,
      "balance_loss_mlp": 1.02569962,
      "epoch": 0.7355257620393194,
      "flos": 16289978647680.0,
      "grad_norm": 2.1567325284536127,
      "language_loss": 0.81925863,
      "learning_rate": 6.897969288538825e-07,
      "loss": 0.84105736,
      "num_input_tokens_seen": 131433705,
      "step": 6117,
      "time_per_iteration": 3.2744300365448
    },
    {
      "auxiliary_loss_clip": 0.01134461,
      "auxiliary_loss_mlp": 0.01026816,
      "balance_loss_clip": 1.04419601,
      "balance_loss_mlp": 1.02014959,
      "epoch": 0.7356460049299585,
      "flos": 18114240631680.0,
      "grad_norm": 1.5929627322204443,
      "language_loss": 0.81296486,
      "learning_rate": 6.892084833455452e-07,
      "loss": 0.83457768,
      "num_input_tokens_seen": 131453275,
      "step": 6118,
      "time_per_iteration": 2.5064010620117188
    },
    {
      "auxiliary_loss_clip": 0.01150812,
      "auxiliary_loss_mlp": 0.01023778,
      "balance_loss_clip": 1.04759479,
      "balance_loss_mlp": 1.01707506,
      "epoch": 0.7357662478205976,
      "flos": 21325193118720.0,
      "grad_norm": 1.433064373283806,
      "language_loss": 0.83858848,
      "learning_rate": 6.886202366832384e-07,
      "loss": 0.8603344,
      "num_input_tokens_seen": 131474960,
      "step": 6119,
      "time_per_iteration": 3.969604969024658
    },
    {
      "auxiliary_loss_clip": 0.01107655,
      "auxiliary_loss_mlp": 0.01028203,
      "balance_loss_clip": 1.04322886,
      "balance_loss_mlp": 1.02080035,
      "epoch": 0.7358864907112367,
      "flos": 14246841139200.0,
      "grad_norm": 1.7917890606995348,
      "language_loss": 0.73691416,
      "learning_rate": 6.880321889561987e-07,
      "loss": 0.75827277,
      "num_input_tokens_seen": 131492935,
      "step": 6120,
      "time_per_iteration": 2.558135509490967
    },
    {
      "auxiliary_loss_clip": 0.01119059,
      "auxiliary_loss_mlp": 0.01027369,
      "balance_loss_clip": 1.04197633,
      "balance_loss_mlp": 1.01953959,
      "epoch": 0.7360067336018757,
      "flos": 22309684058880.0,
      "grad_norm": 1.9773859572101127,
      "language_loss": 0.65461171,
      "learning_rate": 6.874443402536338e-07,
      "loss": 0.67607605,
      "num_input_tokens_seen": 131512025,
      "step": 6121,
      "time_per_iteration": 2.5719122886657715
    },
    {
      "auxiliary_loss_clip": 0.01141796,
      "auxiliary_loss_mlp": 0.01024,
      "balance_loss_clip": 1.04685044,
      "balance_loss_mlp": 1.01654315,
      "epoch": 0.7361269764925149,
      "flos": 25554607833600.0,
      "grad_norm": 1.6605722970797028,
      "language_loss": 0.80656463,
      "learning_rate": 6.868566906647177e-07,
      "loss": 0.82822263,
      "num_input_tokens_seen": 131532975,
      "step": 6122,
      "time_per_iteration": 2.551800489425659
    },
    {
      "auxiliary_loss_clip": 0.01151533,
      "auxiliary_loss_mlp": 0.01031928,
      "balance_loss_clip": 1.04578876,
      "balance_loss_mlp": 1.02465057,
      "epoch": 0.736247219383154,
      "flos": 20376505059840.0,
      "grad_norm": 2.103444210531492,
      "language_loss": 0.83497679,
      "learning_rate": 6.862692402785984e-07,
      "loss": 0.85681146,
      "num_input_tokens_seen": 131553225,
      "step": 6123,
      "time_per_iteration": 2.51865291595459
    },
    {
      "auxiliary_loss_clip": 0.01022022,
      "auxiliary_loss_mlp": 0.01001658,
      "balance_loss_clip": 1.01061964,
      "balance_loss_mlp": 1.00063312,
      "epoch": 0.736367462273793,
      "flos": 70339525735680.0,
      "grad_norm": 0.677095504234691,
      "language_loss": 0.49643466,
      "learning_rate": 6.856819891843899e-07,
      "loss": 0.51667142,
      "num_input_tokens_seen": 131617930,
      "step": 6124,
      "time_per_iteration": 3.2114145755767822
    },
    {
      "auxiliary_loss_clip": 0.01098423,
      "auxiliary_loss_mlp": 0.01030251,
      "balance_loss_clip": 1.04441833,
      "balance_loss_mlp": 1.02333939,
      "epoch": 0.7364877051644322,
      "flos": 22412711243520.0,
      "grad_norm": 2.0757941262500985,
      "language_loss": 0.71915007,
      "learning_rate": 6.8509493747118e-07,
      "loss": 0.74043679,
      "num_input_tokens_seen": 131636740,
      "step": 6125,
      "time_per_iteration": 2.6145901679992676
    },
    {
      "auxiliary_loss_clip": 0.01165569,
      "auxiliary_loss_mlp": 0.01027095,
      "balance_loss_clip": 1.04850543,
      "balance_loss_mlp": 1.020226,
      "epoch": 0.7366079480550712,
      "flos": 12130266274560.0,
      "grad_norm": 2.5559394421903106,
      "language_loss": 0.88145477,
      "learning_rate": 6.845080852280221e-07,
      "loss": 0.90338135,
      "num_input_tokens_seen": 131653810,
      "step": 6126,
      "time_per_iteration": 2.4200439453125
    },
    {
      "auxiliary_loss_clip": 0.01123622,
      "auxiliary_loss_mlp": 0.01027826,
      "balance_loss_clip": 1.04285884,
      "balance_loss_mlp": 1.02083731,
      "epoch": 0.7367281909457103,
      "flos": 15049336844160.0,
      "grad_norm": 1.6873112870122144,
      "language_loss": 0.74205244,
      "learning_rate": 6.839214325439409e-07,
      "loss": 0.76356685,
      "num_input_tokens_seen": 131671505,
      "step": 6127,
      "time_per_iteration": 2.549854040145874
    },
    {
      "auxiliary_loss_clip": 0.01135827,
      "auxiliary_loss_mlp": 0.01025809,
      "balance_loss_clip": 1.04903531,
      "balance_loss_mlp": 1.01878214,
      "epoch": 0.7368484338363495,
      "flos": 23510752053120.0,
      "grad_norm": 1.6432480560067482,
      "language_loss": 0.71636748,
      "learning_rate": 6.833349795079327e-07,
      "loss": 0.73798382,
      "num_input_tokens_seen": 131690615,
      "step": 6128,
      "time_per_iteration": 2.5685598850250244
    },
    {
      "auxiliary_loss_clip": 0.01126544,
      "auxiliary_loss_mlp": 0.01022974,
      "balance_loss_clip": 1.04784942,
      "balance_loss_mlp": 1.01630688,
      "epoch": 0.7369686767269885,
      "flos": 27417833095680.0,
      "grad_norm": 1.8348415970836915,
      "language_loss": 0.69116414,
      "learning_rate": 6.827487262089613e-07,
      "loss": 0.7126593,
      "num_input_tokens_seen": 131711120,
      "step": 6129,
      "time_per_iteration": 2.617788553237915
    },
    {
      "auxiliary_loss_clip": 0.01036349,
      "auxiliary_loss_mlp": 0.01000504,
      "balance_loss_clip": 1.01138496,
      "balance_loss_mlp": 0.99941331,
      "epoch": 0.7370889196176276,
      "flos": 70293343824000.0,
      "grad_norm": 0.9937897978138823,
      "language_loss": 0.56776977,
      "learning_rate": 6.821626727359606e-07,
      "loss": 0.58813828,
      "num_input_tokens_seen": 131776680,
      "step": 6130,
      "time_per_iteration": 3.182039499282837
    },
    {
      "auxiliary_loss_clip": 0.01134605,
      "auxiliary_loss_mlp": 0.0102789,
      "balance_loss_clip": 1.04698098,
      "balance_loss_mlp": 1.02032018,
      "epoch": 0.7372091625082667,
      "flos": 18040839189120.0,
      "grad_norm": 2.3629747551010922,
      "language_loss": 0.77015388,
      "learning_rate": 6.815768191778348e-07,
      "loss": 0.7917788,
      "num_input_tokens_seen": 131794760,
      "step": 6131,
      "time_per_iteration": 2.4929380416870117
    },
    {
      "auxiliary_loss_clip": 0.01146193,
      "auxiliary_loss_mlp": 0.01028099,
      "balance_loss_clip": 1.04428244,
      "balance_loss_mlp": 1.02055931,
      "epoch": 0.7373294053989058,
      "flos": 33726331854720.0,
      "grad_norm": 1.9128193068857957,
      "language_loss": 0.72810936,
      "learning_rate": 6.809911656234569e-07,
      "loss": 0.74985224,
      "num_input_tokens_seen": 131816735,
      "step": 6132,
      "time_per_iteration": 3.344209909439087
    },
    {
      "auxiliary_loss_clip": 0.0112622,
      "auxiliary_loss_mlp": 0.01030979,
      "balance_loss_clip": 1.04263628,
      "balance_loss_mlp": 1.02374637,
      "epoch": 0.7374496482895448,
      "flos": 21506326427520.0,
      "grad_norm": 2.004544754553179,
      "language_loss": 0.78270507,
      "learning_rate": 6.804057121616707e-07,
      "loss": 0.80427706,
      "num_input_tokens_seen": 131834940,
      "step": 6133,
      "time_per_iteration": 2.547750949859619
    },
    {
      "auxiliary_loss_clip": 0.01153512,
      "auxiliary_loss_mlp": 0.01024653,
      "balance_loss_clip": 1.04721165,
      "balance_loss_mlp": 1.01759911,
      "epoch": 0.737569891180184,
      "flos": 24936908624640.0,
      "grad_norm": 1.9061578368619438,
      "language_loss": 0.71894169,
      "learning_rate": 6.798204588812888e-07,
      "loss": 0.74072337,
      "num_input_tokens_seen": 131854355,
      "step": 6134,
      "time_per_iteration": 2.5295207500457764
    },
    {
      "auxiliary_loss_clip": 0.01088549,
      "auxiliary_loss_mlp": 0.00760957,
      "balance_loss_clip": 1.03970265,
      "balance_loss_mlp": 1.00020766,
      "epoch": 0.7376901340708231,
      "flos": 20664544222080.0,
      "grad_norm": 1.6768866306767447,
      "language_loss": 0.75310707,
      "learning_rate": 6.792354058710937e-07,
      "loss": 0.77160209,
      "num_input_tokens_seen": 131871825,
      "step": 6135,
      "time_per_iteration": 2.6354610919952393
    },
    {
      "auxiliary_loss_clip": 0.01159292,
      "auxiliary_loss_mlp": 0.01019925,
      "balance_loss_clip": 1.04634428,
      "balance_loss_mlp": 1.01336837,
      "epoch": 0.7378103769614621,
      "flos": 23805794367360.0,
      "grad_norm": 1.90204413442809,
      "language_loss": 0.64900696,
      "learning_rate": 6.786505532198374e-07,
      "loss": 0.67079914,
      "num_input_tokens_seen": 131890770,
      "step": 6136,
      "time_per_iteration": 2.48842191696167
    },
    {
      "auxiliary_loss_clip": 0.01167388,
      "auxiliary_loss_mlp": 0.01027036,
      "balance_loss_clip": 1.04974973,
      "balance_loss_mlp": 1.01974916,
      "epoch": 0.7379306198521013,
      "flos": 22237216369920.0,
      "grad_norm": 1.8599968147522656,
      "language_loss": 0.85082793,
      "learning_rate": 6.780659010162411e-07,
      "loss": 0.8727721,
      "num_input_tokens_seen": 131909720,
      "step": 6137,
      "time_per_iteration": 2.5080342292785645
    },
    {
      "auxiliary_loss_clip": 0.01129209,
      "auxiliary_loss_mlp": 0.01027177,
      "balance_loss_clip": 1.04591942,
      "balance_loss_mlp": 1.02073038,
      "epoch": 0.7380508627427403,
      "flos": 14903108576640.0,
      "grad_norm": 1.8481285965914338,
      "language_loss": 0.83218491,
      "learning_rate": 6.774814493489975e-07,
      "loss": 0.85374874,
      "num_input_tokens_seen": 131927395,
      "step": 6138,
      "time_per_iteration": 2.5578792095184326
    },
    {
      "auxiliary_loss_clip": 0.011462,
      "auxiliary_loss_mlp": 0.01033372,
      "balance_loss_clip": 1.04509962,
      "balance_loss_mlp": 1.02675033,
      "epoch": 0.7381711056333794,
      "flos": 21685843624320.0,
      "grad_norm": 1.646596254176413,
      "language_loss": 0.65994352,
      "learning_rate": 6.768971983067655e-07,
      "loss": 0.68173927,
      "num_input_tokens_seen": 131947725,
      "step": 6139,
      "time_per_iteration": 2.4776556491851807
    },
    {
      "auxiliary_loss_clip": 0.01058477,
      "auxiliary_loss_mlp": 0.01001773,
      "balance_loss_clip": 1.01068783,
      "balance_loss_mlp": 1.00067675,
      "epoch": 0.7382913485240186,
      "flos": 52404263596800.0,
      "grad_norm": 1.0033107505907077,
      "language_loss": 0.67793834,
      "learning_rate": 6.763131479781772e-07,
      "loss": 0.69854081,
      "num_input_tokens_seen": 131997485,
      "step": 6140,
      "time_per_iteration": 2.883711338043213
    },
    {
      "auxiliary_loss_clip": 0.01131139,
      "auxiliary_loss_mlp": 0.01029041,
      "balance_loss_clip": 1.04742503,
      "balance_loss_mlp": 1.02244306,
      "epoch": 0.7384115914146576,
      "flos": 21798818876160.0,
      "grad_norm": 1.823553578936655,
      "language_loss": 0.7588678,
      "learning_rate": 6.757292984518316e-07,
      "loss": 0.78046966,
      "num_input_tokens_seen": 132016885,
      "step": 6141,
      "time_per_iteration": 2.5236330032348633
    },
    {
      "auxiliary_loss_clip": 0.01050824,
      "auxiliary_loss_mlp": 0.01003425,
      "balance_loss_clip": 1.01204967,
      "balance_loss_mlp": 1.00231063,
      "epoch": 0.7385318343052967,
      "flos": 61494331662720.0,
      "grad_norm": 0.7497340849319888,
      "language_loss": 0.56442738,
      "learning_rate": 6.751456498162981e-07,
      "loss": 0.58496988,
      "num_input_tokens_seen": 132075920,
      "step": 6142,
      "time_per_iteration": 2.9882025718688965
    },
    {
      "auxiliary_loss_clip": 0.01150583,
      "auxiliary_loss_mlp": 0.01024695,
      "balance_loss_clip": 1.04466605,
      "balance_loss_mlp": 1.01801658,
      "epoch": 0.7386520771959358,
      "flos": 17013757697280.0,
      "grad_norm": 1.8538237747670587,
      "language_loss": 0.85803211,
      "learning_rate": 6.745622021601174e-07,
      "loss": 0.87978488,
      "num_input_tokens_seen": 132092945,
      "step": 6143,
      "time_per_iteration": 3.2462685108184814
    },
    {
      "auxiliary_loss_clip": 0.01128603,
      "auxiliary_loss_mlp": 0.01025475,
      "balance_loss_clip": 1.04579973,
      "balance_loss_mlp": 1.01805162,
      "epoch": 0.7387723200865749,
      "flos": 18770759464320.0,
      "grad_norm": 1.8143213925104127,
      "language_loss": 0.69765717,
      "learning_rate": 6.739789555717954e-07,
      "loss": 0.71919799,
      "num_input_tokens_seen": 132109920,
      "step": 6144,
      "time_per_iteration": 2.5487778186798096
    },
    {
      "auxiliary_loss_clip": 0.01163313,
      "auxiliary_loss_mlp": 0.010244,
      "balance_loss_clip": 1.04697871,
      "balance_loss_mlp": 1.01718152,
      "epoch": 0.738892562977214,
      "flos": 22525542840960.0,
      "grad_norm": 1.9445961306708712,
      "language_loss": 0.77170813,
      "learning_rate": 6.733959101398124e-07,
      "loss": 0.79358524,
      "num_input_tokens_seen": 132128050,
      "step": 6145,
      "time_per_iteration": 4.044026851654053
    },
    {
      "auxiliary_loss_clip": 0.01136018,
      "auxiliary_loss_mlp": 0.01024581,
      "balance_loss_clip": 1.04468107,
      "balance_loss_mlp": 1.01773572,
      "epoch": 0.7390128058678531,
      "flos": 21501478091520.0,
      "grad_norm": 1.7133266615157359,
      "language_loss": 0.81439817,
      "learning_rate": 6.728130659526143e-07,
      "loss": 0.83600414,
      "num_input_tokens_seen": 132145860,
      "step": 6146,
      "time_per_iteration": 2.5228497982025146
    },
    {
      "auxiliary_loss_clip": 0.01140278,
      "auxiliary_loss_mlp": 0.01029374,
      "balance_loss_clip": 1.04778123,
      "balance_loss_mlp": 1.02225721,
      "epoch": 0.7391330487584922,
      "flos": 25776176878080.0,
      "grad_norm": 2.3783216644567795,
      "language_loss": 0.70805442,
      "learning_rate": 6.7223042309862e-07,
      "loss": 0.72975099,
      "num_input_tokens_seen": 132166060,
      "step": 6147,
      "time_per_iteration": 2.5813348293304443
    },
    {
      "auxiliary_loss_clip": 0.01148582,
      "auxiliary_loss_mlp": 0.01028027,
      "balance_loss_clip": 1.04497266,
      "balance_loss_mlp": 1.02106202,
      "epoch": 0.7392532916491312,
      "flos": 28366736636160.0,
      "grad_norm": 2.3630964804722545,
      "language_loss": 0.73700744,
      "learning_rate": 6.716479816662144e-07,
      "loss": 0.75877357,
      "num_input_tokens_seen": 132187790,
      "step": 6148,
      "time_per_iteration": 2.568819284439087
    },
    {
      "auxiliary_loss_clip": 0.01141577,
      "auxiliary_loss_mlp": 0.01023097,
      "balance_loss_clip": 1.04539204,
      "balance_loss_mlp": 1.01606989,
      "epoch": 0.7393735345397703,
      "flos": 23585877348480.0,
      "grad_norm": 1.8353345491842068,
      "language_loss": 0.72944868,
      "learning_rate": 6.710657417437531e-07,
      "loss": 0.75109535,
      "num_input_tokens_seen": 132207495,
      "step": 6149,
      "time_per_iteration": 2.55751371383667
    },
    {
      "auxiliary_loss_clip": 0.01135284,
      "auxiliary_loss_mlp": 0.01020349,
      "balance_loss_clip": 1.04534888,
      "balance_loss_mlp": 1.01374221,
      "epoch": 0.7394937774304094,
      "flos": 19974772373760.0,
      "grad_norm": 2.482095309706797,
      "language_loss": 0.80376053,
      "learning_rate": 6.704837034195628e-07,
      "loss": 0.82531691,
      "num_input_tokens_seen": 132225960,
      "step": 6150,
      "time_per_iteration": 2.514275074005127
    },
    {
      "auxiliary_loss_clip": 0.01143688,
      "auxiliary_loss_mlp": 0.01030188,
      "balance_loss_clip": 1.04574323,
      "balance_loss_mlp": 1.02292848,
      "epoch": 0.7396140203210485,
      "flos": 23478037741440.0,
      "grad_norm": 1.6466144073528222,
      "language_loss": 0.84782302,
      "learning_rate": 6.699018667819376e-07,
      "loss": 0.86956179,
      "num_input_tokens_seen": 132245360,
      "step": 6151,
      "time_per_iteration": 2.507328510284424
    },
    {
      "auxiliary_loss_clip": 0.01144925,
      "auxiliary_loss_mlp": 0.01028532,
      "balance_loss_clip": 1.04324925,
      "balance_loss_mlp": 1.02113485,
      "epoch": 0.7397342632116876,
      "flos": 25555433846400.0,
      "grad_norm": 1.5306626590793733,
      "language_loss": 0.72938156,
      "learning_rate": 6.693202319191415e-07,
      "loss": 0.75111616,
      "num_input_tokens_seen": 132267095,
      "step": 6152,
      "time_per_iteration": 2.551039934158325
    },
    {
      "auxiliary_loss_clip": 0.01165913,
      "auxiliary_loss_mlp": 0.01030111,
      "balance_loss_clip": 1.05101967,
      "balance_loss_mlp": 1.02299428,
      "epoch": 0.7398545061023267,
      "flos": 24755021130240.0,
      "grad_norm": 1.776993029654492,
      "language_loss": 0.74641693,
      "learning_rate": 6.687387989194084e-07,
      "loss": 0.76837718,
      "num_input_tokens_seen": 132286610,
      "step": 6153,
      "time_per_iteration": 2.4801669120788574
    },
    {
      "auxiliary_loss_clip": 0.01128423,
      "auxiliary_loss_mlp": 0.01022082,
      "balance_loss_clip": 1.04474831,
      "balance_loss_mlp": 1.01516461,
      "epoch": 0.7399747489929658,
      "flos": 16508602776960.0,
      "grad_norm": 1.8371343245252565,
      "language_loss": 0.7927689,
      "learning_rate": 6.681575678709404e-07,
      "loss": 0.81427395,
      "num_input_tokens_seen": 132305300,
      "step": 6154,
      "time_per_iteration": 2.510317802429199
    },
    {
      "auxiliary_loss_clip": 0.01148033,
      "auxiliary_loss_mlp": 0.01029012,
      "balance_loss_clip": 1.04461813,
      "balance_loss_mlp": 1.02190685,
      "epoch": 0.7400949918836048,
      "flos": 24097065753600.0,
      "grad_norm": 2.0671192980471464,
      "language_loss": 0.70752859,
      "learning_rate": 6.67576538861911e-07,
      "loss": 0.72929901,
      "num_input_tokens_seen": 132323875,
      "step": 6155,
      "time_per_iteration": 2.4956181049346924
    },
    {
      "auxiliary_loss_clip": 0.01133983,
      "auxiliary_loss_mlp": 0.01022771,
      "balance_loss_clip": 1.04637492,
      "balance_loss_mlp": 1.01630664,
      "epoch": 0.740215234774244,
      "flos": 21802517976960.0,
      "grad_norm": 2.631590792316345,
      "language_loss": 0.82064962,
      "learning_rate": 6.669957119804612e-07,
      "loss": 0.84221721,
      "num_input_tokens_seen": 132345510,
      "step": 6156,
      "time_per_iteration": 2.5605835914611816
    },
    {
      "auxiliary_loss_clip": 0.01144329,
      "auxiliary_loss_mlp": 0.01024965,
      "balance_loss_clip": 1.04542327,
      "balance_loss_mlp": 1.01712656,
      "epoch": 0.7403354776648831,
      "flos": 18733196816640.0,
      "grad_norm": 2.8954011904520573,
      "language_loss": 0.72540522,
      "learning_rate": 6.66415087314702e-07,
      "loss": 0.74709821,
      "num_input_tokens_seen": 132360465,
      "step": 6157,
      "time_per_iteration": 2.5082178115844727
    },
    {
      "auxiliary_loss_clip": 0.01137277,
      "auxiliary_loss_mlp": 0.01026713,
      "balance_loss_clip": 1.04382348,
      "balance_loss_mlp": 1.01998413,
      "epoch": 0.7404557205555221,
      "flos": 16909581277440.0,
      "grad_norm": 2.188600692675026,
      "language_loss": 0.72843164,
      "learning_rate": 6.65834664952714e-07,
      "loss": 0.75007153,
      "num_input_tokens_seen": 132377915,
      "step": 6158,
      "time_per_iteration": 3.2086164951324463
    },
    {
      "auxiliary_loss_clip": 0.01123061,
      "auxiliary_loss_mlp": 0.01027088,
      "balance_loss_clip": 1.04206991,
      "balance_loss_mlp": 1.01990891,
      "epoch": 0.7405759634461613,
      "flos": 21214408596480.0,
      "grad_norm": 1.6569265027966444,
      "language_loss": 0.75860471,
      "learning_rate": 6.652544449825457e-07,
      "loss": 0.78010619,
      "num_input_tokens_seen": 132398170,
      "step": 6159,
      "time_per_iteration": 2.5593631267547607
    },
    {
      "auxiliary_loss_clip": 0.0114183,
      "auxiliary_loss_mlp": 0.01027589,
      "balance_loss_clip": 1.04601073,
      "balance_loss_mlp": 1.02042484,
      "epoch": 0.7406962063368003,
      "flos": 20480106862080.0,
      "grad_norm": 1.6266490807209797,
      "language_loss": 0.76552564,
      "learning_rate": 6.646744274922182e-07,
      "loss": 0.78721988,
      "num_input_tokens_seen": 132416615,
      "step": 6160,
      "time_per_iteration": 2.5394034385681152
    },
    {
      "auxiliary_loss_clip": 0.01137258,
      "auxiliary_loss_mlp": 0.01030683,
      "balance_loss_clip": 1.04361892,
      "balance_loss_mlp": 1.0239656,
      "epoch": 0.7408164492274394,
      "flos": 19791915212160.0,
      "grad_norm": 2.6144675404741156,
      "language_loss": 0.75606811,
      "learning_rate": 6.640946125697171e-07,
      "loss": 0.77774757,
      "num_input_tokens_seen": 132434145,
      "step": 6161,
      "time_per_iteration": 2.5034382343292236
    },
    {
      "auxiliary_loss_clip": 0.01151883,
      "auxiliary_loss_mlp": 0.01024748,
      "balance_loss_clip": 1.04524815,
      "balance_loss_mlp": 1.01689816,
      "epoch": 0.7409366921180786,
      "flos": 29204855654400.0,
      "grad_norm": 1.7720477352127584,
      "language_loss": 0.75376821,
      "learning_rate": 6.635150003030017e-07,
      "loss": 0.77553445,
      "num_input_tokens_seen": 132452670,
      "step": 6162,
      "time_per_iteration": 2.570927381515503
    },
    {
      "auxiliary_loss_clip": 0.01106427,
      "auxiliary_loss_mlp": 0.01025119,
      "balance_loss_clip": 1.03890634,
      "balance_loss_mlp": 1.01816344,
      "epoch": 0.7410569350087176,
      "flos": 22930004960640.0,
      "grad_norm": 1.944547130207029,
      "language_loss": 0.86129034,
      "learning_rate": 6.629355907799981e-07,
      "loss": 0.88260579,
      "num_input_tokens_seen": 132472475,
      "step": 6163,
      "time_per_iteration": 2.5772950649261475
    },
    {
      "auxiliary_loss_clip": 0.01154401,
      "auxiliary_loss_mlp": 0.01028229,
      "balance_loss_clip": 1.04672194,
      "balance_loss_mlp": 1.02100825,
      "epoch": 0.7411771778993567,
      "flos": 30440397726720.0,
      "grad_norm": 1.687966527970153,
      "language_loss": 0.69099963,
      "learning_rate": 6.623563840886015e-07,
      "loss": 0.71282601,
      "num_input_tokens_seen": 132493400,
      "step": 6164,
      "time_per_iteration": 2.5851614475250244
    },
    {
      "auxiliary_loss_clip": 0.01146708,
      "auxiliary_loss_mlp": 0.0102506,
      "balance_loss_clip": 1.04458117,
      "balance_loss_mlp": 1.01829481,
      "epoch": 0.7412974207899958,
      "flos": 20522050968960.0,
      "grad_norm": 1.6446349389779142,
      "language_loss": 0.69849735,
      "learning_rate": 6.617773803166795e-07,
      "loss": 0.72021502,
      "num_input_tokens_seen": 132511725,
      "step": 6165,
      "time_per_iteration": 2.4932861328125
    },
    {
      "auxiliary_loss_clip": 0.01141592,
      "auxiliary_loss_mlp": 0.00761077,
      "balance_loss_clip": 1.04529858,
      "balance_loss_mlp": 1.00029135,
      "epoch": 0.7414176636806349,
      "flos": 22090700793600.0,
      "grad_norm": 2.2104007692142593,
      "language_loss": 0.81623411,
      "learning_rate": 6.611985795520634e-07,
      "loss": 0.83526075,
      "num_input_tokens_seen": 132530270,
      "step": 6166,
      "time_per_iteration": 2.5477118492126465
    },
    {
      "auxiliary_loss_clip": 0.0113232,
      "auxiliary_loss_mlp": 0.01027566,
      "balance_loss_clip": 1.04623342,
      "balance_loss_mlp": 1.02059197,
      "epoch": 0.7415379065712739,
      "flos": 25155245445120.0,
      "grad_norm": 1.942712286550165,
      "language_loss": 0.77781779,
      "learning_rate": 6.606199818825588e-07,
      "loss": 0.79941666,
      "num_input_tokens_seen": 132550725,
      "step": 6167,
      "time_per_iteration": 2.5791823863983154
    },
    {
      "auxiliary_loss_clip": 0.0114033,
      "auxiliary_loss_mlp": 0.01028456,
      "balance_loss_clip": 1.04209375,
      "balance_loss_mlp": 1.02136326,
      "epoch": 0.7416581494619131,
      "flos": 16871731320960.0,
      "grad_norm": 1.8510703675873201,
      "language_loss": 0.82116914,
      "learning_rate": 6.600415873959377e-07,
      "loss": 0.84285694,
      "num_input_tokens_seen": 132568600,
      "step": 6168,
      "time_per_iteration": 2.5078556537628174
    },
    {
      "auxiliary_loss_clip": 0.0109157,
      "auxiliary_loss_mlp": 0.00759558,
      "balance_loss_clip": 1.0376941,
      "balance_loss_mlp": 1.00021398,
      "epoch": 0.7417783923525522,
      "flos": 28438881102720.0,
      "grad_norm": 2.2725349970653497,
      "language_loss": 0.64730954,
      "learning_rate": 6.594633961799437e-07,
      "loss": 0.66582084,
      "num_input_tokens_seen": 132587640,
      "step": 6169,
      "time_per_iteration": 3.4517016410827637
    },
    {
      "auxiliary_loss_clip": 0.01132239,
      "auxiliary_loss_mlp": 0.01030697,
      "balance_loss_clip": 1.0447346,
      "balance_loss_mlp": 1.02362823,
      "epoch": 0.7418986352431912,
      "flos": 20084299920000.0,
      "grad_norm": 1.5555099974047548,
      "language_loss": 0.81813264,
      "learning_rate": 6.588854083222857e-07,
      "loss": 0.83976197,
      "num_input_tokens_seen": 132607075,
      "step": 6170,
      "time_per_iteration": 2.572512626647949
    },
    {
      "auxiliary_loss_clip": 0.01139197,
      "auxiliary_loss_mlp": 0.01025401,
      "balance_loss_clip": 1.04662967,
      "balance_loss_mlp": 1.01789081,
      "epoch": 0.7420188781338304,
      "flos": 18259571059200.0,
      "grad_norm": 1.8318278030641693,
      "language_loss": 0.80893505,
      "learning_rate": 6.583076239106444e-07,
      "loss": 0.83058107,
      "num_input_tokens_seen": 132625580,
      "step": 6171,
      "time_per_iteration": 3.9722211360931396
    },
    {
      "auxiliary_loss_clip": 0.01143202,
      "auxiliary_loss_mlp": 0.01025115,
      "balance_loss_clip": 1.04547739,
      "balance_loss_mlp": 1.0177927,
      "epoch": 0.7421391210244694,
      "flos": 13771994319360.0,
      "grad_norm": 3.317409160506621,
      "language_loss": 0.75216711,
      "learning_rate": 6.577300430326707e-07,
      "loss": 0.77385026,
      "num_input_tokens_seen": 132640525,
      "step": 6172,
      "time_per_iteration": 2.4950003623962402
    },
    {
      "auxiliary_loss_clip": 0.01121405,
      "auxiliary_loss_mlp": 0.01023046,
      "balance_loss_clip": 1.04563355,
      "balance_loss_mlp": 1.01645374,
      "epoch": 0.7422593639151085,
      "flos": 15961683317760.0,
      "grad_norm": 1.9282927725839085,
      "language_loss": 0.71799242,
      "learning_rate": 6.571526657759821e-07,
      "loss": 0.73943698,
      "num_input_tokens_seen": 132656265,
      "step": 6173,
      "time_per_iteration": 2.5470693111419678
    },
    {
      "auxiliary_loss_clip": 0.0114336,
      "auxiliary_loss_mlp": 0.010252,
      "balance_loss_clip": 1.04318297,
      "balance_loss_mlp": 1.01839972,
      "epoch": 0.7423796068057477,
      "flos": 30114400867200.0,
      "grad_norm": 1.6130269624329108,
      "language_loss": 0.70783198,
      "learning_rate": 6.565754922281663e-07,
      "loss": 0.72951758,
      "num_input_tokens_seen": 132678510,
      "step": 6174,
      "time_per_iteration": 2.5488967895507812
    },
    {
      "auxiliary_loss_clip": 0.01136095,
      "auxiliary_loss_mlp": 0.01023199,
      "balance_loss_clip": 1.04440928,
      "balance_loss_mlp": 1.01633,
      "epoch": 0.7424998496963867,
      "flos": 20521907314560.0,
      "grad_norm": 1.7320889884932456,
      "language_loss": 0.78373528,
      "learning_rate": 6.559985224767801e-07,
      "loss": 0.80532825,
      "num_input_tokens_seen": 132696385,
      "step": 6175,
      "time_per_iteration": 2.5369863510131836
    },
    {
      "auxiliary_loss_clip": 0.01123514,
      "auxiliary_loss_mlp": 0.01028778,
      "balance_loss_clip": 1.04377913,
      "balance_loss_mlp": 1.02162862,
      "epoch": 0.7426200925870258,
      "flos": 21871573873920.0,
      "grad_norm": 2.6196292202306553,
      "language_loss": 0.75514561,
      "learning_rate": 6.55421756609349e-07,
      "loss": 0.77666855,
      "num_input_tokens_seen": 132714640,
      "step": 6176,
      "time_per_iteration": 2.53421950340271
    },
    {
      "auxiliary_loss_clip": 0.01149731,
      "auxiliary_loss_mlp": 0.01027268,
      "balance_loss_clip": 1.04975557,
      "balance_loss_mlp": 1.01991618,
      "epoch": 0.7427403354776649,
      "flos": 26432049265920.0,
      "grad_norm": 1.8324216644891627,
      "language_loss": 0.78696227,
      "learning_rate": 6.54845194713369e-07,
      "loss": 0.80873233,
      "num_input_tokens_seen": 132735590,
      "step": 6177,
      "time_per_iteration": 2.5513641834259033
    },
    {
      "auxiliary_loss_clip": 0.01150066,
      "auxiliary_loss_mlp": 0.01025688,
      "balance_loss_clip": 1.04828858,
      "balance_loss_mlp": 1.0188508,
      "epoch": 0.742860578368304,
      "flos": 19898390102400.0,
      "grad_norm": 2.452339234687682,
      "language_loss": 0.79440141,
      "learning_rate": 6.542688368763034e-07,
      "loss": 0.81615889,
      "num_input_tokens_seen": 132753995,
      "step": 6178,
      "time_per_iteration": 2.4721994400024414
    },
    {
      "auxiliary_loss_clip": 0.01149363,
      "auxiliary_loss_mlp": 0.01026019,
      "balance_loss_clip": 1.04787886,
      "balance_loss_mlp": 1.01943862,
      "epoch": 0.742980821258943,
      "flos": 24827201510400.0,
      "grad_norm": 1.5251177458905527,
      "language_loss": 0.7728911,
      "learning_rate": 6.536926831855854e-07,
      "loss": 0.79464489,
      "num_input_tokens_seen": 132773160,
      "step": 6179,
      "time_per_iteration": 2.5258538722991943
    },
    {
      "auxiliary_loss_clip": 0.01134964,
      "auxiliary_loss_mlp": 0.01025105,
      "balance_loss_clip": 1.04668248,
      "balance_loss_mlp": 1.01805341,
      "epoch": 0.7431010641495821,
      "flos": 25228646887680.0,
      "grad_norm": 2.1912590599212947,
      "language_loss": 0.73036557,
      "learning_rate": 6.531167337286165e-07,
      "loss": 0.75196624,
      "num_input_tokens_seen": 132793180,
      "step": 6180,
      "time_per_iteration": 2.5401129722595215
    },
    {
      "auxiliary_loss_clip": 0.01135142,
      "auxiliary_loss_mlp": 0.01025519,
      "balance_loss_clip": 1.04718065,
      "balance_loss_mlp": 1.01832175,
      "epoch": 0.7432213070402213,
      "flos": 21762369550080.0,
      "grad_norm": 1.6145057994589487,
      "language_loss": 0.79473388,
      "learning_rate": 6.52540988592768e-07,
      "loss": 0.81634045,
      "num_input_tokens_seen": 132814200,
      "step": 6181,
      "time_per_iteration": 2.5438971519470215
    },
    {
      "auxiliary_loss_clip": 0.01141013,
      "auxiliary_loss_mlp": 0.01026148,
      "balance_loss_clip": 1.046404,
      "balance_loss_mlp": 1.01889372,
      "epoch": 0.7433415499308603,
      "flos": 14793832425600.0,
      "grad_norm": 1.883145803787023,
      "language_loss": 0.8327136,
      "learning_rate": 6.519654478653814e-07,
      "loss": 0.85438514,
      "num_input_tokens_seen": 132832565,
      "step": 6182,
      "time_per_iteration": 2.50787091255188
    },
    {
      "auxiliary_loss_clip": 0.01039672,
      "auxiliary_loss_mlp": 0.01000864,
      "balance_loss_clip": 1.00979686,
      "balance_loss_mlp": 0.99971402,
      "epoch": 0.7434617928214994,
      "flos": 67155577297920.0,
      "grad_norm": 0.7434844906592243,
      "language_loss": 0.56135583,
      "learning_rate": 6.51390111633763e-07,
      "loss": 0.58176124,
      "num_input_tokens_seen": 132897840,
      "step": 6183,
      "time_per_iteration": 3.17803955078125
    },
    {
      "auxiliary_loss_clip": 0.01092535,
      "auxiliary_loss_mlp": 0.01023415,
      "balance_loss_clip": 1.03901339,
      "balance_loss_mlp": 1.01658762,
      "epoch": 0.7435820357121385,
      "flos": 27377576928000.0,
      "grad_norm": 1.5171715445883973,
      "language_loss": 0.76304036,
      "learning_rate": 6.508149799851932e-07,
      "loss": 0.78419983,
      "num_input_tokens_seen": 132919505,
      "step": 6184,
      "time_per_iteration": 3.4580113887786865
    },
    {
      "auxiliary_loss_clip": 0.01133643,
      "auxiliary_loss_mlp": 0.01020428,
      "balance_loss_clip": 1.0451448,
      "balance_loss_mlp": 1.01401114,
      "epoch": 0.7437022786027776,
      "flos": 23987645948160.0,
      "grad_norm": 1.947401492562077,
      "language_loss": 0.61395884,
      "learning_rate": 6.502400530069183e-07,
      "loss": 0.63549948,
      "num_input_tokens_seen": 132939390,
      "step": 6185,
      "time_per_iteration": 2.5627334117889404
    },
    {
      "auxiliary_loss_clip": 0.01126574,
      "auxiliary_loss_mlp": 0.01029526,
      "balance_loss_clip": 1.0474,
      "balance_loss_mlp": 1.02150941,
      "epoch": 0.7438225214934167,
      "flos": 21866761451520.0,
      "grad_norm": 1.6370682420757459,
      "language_loss": 0.68387985,
      "learning_rate": 6.496653307861535e-07,
      "loss": 0.70544076,
      "num_input_tokens_seen": 132960060,
      "step": 6186,
      "time_per_iteration": 2.5780794620513916
    },
    {
      "auxiliary_loss_clip": 0.01155658,
      "auxiliary_loss_mlp": 0.0102911,
      "balance_loss_clip": 1.04650307,
      "balance_loss_mlp": 1.02232063,
      "epoch": 0.7439427643840558,
      "flos": 20230097224320.0,
      "grad_norm": 1.944131808770365,
      "language_loss": 0.65853155,
      "learning_rate": 6.490908134100857e-07,
      "loss": 0.68037921,
      "num_input_tokens_seen": 132978525,
      "step": 6187,
      "time_per_iteration": 2.4751555919647217
    },
    {
      "auxiliary_loss_clip": 0.01155397,
      "auxiliary_loss_mlp": 0.01026809,
      "balance_loss_clip": 1.04768372,
      "balance_loss_mlp": 1.01997864,
      "epoch": 0.7440630072746949,
      "flos": 20849915335680.0,
      "grad_norm": 2.0537084445751255,
      "language_loss": 0.69182646,
      "learning_rate": 6.48516500965866e-07,
      "loss": 0.71364856,
      "num_input_tokens_seen": 132998460,
      "step": 6188,
      "time_per_iteration": 2.5139377117156982
    },
    {
      "auxiliary_loss_clip": 0.01152989,
      "auxiliary_loss_mlp": 0.01023896,
      "balance_loss_clip": 1.04461336,
      "balance_loss_mlp": 1.01697326,
      "epoch": 0.7441832501653339,
      "flos": 26503762769280.0,
      "grad_norm": 1.6161538533865363,
      "language_loss": 0.81556493,
      "learning_rate": 6.479423935406192e-07,
      "loss": 0.8373338,
      "num_input_tokens_seen": 133018445,
      "step": 6189,
      "time_per_iteration": 2.5277135372161865
    },
    {
      "auxiliary_loss_clip": 0.01033325,
      "auxiliary_loss_mlp": 0.01002864,
      "balance_loss_clip": 1.0107131,
      "balance_loss_mlp": 1.00189877,
      "epoch": 0.7443034930559731,
      "flos": 68602848088320.0,
      "grad_norm": 0.8098749032001336,
      "language_loss": 0.61988962,
      "learning_rate": 6.473684912214357e-07,
      "loss": 0.64025152,
      "num_input_tokens_seen": 133082005,
      "step": 6190,
      "time_per_iteration": 3.256326198577881
    },
    {
      "auxiliary_loss_clip": 0.01151879,
      "auxiliary_loss_mlp": 0.01029996,
      "balance_loss_clip": 1.04806519,
      "balance_loss_mlp": 1.02252734,
      "epoch": 0.7444237359466122,
      "flos": 18654982951680.0,
      "grad_norm": 1.9418326515742503,
      "language_loss": 0.69673413,
      "learning_rate": 6.467947940953778e-07,
      "loss": 0.71855283,
      "num_input_tokens_seen": 133100530,
      "step": 6191,
      "time_per_iteration": 2.4932548999786377
    },
    {
      "auxiliary_loss_clip": 0.0113491,
      "auxiliary_loss_mlp": 0.01025062,
      "balance_loss_clip": 1.04351151,
      "balance_loss_mlp": 1.01839256,
      "epoch": 0.7445439788372512,
      "flos": 22817604326400.0,
      "grad_norm": 1.862387235930378,
      "language_loss": 0.72491407,
      "learning_rate": 6.462213022494732e-07,
      "loss": 0.74651378,
      "num_input_tokens_seen": 133119775,
      "step": 6192,
      "time_per_iteration": 2.537412405014038
    },
    {
      "auxiliary_loss_clip": 0.01049656,
      "auxiliary_loss_mlp": 0.01000739,
      "balance_loss_clip": 1.01045442,
      "balance_loss_mlp": 0.99966013,
      "epoch": 0.7446642217278904,
      "flos": 67045690615680.0,
      "grad_norm": 2.6854363585244263,
      "language_loss": 0.61095309,
      "learning_rate": 6.456480157707201e-07,
      "loss": 0.63145703,
      "num_input_tokens_seen": 133184550,
      "step": 6193,
      "time_per_iteration": 3.0377211570739746
    },
    {
      "auxiliary_loss_clip": 0.01116945,
      "auxiliary_loss_mlp": 0.01024211,
      "balance_loss_clip": 1.04308438,
      "balance_loss_mlp": 1.01709771,
      "epoch": 0.7447844646185294,
      "flos": 17417465631360.0,
      "grad_norm": 2.05414268661914,
      "language_loss": 0.85190415,
      "learning_rate": 6.450749347460866e-07,
      "loss": 0.87331569,
      "num_input_tokens_seen": 133201525,
      "step": 6194,
      "time_per_iteration": 2.538580894470215
    },
    {
      "auxiliary_loss_clip": 0.01167593,
      "auxiliary_loss_mlp": 0.01026336,
      "balance_loss_clip": 1.04921508,
      "balance_loss_mlp": 1.01898718,
      "epoch": 0.7449047075091685,
      "flos": 26615876094720.0,
      "grad_norm": 1.7320051721240353,
      "language_loss": 0.79471231,
      "learning_rate": 6.445020592625083e-07,
      "loss": 0.8166517,
      "num_input_tokens_seen": 133222175,
      "step": 6195,
      "time_per_iteration": 2.497760057449341
    },
    {
      "auxiliary_loss_clip": 0.01163742,
      "auxiliary_loss_mlp": 0.01022491,
      "balance_loss_clip": 1.04662657,
      "balance_loss_mlp": 1.01549983,
      "epoch": 0.7450249503998077,
      "flos": 14170458867840.0,
      "grad_norm": 2.6964442897500964,
      "language_loss": 0.80402261,
      "learning_rate": 6.4392938940689e-07,
      "loss": 0.82588494,
      "num_input_tokens_seen": 133237590,
      "step": 6196,
      "time_per_iteration": 3.1899240016937256
    },
    {
      "auxiliary_loss_clip": 0.01105921,
      "auxiliary_loss_mlp": 0.00760835,
      "balance_loss_clip": 1.04297066,
      "balance_loss_mlp": 1.00020301,
      "epoch": 0.7451451932904467,
      "flos": 19606687752960.0,
      "grad_norm": 2.3481652368734465,
      "language_loss": 0.71362913,
      "learning_rate": 6.433569252661049e-07,
      "loss": 0.73229671,
      "num_input_tokens_seen": 133255590,
      "step": 6197,
      "time_per_iteration": 4.112748622894287
    },
    {
      "auxiliary_loss_clip": 0.01118376,
      "auxiliary_loss_mlp": 0.01023584,
      "balance_loss_clip": 1.04264414,
      "balance_loss_mlp": 1.01716161,
      "epoch": 0.7452654361810858,
      "flos": 12495405980160.0,
      "grad_norm": 1.7625281564161643,
      "language_loss": 0.71511972,
      "learning_rate": 6.427846669269952e-07,
      "loss": 0.73653924,
      "num_input_tokens_seen": 133273210,
      "step": 6198,
      "time_per_iteration": 2.5607643127441406
    },
    {
      "auxiliary_loss_clip": 0.01168595,
      "auxiliary_loss_mlp": 0.01028533,
      "balance_loss_clip": 1.05155122,
      "balance_loss_mlp": 1.02195609,
      "epoch": 0.7453856790717249,
      "flos": 22127329687680.0,
      "grad_norm": 1.9546354663939223,
      "language_loss": 0.82365346,
      "learning_rate": 6.422126144763729e-07,
      "loss": 0.84562474,
      "num_input_tokens_seen": 133292600,
      "step": 6199,
      "time_per_iteration": 2.470552444458008
    },
    {
      "auxiliary_loss_clip": 0.01121788,
      "auxiliary_loss_mlp": 0.00761184,
      "balance_loss_clip": 1.04064679,
      "balance_loss_mlp": 1.0002017,
      "epoch": 0.745505921962364,
      "flos": 20010682995840.0,
      "grad_norm": 2.0362318241114186,
      "language_loss": 0.77421319,
      "learning_rate": 6.416407680010174e-07,
      "loss": 0.7930429,
      "num_input_tokens_seen": 133306960,
      "step": 6200,
      "time_per_iteration": 2.556100368499756
    },
    {
      "auxiliary_loss_clip": 0.01118633,
      "auxiliary_loss_mlp": 0.01028224,
      "balance_loss_clip": 1.04396021,
      "balance_loss_mlp": 1.02143836,
      "epoch": 0.745626164853003,
      "flos": 24677884673280.0,
      "grad_norm": 2.0249996112742257,
      "language_loss": 0.80610418,
      "learning_rate": 6.410691275876774e-07,
      "loss": 0.82757282,
      "num_input_tokens_seen": 133326380,
      "step": 6201,
      "time_per_iteration": 2.618307590484619
    },
    {
      "auxiliary_loss_clip": 0.0114587,
      "auxiliary_loss_mlp": 0.01022999,
      "balance_loss_clip": 1.04893613,
      "balance_loss_mlp": 1.01635563,
      "epoch": 0.7457464077436422,
      "flos": 14538830797440.0,
      "grad_norm": 2.2401438240346288,
      "language_loss": 0.77095437,
      "learning_rate": 6.404976933230704e-07,
      "loss": 0.79264307,
      "num_input_tokens_seen": 133342900,
      "step": 6202,
      "time_per_iteration": 2.5525009632110596
    },
    {
      "auxiliary_loss_clip": 0.01143448,
      "auxiliary_loss_mlp": 0.01022944,
      "balance_loss_clip": 1.04619694,
      "balance_loss_mlp": 1.01566005,
      "epoch": 0.7458666506342813,
      "flos": 34021194600960.0,
      "grad_norm": 2.2203568004828798,
      "language_loss": 0.72475088,
      "learning_rate": 6.399264652938813e-07,
      "loss": 0.74641478,
      "num_input_tokens_seen": 133363805,
      "step": 6203,
      "time_per_iteration": 2.626080274581909
    },
    {
      "auxiliary_loss_clip": 0.01137248,
      "auxiliary_loss_mlp": 0.01027195,
      "balance_loss_clip": 1.04484487,
      "balance_loss_mlp": 1.02005136,
      "epoch": 0.7459868935249203,
      "flos": 24279025075200.0,
      "grad_norm": 1.91052677467262,
      "language_loss": 0.74709845,
      "learning_rate": 6.393554435867679e-07,
      "loss": 0.76874286,
      "num_input_tokens_seen": 133384655,
      "step": 6204,
      "time_per_iteration": 2.5494868755340576
    },
    {
      "auxiliary_loss_clip": 0.01119765,
      "auxiliary_loss_mlp": 0.01025691,
      "balance_loss_clip": 1.04212856,
      "balance_loss_mlp": 1.01838982,
      "epoch": 0.7461071364155595,
      "flos": 21908777385600.0,
      "grad_norm": 5.854153783758928,
      "language_loss": 0.83494306,
      "learning_rate": 6.387846282883502e-07,
      "loss": 0.85639763,
      "num_input_tokens_seen": 133401185,
      "step": 6205,
      "time_per_iteration": 2.580301284790039
    },
    {
      "auxiliary_loss_clip": 0.01163707,
      "auxiliary_loss_mlp": 0.01025772,
      "balance_loss_clip": 1.04860544,
      "balance_loss_mlp": 1.01872075,
      "epoch": 0.7462273793061985,
      "flos": 22889712879360.0,
      "grad_norm": 1.9888718236730922,
      "language_loss": 0.76922107,
      "learning_rate": 6.38214019485223e-07,
      "loss": 0.79111588,
      "num_input_tokens_seen": 133420010,
      "step": 6206,
      "time_per_iteration": 2.471585988998413
    },
    {
      "auxiliary_loss_clip": 0.0109036,
      "auxiliary_loss_mlp": 0.01024073,
      "balance_loss_clip": 1.03860855,
      "balance_loss_mlp": 1.01737094,
      "epoch": 0.7463476221968376,
      "flos": 19968451580160.0,
      "grad_norm": 1.7473812132094726,
      "language_loss": 0.71357512,
      "learning_rate": 6.376436172639461e-07,
      "loss": 0.73471951,
      "num_input_tokens_seen": 133437855,
      "step": 6207,
      "time_per_iteration": 2.598330497741699
    },
    {
      "auxiliary_loss_clip": 0.01081242,
      "auxiliary_loss_mlp": 0.0102782,
      "balance_loss_clip": 1.03997993,
      "balance_loss_mlp": 1.020226,
      "epoch": 0.7464678650874768,
      "flos": 16836610798080.0,
      "grad_norm": 2.964567765805751,
      "language_loss": 0.64868027,
      "learning_rate": 6.370734217110487e-07,
      "loss": 0.6697709,
      "num_input_tokens_seen": 133456600,
      "step": 6208,
      "time_per_iteration": 2.7120487689971924
    },
    {
      "auxiliary_loss_clip": 0.01142776,
      "auxiliary_loss_mlp": 0.01027763,
      "balance_loss_clip": 1.04701436,
      "balance_loss_mlp": 1.01978779,
      "epoch": 0.7465881079781158,
      "flos": 48100869843840.0,
      "grad_norm": 1.3741774626888217,
      "language_loss": 0.64222562,
      "learning_rate": 6.36503432913031e-07,
      "loss": 0.66393101,
      "num_input_tokens_seen": 133479745,
      "step": 6209,
      "time_per_iteration": 2.922898292541504
    },
    {
      "auxiliary_loss_clip": 0.01150229,
      "auxiliary_loss_mlp": 0.01028905,
      "balance_loss_clip": 1.04766035,
      "balance_loss_mlp": 1.02104867,
      "epoch": 0.7467083508687549,
      "flos": 19677359761920.0,
      "grad_norm": 3.9942417438002353,
      "language_loss": 0.68906212,
      "learning_rate": 6.359336509563569e-07,
      "loss": 0.71085346,
      "num_input_tokens_seen": 133495765,
      "step": 6210,
      "time_per_iteration": 3.223353385925293
    },
    {
      "auxiliary_loss_clip": 0.01109996,
      "auxiliary_loss_mlp": 0.01028084,
      "balance_loss_clip": 1.04169154,
      "balance_loss_mlp": 1.02106893,
      "epoch": 0.7468285937593939,
      "flos": 17895436934400.0,
      "grad_norm": 1.7070421499167272,
      "language_loss": 0.80409968,
      "learning_rate": 6.353640759274641e-07,
      "loss": 0.82548046,
      "num_input_tokens_seen": 133514655,
      "step": 6211,
      "time_per_iteration": 2.5277109146118164
    },
    {
      "auxiliary_loss_clip": 0.01148584,
      "auxiliary_loss_mlp": 0.01027856,
      "balance_loss_clip": 1.04336596,
      "balance_loss_mlp": 1.0207001,
      "epoch": 0.7469488366500331,
      "flos": 23141446369920.0,
      "grad_norm": 3.3349130226226418,
      "language_loss": 0.73958749,
      "learning_rate": 6.347947079127556e-07,
      "loss": 0.76135188,
      "num_input_tokens_seen": 133532555,
      "step": 6212,
      "time_per_iteration": 2.5230050086975098
    },
    {
      "auxiliary_loss_clip": 0.01132835,
      "auxiliary_loss_mlp": 0.0102861,
      "balance_loss_clip": 1.04498696,
      "balance_loss_mlp": 1.02149951,
      "epoch": 0.7470690795406721,
      "flos": 16690849407360.0,
      "grad_norm": 2.9075449048490563,
      "language_loss": 0.76228839,
      "learning_rate": 6.342255469986053e-07,
      "loss": 0.78390288,
      "num_input_tokens_seen": 133551300,
      "step": 6213,
      "time_per_iteration": 2.5390121936798096
    },
    {
      "auxiliary_loss_clip": 0.01164927,
      "auxiliary_loss_mlp": 0.01023461,
      "balance_loss_clip": 1.04870117,
      "balance_loss_mlp": 1.01661503,
      "epoch": 0.7471893224313112,
      "flos": 25192700352000.0,
      "grad_norm": 1.7048372786251287,
      "language_loss": 0.76533854,
      "learning_rate": 6.336565932713533e-07,
      "loss": 0.78722239,
      "num_input_tokens_seen": 133570725,
      "step": 6214,
      "time_per_iteration": 2.5210957527160645
    },
    {
      "auxiliary_loss_clip": 0.0113371,
      "auxiliary_loss_mlp": 0.01027818,
      "balance_loss_clip": 1.04558325,
      "balance_loss_mlp": 1.02040899,
      "epoch": 0.7473095653219504,
      "flos": 22526225199360.0,
      "grad_norm": 1.7392802217638244,
      "language_loss": 0.77739358,
      "learning_rate": 6.330878468173088e-07,
      "loss": 0.79900885,
      "num_input_tokens_seen": 133590790,
      "step": 6215,
      "time_per_iteration": 2.559904098510742
    },
    {
      "auxiliary_loss_clip": 0.01141581,
      "auxiliary_loss_mlp": 0.01026781,
      "balance_loss_clip": 1.04366863,
      "balance_loss_mlp": 1.01977694,
      "epoch": 0.7474298082125894,
      "flos": 18113989236480.0,
      "grad_norm": 1.700542724937735,
      "language_loss": 0.73017275,
      "learning_rate": 6.32519307722752e-07,
      "loss": 0.75185633,
      "num_input_tokens_seen": 133608685,
      "step": 6216,
      "time_per_iteration": 2.4790258407592773
    },
    {
      "auxiliary_loss_clip": 0.01022403,
      "auxiliary_loss_mlp": 0.01001973,
      "balance_loss_clip": 1.01054096,
      "balance_loss_mlp": 1.00092399,
      "epoch": 0.7475500511032285,
      "flos": 62086535193600.0,
      "grad_norm": 0.8994235518745616,
      "language_loss": 0.54997462,
      "learning_rate": 6.31950976073929e-07,
      "loss": 0.57021838,
      "num_input_tokens_seen": 133662775,
      "step": 6217,
      "time_per_iteration": 3.1623966693878174
    },
    {
      "auxiliary_loss_clip": 0.01104223,
      "auxiliary_loss_mlp": 0.01027959,
      "balance_loss_clip": 1.04217434,
      "balance_loss_mlp": 1.02008569,
      "epoch": 0.7476702939938676,
      "flos": 17785586165760.0,
      "grad_norm": 1.9995919926693717,
      "language_loss": 0.81019992,
      "learning_rate": 6.31382851957055e-07,
      "loss": 0.83152175,
      "num_input_tokens_seen": 133679595,
      "step": 6218,
      "time_per_iteration": 2.593344211578369
    },
    {
      "auxiliary_loss_clip": 0.01122455,
      "auxiliary_loss_mlp": 0.00760589,
      "balance_loss_clip": 1.04594731,
      "balance_loss_mlp": 1.00021648,
      "epoch": 0.7477905368845067,
      "flos": 27927944092800.0,
      "grad_norm": 1.9310954025710074,
      "language_loss": 0.71458554,
      "learning_rate": 6.308149354583143e-07,
      "loss": 0.73341602,
      "num_input_tokens_seen": 133699000,
      "step": 6219,
      "time_per_iteration": 2.6697282791137695
    },
    {
      "auxiliary_loss_clip": 0.01155813,
      "auxiliary_loss_mlp": 0.01031158,
      "balance_loss_clip": 1.04795444,
      "balance_loss_mlp": 1.02284956,
      "epoch": 0.7479107797751458,
      "flos": 26870374932480.0,
      "grad_norm": 1.8553642674760444,
      "language_loss": 0.81663883,
      "learning_rate": 6.302472266638586e-07,
      "loss": 0.83850861,
      "num_input_tokens_seen": 133719540,
      "step": 6220,
      "time_per_iteration": 2.5561609268188477
    },
    {
      "auxiliary_loss_clip": 0.01172719,
      "auxiliary_loss_mlp": 0.01031237,
      "balance_loss_clip": 1.05093002,
      "balance_loss_mlp": 1.02310753,
      "epoch": 0.7480310226657849,
      "flos": 33943375785600.0,
      "grad_norm": 2.5993052402991217,
      "language_loss": 0.70253623,
      "learning_rate": 6.296797256598101e-07,
      "loss": 0.72457576,
      "num_input_tokens_seen": 133741020,
      "step": 6221,
      "time_per_iteration": 2.600849151611328
    },
    {
      "auxiliary_loss_clip": 0.01114294,
      "auxiliary_loss_mlp": 0.01025195,
      "balance_loss_clip": 1.04201365,
      "balance_loss_mlp": 1.01836991,
      "epoch": 0.748151265556424,
      "flos": 24826555065600.0,
      "grad_norm": 1.6543087262106295,
      "language_loss": 0.814179,
      "learning_rate": 6.291124325322576e-07,
      "loss": 0.83557391,
      "num_input_tokens_seen": 133761145,
      "step": 6222,
      "time_per_iteration": 3.377903938293457
    },
    {
      "auxiliary_loss_clip": 0.01141815,
      "auxiliary_loss_mlp": 0.01023043,
      "balance_loss_clip": 1.04591477,
      "balance_loss_mlp": 1.01595628,
      "epoch": 0.748271508447063,
      "flos": 38399351535360.0,
      "grad_norm": 1.6001818472873717,
      "language_loss": 0.62480676,
      "learning_rate": 6.285453473672595e-07,
      "loss": 0.64645541,
      "num_input_tokens_seen": 133783715,
      "step": 6223,
      "time_per_iteration": 4.21563458442688
    },
    {
      "auxiliary_loss_clip": 0.01162535,
      "auxiliary_loss_mlp": 0.01026449,
      "balance_loss_clip": 1.04682112,
      "balance_loss_mlp": 1.01948714,
      "epoch": 0.7483917513377022,
      "flos": 21541842000000.0,
      "grad_norm": 1.9302349127005114,
      "language_loss": 0.7559973,
      "learning_rate": 6.279784702508415e-07,
      "loss": 0.77788711,
      "num_input_tokens_seen": 133804465,
      "step": 6224,
      "time_per_iteration": 2.4753615856170654
    },
    {
      "auxiliary_loss_clip": 0.0102601,
      "auxiliary_loss_mlp": 0.01001275,
      "balance_loss_clip": 1.00908518,
      "balance_loss_mlp": 1.00010681,
      "epoch": 0.7485119942283412,
      "flos": 62314532772480.0,
      "grad_norm": 0.7727059596762108,
      "language_loss": 0.58593297,
      "learning_rate": 6.274118012689979e-07,
      "loss": 0.60620582,
      "num_input_tokens_seen": 133866365,
      "step": 6225,
      "time_per_iteration": 3.24021577835083
    },
    {
      "auxiliary_loss_clip": 0.01129283,
      "auxiliary_loss_mlp": 0.01029064,
      "balance_loss_clip": 1.04390049,
      "balance_loss_mlp": 1.02111602,
      "epoch": 0.7486322371189803,
      "flos": 29937613104000.0,
      "grad_norm": 1.4448072906289855,
      "language_loss": 0.67984658,
      "learning_rate": 6.268453405076943e-07,
      "loss": 0.70143008,
      "num_input_tokens_seen": 133888760,
      "step": 6226,
      "time_per_iteration": 2.6113390922546387
    },
    {
      "auxiliary_loss_clip": 0.01138838,
      "auxiliary_loss_mlp": 0.01025389,
      "balance_loss_clip": 1.04516077,
      "balance_loss_mlp": 1.01911569,
      "epoch": 0.7487524800096195,
      "flos": 18949414734720.0,
      "grad_norm": 2.1166555255141954,
      "language_loss": 0.82033408,
      "learning_rate": 6.262790880528592e-07,
      "loss": 0.84197634,
      "num_input_tokens_seen": 133906380,
      "step": 6227,
      "time_per_iteration": 2.51828932762146
    },
    {
      "auxiliary_loss_clip": 0.01130789,
      "auxiliary_loss_mlp": 0.01026119,
      "balance_loss_clip": 1.04139209,
      "balance_loss_mlp": 1.0187701,
      "epoch": 0.7488727229002585,
      "flos": 18697393935360.0,
      "grad_norm": 2.372877871825076,
      "language_loss": 0.79803568,
      "learning_rate": 6.257130439903951e-07,
      "loss": 0.81960475,
      "num_input_tokens_seen": 133922875,
      "step": 6228,
      "time_per_iteration": 2.5349538326263428
    },
    {
      "auxiliary_loss_clip": 0.01169141,
      "auxiliary_loss_mlp": 0.01030798,
      "balance_loss_clip": 1.0506469,
      "balance_loss_mlp": 1.02367508,
      "epoch": 0.7489929657908976,
      "flos": 23623368168960.0,
      "grad_norm": 1.7849136989866308,
      "language_loss": 0.812433,
      "learning_rate": 6.251472084061695e-07,
      "loss": 0.83443236,
      "num_input_tokens_seen": 133941795,
      "step": 6229,
      "time_per_iteration": 2.4966917037963867
    },
    {
      "auxiliary_loss_clip": 0.01150643,
      "auxiliary_loss_mlp": 0.01027923,
      "balance_loss_clip": 1.04748654,
      "balance_loss_mlp": 1.02101469,
      "epoch": 0.7491132086815367,
      "flos": 20551533056640.0,
      "grad_norm": 4.99044604885606,
      "language_loss": 0.88620651,
      "learning_rate": 6.245815813860191e-07,
      "loss": 0.90799212,
      "num_input_tokens_seen": 133957305,
      "step": 6230,
      "time_per_iteration": 2.4961092472076416
    },
    {
      "auxiliary_loss_clip": 0.01167032,
      "auxiliary_loss_mlp": 0.0102541,
      "balance_loss_clip": 1.04863238,
      "balance_loss_mlp": 1.01759005,
      "epoch": 0.7492334515721758,
      "flos": 23003011353600.0,
      "grad_norm": 1.9230898194822925,
      "language_loss": 0.70347106,
      "learning_rate": 6.240161630157495e-07,
      "loss": 0.7253955,
      "num_input_tokens_seen": 133976660,
      "step": 6231,
      "time_per_iteration": 2.472421884536743
    },
    {
      "auxiliary_loss_clip": 0.01167459,
      "auxiliary_loss_mlp": 0.01029277,
      "balance_loss_clip": 1.04819429,
      "balance_loss_mlp": 1.02085495,
      "epoch": 0.7493536944628149,
      "flos": 16398823835520.0,
      "grad_norm": 2.54103356269205,
      "language_loss": 0.6970799,
      "learning_rate": 6.23450953381133e-07,
      "loss": 0.71904725,
      "num_input_tokens_seen": 133994750,
      "step": 6232,
      "time_per_iteration": 2.459921360015869
    },
    {
      "auxiliary_loss_clip": 0.01130608,
      "auxiliary_loss_mlp": 0.01021578,
      "balance_loss_clip": 1.04411805,
      "balance_loss_mlp": 1.01509261,
      "epoch": 0.749473937353454,
      "flos": 15338561155200.0,
      "grad_norm": 2.5526628539795913,
      "language_loss": 0.67994773,
      "learning_rate": 6.228859525679131e-07,
      "loss": 0.70146954,
      "num_input_tokens_seen": 134009165,
      "step": 6233,
      "time_per_iteration": 2.5213265419006348
    },
    {
      "auxiliary_loss_clip": 0.01153059,
      "auxiliary_loss_mlp": 0.01024283,
      "balance_loss_clip": 1.04804552,
      "balance_loss_mlp": 1.01781583,
      "epoch": 0.7495941802440931,
      "flos": 18951138587520.0,
      "grad_norm": 2.0380430648144823,
      "language_loss": 0.80231297,
      "learning_rate": 6.223211606617986e-07,
      "loss": 0.82408631,
      "num_input_tokens_seen": 134027585,
      "step": 6234,
      "time_per_iteration": 2.5344016551971436
    },
    {
      "auxiliary_loss_clip": 0.01150538,
      "auxiliary_loss_mlp": 0.01026749,
      "balance_loss_clip": 1.05020845,
      "balance_loss_mlp": 1.02028513,
      "epoch": 0.7497144231347321,
      "flos": 22492469393280.0,
      "grad_norm": 3.1885884685393346,
      "language_loss": 0.83864295,
      "learning_rate": 6.217565777484701e-07,
      "loss": 0.86041582,
      "num_input_tokens_seen": 134046680,
      "step": 6235,
      "time_per_iteration": 2.4919426441192627
    },
    {
      "auxiliary_loss_clip": 0.01137028,
      "auxiliary_loss_mlp": 0.0076013,
      "balance_loss_clip": 1.04636633,
      "balance_loss_mlp": 1.00022376,
      "epoch": 0.7498346660253713,
      "flos": 24243509502720.0,
      "grad_norm": 1.771204028295865,
      "language_loss": 0.80247426,
      "learning_rate": 6.211922039135722e-07,
      "loss": 0.8214457,
      "num_input_tokens_seen": 134066825,
      "step": 6236,
      "time_per_iteration": 3.285532236099243
    },
    {
      "auxiliary_loss_clip": 0.01166463,
      "auxiliary_loss_mlp": 0.01025736,
      "balance_loss_clip": 1.04947138,
      "balance_loss_mlp": 1.01903915,
      "epoch": 0.7499549089160104,
      "flos": 24387080163840.0,
      "grad_norm": 2.1068868092801103,
      "language_loss": 0.80884236,
      "learning_rate": 6.206280392427201e-07,
      "loss": 0.83076435,
      "num_input_tokens_seen": 134086410,
      "step": 6237,
      "time_per_iteration": 2.4761288166046143
    },
    {
      "auxiliary_loss_clip": 0.01143369,
      "auxiliary_loss_mlp": 0.01031811,
      "balance_loss_clip": 1.04351485,
      "balance_loss_mlp": 1.02446759,
      "epoch": 0.7500751518066494,
      "flos": 34057320704640.0,
      "grad_norm": 1.5091953904784017,
      "language_loss": 0.73846662,
      "learning_rate": 6.200640838214983e-07,
      "loss": 0.76021838,
      "num_input_tokens_seen": 134109185,
      "step": 6238,
      "time_per_iteration": 2.601134777069092
    },
    {
      "auxiliary_loss_clip": 0.01163226,
      "auxiliary_loss_mlp": 0.0102957,
      "balance_loss_clip": 1.04758584,
      "balance_loss_mlp": 1.02310848,
      "epoch": 0.7501953946972886,
      "flos": 18843586289280.0,
      "grad_norm": 1.9389816835684843,
      "language_loss": 0.67235982,
      "learning_rate": 6.195003377354578e-07,
      "loss": 0.69428772,
      "num_input_tokens_seen": 134128455,
      "step": 6239,
      "time_per_iteration": 2.4516124725341797
    },
    {
      "auxiliary_loss_clip": 0.0114817,
      "auxiliary_loss_mlp": 0.0102854,
      "balance_loss_clip": 1.04513168,
      "balance_loss_mlp": 1.02085066,
      "epoch": 0.7503156375879276,
      "flos": 20257675891200.0,
      "grad_norm": 2.638826809105794,
      "language_loss": 0.7319572,
      "learning_rate": 6.189368010701183e-07,
      "loss": 0.75372434,
      "num_input_tokens_seen": 134145515,
      "step": 6240,
      "time_per_iteration": 2.499545097351074
    },
    {
      "auxiliary_loss_clip": 0.01154288,
      "auxiliary_loss_mlp": 0.01021391,
      "balance_loss_clip": 1.04437017,
      "balance_loss_mlp": 1.01459289,
      "epoch": 0.7504358804785667,
      "flos": 13480040574720.0,
      "grad_norm": 1.8034852588708608,
      "language_loss": 0.76735997,
      "learning_rate": 6.183734739109683e-07,
      "loss": 0.78911674,
      "num_input_tokens_seen": 134163335,
      "step": 6241,
      "time_per_iteration": 2.467649459838867
    },
    {
      "auxiliary_loss_clip": 0.01156224,
      "auxiliary_loss_mlp": 0.01029195,
      "balance_loss_clip": 1.04915023,
      "balance_loss_mlp": 1.0214107,
      "epoch": 0.7505561233692057,
      "flos": 29461042431360.0,
      "grad_norm": 1.9215701819878643,
      "language_loss": 0.68897903,
      "learning_rate": 6.178103563434629e-07,
      "loss": 0.71083319,
      "num_input_tokens_seen": 134182335,
      "step": 6242,
      "time_per_iteration": 2.5358595848083496
    },
    {
      "auxiliary_loss_clip": 0.0116552,
      "auxiliary_loss_mlp": 0.01025141,
      "balance_loss_clip": 1.04878569,
      "balance_loss_mlp": 1.01825666,
      "epoch": 0.7506763662598449,
      "flos": 20302457172480.0,
      "grad_norm": 1.7171033807334242,
      "language_loss": 0.83774853,
      "learning_rate": 6.172474484530283e-07,
      "loss": 0.85965514,
      "num_input_tokens_seen": 134201070,
      "step": 6243,
      "time_per_iteration": 2.47774076461792
    },
    {
      "auxiliary_loss_clip": 0.01125085,
      "auxiliary_loss_mlp": 0.01026248,
      "balance_loss_clip": 1.04063988,
      "balance_loss_mlp": 1.01869583,
      "epoch": 0.750796609150484,
      "flos": 37230961939200.0,
      "grad_norm": 1.602233524873449,
      "language_loss": 0.75927943,
      "learning_rate": 6.166847503250563e-07,
      "loss": 0.78079277,
      "num_input_tokens_seen": 134223310,
      "step": 6244,
      "time_per_iteration": 2.6686129570007324
    },
    {
      "auxiliary_loss_clip": 0.01138057,
      "auxiliary_loss_mlp": 0.01027409,
      "balance_loss_clip": 1.04449058,
      "balance_loss_mlp": 1.02070689,
      "epoch": 0.750916852041123,
      "flos": 19609417186560.0,
      "grad_norm": 2.2627598009922743,
      "language_loss": 0.79118818,
      "learning_rate": 6.161222620449078e-07,
      "loss": 0.81284285,
      "num_input_tokens_seen": 134242085,
      "step": 6245,
      "time_per_iteration": 2.528900146484375
    },
    {
      "auxiliary_loss_clip": 0.01127078,
      "auxiliary_loss_mlp": 0.01030012,
      "balance_loss_clip": 1.04624796,
      "balance_loss_mlp": 1.02321458,
      "epoch": 0.7510370949317622,
      "flos": 25112690807040.0,
      "grad_norm": 2.2601726393376462,
      "language_loss": 0.80108911,
      "learning_rate": 6.155599836979117e-07,
      "loss": 0.82266003,
      "num_input_tokens_seen": 134260770,
      "step": 6246,
      "time_per_iteration": 2.597461223602295
    },
    {
      "auxiliary_loss_clip": 0.01111689,
      "auxiliary_loss_mlp": 0.01026548,
      "balance_loss_clip": 1.04207587,
      "balance_loss_mlp": 1.01945829,
      "epoch": 0.7511573378224012,
      "flos": 19062282245760.0,
      "grad_norm": 1.9058818973385154,
      "language_loss": 0.8140046,
      "learning_rate": 6.149979153693649e-07,
      "loss": 0.83538699,
      "num_input_tokens_seen": 134278025,
      "step": 6247,
      "time_per_iteration": 2.5822956562042236
    },
    {
      "auxiliary_loss_clip": 0.01146268,
      "auxiliary_loss_mlp": 0.01026444,
      "balance_loss_clip": 1.04459655,
      "balance_loss_mlp": 1.01924956,
      "epoch": 0.7512775807130403,
      "flos": 19937676602880.0,
      "grad_norm": 1.842419003075821,
      "language_loss": 0.76897597,
      "learning_rate": 6.144360571445343e-07,
      "loss": 0.79070312,
      "num_input_tokens_seen": 134297170,
      "step": 6248,
      "time_per_iteration": 3.279876232147217
    },
    {
      "auxiliary_loss_clip": 0.01148575,
      "auxiliary_loss_mlp": 0.01024347,
      "balance_loss_clip": 1.04829168,
      "balance_loss_mlp": 1.01716208,
      "epoch": 0.7513978236036795,
      "flos": 20739920912640.0,
      "grad_norm": 1.6579890181522157,
      "language_loss": 0.79871202,
      "learning_rate": 6.138744091086509e-07,
      "loss": 0.82044125,
      "num_input_tokens_seen": 134316755,
      "step": 6249,
      "time_per_iteration": 4.070315361022949
    },
    {
      "auxiliary_loss_clip": 0.01125105,
      "auxiliary_loss_mlp": 0.01025544,
      "balance_loss_clip": 1.04602599,
      "balance_loss_mlp": 1.01896119,
      "epoch": 0.7515180664943185,
      "flos": 27563163523200.0,
      "grad_norm": 3.1489178011242736,
      "language_loss": 0.72777104,
      "learning_rate": 6.133129713469183e-07,
      "loss": 0.74927759,
      "num_input_tokens_seen": 134335960,
      "step": 6250,
      "time_per_iteration": 2.6243326663970947
    },
    {
      "auxiliary_loss_clip": 0.01132516,
      "auxiliary_loss_mlp": 0.01023806,
      "balance_loss_clip": 1.04316998,
      "balance_loss_mlp": 1.01701427,
      "epoch": 0.7516383093849576,
      "flos": 33803181002880.0,
      "grad_norm": 1.8011596404267367,
      "language_loss": 0.6419872,
      "learning_rate": 6.127517439445053e-07,
      "loss": 0.66355044,
      "num_input_tokens_seen": 134356805,
      "step": 6251,
      "time_per_iteration": 2.697223663330078
    },
    {
      "auxiliary_loss_clip": 0.01100254,
      "auxiliary_loss_mlp": 0.01028183,
      "balance_loss_clip": 1.04193938,
      "balance_loss_mlp": 1.02183175,
      "epoch": 0.7517585522755967,
      "flos": 29746172592000.0,
      "grad_norm": 1.9460801815966915,
      "language_loss": 0.81582463,
      "learning_rate": 6.121907269865498e-07,
      "loss": 0.83710903,
      "num_input_tokens_seen": 134376295,
      "step": 6252,
      "time_per_iteration": 2.64725923538208
    },
    {
      "auxiliary_loss_clip": 0.01023803,
      "auxiliary_loss_mlp": 0.01000779,
      "balance_loss_clip": 1.00918746,
      "balance_loss_mlp": 0.99981368,
      "epoch": 0.7518787951662358,
      "flos": 69807974319360.0,
      "grad_norm": 0.9244843124448847,
      "language_loss": 0.67310929,
      "learning_rate": 6.116299205581577e-07,
      "loss": 0.6933552,
      "num_input_tokens_seen": 134431125,
      "step": 6253,
      "time_per_iteration": 3.1114208698272705
    },
    {
      "auxiliary_loss_clip": 0.01170699,
      "auxiliary_loss_mlp": 0.01026656,
      "balance_loss_clip": 1.05055535,
      "balance_loss_mlp": 1.01883006,
      "epoch": 0.7519990380568748,
      "flos": 34203225749760.0,
      "grad_norm": 2.126519314132762,
      "language_loss": 0.68759048,
      "learning_rate": 6.110693247444018e-07,
      "loss": 0.70956409,
      "num_input_tokens_seen": 134452960,
      "step": 6254,
      "time_per_iteration": 2.6071577072143555
    },
    {
      "auxiliary_loss_clip": 0.01111513,
      "auxiliary_loss_mlp": 0.01022226,
      "balance_loss_clip": 1.04238677,
      "balance_loss_mlp": 1.01600647,
      "epoch": 0.752119280947514,
      "flos": 21725704742400.0,
      "grad_norm": 1.7800000375566,
      "language_loss": 0.82469749,
      "learning_rate": 6.105089396303258e-07,
      "loss": 0.84603488,
      "num_input_tokens_seen": 134471350,
      "step": 6255,
      "time_per_iteration": 2.553864002227783
    },
    {
      "auxiliary_loss_clip": 0.0113732,
      "auxiliary_loss_mlp": 0.01029137,
      "balance_loss_clip": 1.04374504,
      "balance_loss_mlp": 1.02211547,
      "epoch": 0.7522395238381531,
      "flos": 32742774668160.0,
      "grad_norm": 1.8285768815249757,
      "language_loss": 0.75537407,
      "learning_rate": 6.099487653009383e-07,
      "loss": 0.77703863,
      "num_input_tokens_seen": 134490695,
      "step": 6256,
      "time_per_iteration": 2.6359426975250244
    },
    {
      "auxiliary_loss_clip": 0.01148764,
      "auxiliary_loss_mlp": 0.01025441,
      "balance_loss_clip": 1.04555893,
      "balance_loss_mlp": 1.01880443,
      "epoch": 0.7523597667287921,
      "flos": 23476026579840.0,
      "grad_norm": 1.9740247538150926,
      "language_loss": 0.83276796,
      "learning_rate": 6.093888018412192e-07,
      "loss": 0.85451007,
      "num_input_tokens_seen": 134506885,
      "step": 6257,
      "time_per_iteration": 2.5164895057678223
    },
    {
      "auxiliary_loss_clip": 0.01048034,
      "auxiliary_loss_mlp": 0.01001202,
      "balance_loss_clip": 1.00867832,
      "balance_loss_mlp": 1.00009966,
      "epoch": 0.7524800096194313,
      "flos": 67346730501120.0,
      "grad_norm": 0.7095707323809946,
      "language_loss": 0.54666471,
      "learning_rate": 6.088290493361125e-07,
      "loss": 0.56715715,
      "num_input_tokens_seen": 134571770,
      "step": 6258,
      "time_per_iteration": 3.249915599822998
    },
    {
      "auxiliary_loss_clip": 0.01098488,
      "auxiliary_loss_mlp": 0.01028599,
      "balance_loss_clip": 1.04110265,
      "balance_loss_mlp": 1.02133632,
      "epoch": 0.7526002525100703,
      "flos": 13006055681280.0,
      "grad_norm": 2.0360812768208953,
      "language_loss": 0.71430731,
      "learning_rate": 6.082695078705322e-07,
      "loss": 0.73557818,
      "num_input_tokens_seen": 134589250,
      "step": 6259,
      "time_per_iteration": 2.5667903423309326
    },
    {
      "auxiliary_loss_clip": 0.01145227,
      "auxiliary_loss_mlp": 0.01028798,
      "balance_loss_clip": 1.04615259,
      "balance_loss_mlp": 1.0215323,
      "epoch": 0.7527204954007094,
      "flos": 21397229844480.0,
      "grad_norm": 2.1002009197052955,
      "language_loss": 0.68446505,
      "learning_rate": 6.077101775293618e-07,
      "loss": 0.70620525,
      "num_input_tokens_seen": 134608075,
      "step": 6260,
      "time_per_iteration": 2.511019706726074
    },
    {
      "auxiliary_loss_clip": 0.01151298,
      "auxiliary_loss_mlp": 0.01025245,
      "balance_loss_clip": 1.04554451,
      "balance_loss_mlp": 1.01756787,
      "epoch": 0.7528407382913486,
      "flos": 18947188091520.0,
      "grad_norm": 2.3173733357821527,
      "language_loss": 0.82673168,
      "learning_rate": 6.071510583974504e-07,
      "loss": 0.84849709,
      "num_input_tokens_seen": 134623260,
      "step": 6261,
      "time_per_iteration": 2.4617974758148193
    },
    {
      "auxiliary_loss_clip": 0.01166716,
      "auxiliary_loss_mlp": 0.01030779,
      "balance_loss_clip": 1.04835665,
      "balance_loss_mlp": 1.02344429,
      "epoch": 0.7529609811819876,
      "flos": 15231798956160.0,
      "grad_norm": 1.8979817014794291,
      "language_loss": 0.7191686,
      "learning_rate": 6.065921505596161e-07,
      "loss": 0.74114358,
      "num_input_tokens_seen": 134641540,
      "step": 6262,
      "time_per_iteration": 3.129119634628296
    },
    {
      "auxiliary_loss_clip": 0.01121149,
      "auxiliary_loss_mlp": 0.01024487,
      "balance_loss_clip": 1.04439378,
      "balance_loss_mlp": 1.01780558,
      "epoch": 0.7530812240726267,
      "flos": 19354487385600.0,
      "grad_norm": 1.7617995249140659,
      "language_loss": 0.76778251,
      "learning_rate": 6.060334541006445e-07,
      "loss": 0.78923887,
      "num_input_tokens_seen": 134660035,
      "step": 6263,
      "time_per_iteration": 2.5495619773864746
    },
    {
      "auxiliary_loss_clip": 0.01123816,
      "auxiliary_loss_mlp": 0.01029191,
      "balance_loss_clip": 1.03987014,
      "balance_loss_mlp": 1.02245557,
      "epoch": 0.7532014669632658,
      "flos": 27748247328000.0,
      "grad_norm": 1.6351522340215003,
      "language_loss": 0.68991965,
      "learning_rate": 6.05474969105289e-07,
      "loss": 0.71144974,
      "num_input_tokens_seen": 134683025,
      "step": 6264,
      "time_per_iteration": 2.6634681224823
    },
    {
      "auxiliary_loss_clip": 0.01151708,
      "auxiliary_loss_mlp": 0.01024727,
      "balance_loss_clip": 1.04756951,
      "balance_loss_mlp": 1.01759219,
      "epoch": 0.7533217098539049,
      "flos": 14137421333760.0,
      "grad_norm": 2.033676205793127,
      "language_loss": 0.73727804,
      "learning_rate": 6.049166956582725e-07,
      "loss": 0.75904238,
      "num_input_tokens_seen": 134701290,
      "step": 6265,
      "time_per_iteration": 2.492546558380127
    },
    {
      "auxiliary_loss_clip": 0.0114851,
      "auxiliary_loss_mlp": 0.01024319,
      "balance_loss_clip": 1.04667783,
      "balance_loss_mlp": 1.01783097,
      "epoch": 0.753441952744544,
      "flos": 26429068437120.0,
      "grad_norm": 3.5740809725565783,
      "language_loss": 0.87490135,
      "learning_rate": 6.043586338442841e-07,
      "loss": 0.89662975,
      "num_input_tokens_seen": 134720345,
      "step": 6266,
      "time_per_iteration": 2.528953790664673
    },
    {
      "auxiliary_loss_clip": 0.01163866,
      "auxiliary_loss_mlp": 0.01024442,
      "balance_loss_clip": 1.04995918,
      "balance_loss_mlp": 1.01829064,
      "epoch": 0.7535621956351831,
      "flos": 23878621192320.0,
      "grad_norm": 1.7504916463998936,
      "language_loss": 0.72972959,
      "learning_rate": 6.038007837479815e-07,
      "loss": 0.75161266,
      "num_input_tokens_seen": 134741450,
      "step": 6267,
      "time_per_iteration": 2.496040105819702
    },
    {
      "auxiliary_loss_clip": 0.01149559,
      "auxiliary_loss_mlp": 0.01023252,
      "balance_loss_clip": 1.04767919,
      "balance_loss_mlp": 1.01642752,
      "epoch": 0.7536824385258222,
      "flos": 21795873960960.0,
      "grad_norm": 1.879179788386603,
      "language_loss": 0.64003956,
      "learning_rate": 6.032431454539897e-07,
      "loss": 0.66176772,
      "num_input_tokens_seen": 134760295,
      "step": 6268,
      "time_per_iteration": 2.505577564239502
    },
    {
      "auxiliary_loss_clip": 0.01124689,
      "auxiliary_loss_mlp": 0.01025629,
      "balance_loss_clip": 1.04388583,
      "balance_loss_mlp": 1.01900733,
      "epoch": 0.7538026814164612,
      "flos": 28911644933760.0,
      "grad_norm": 1.7480969815319523,
      "language_loss": 0.81738132,
      "learning_rate": 6.026857190469014e-07,
      "loss": 0.83888447,
      "num_input_tokens_seen": 134782050,
      "step": 6269,
      "time_per_iteration": 2.6259500980377197
    },
    {
      "auxiliary_loss_clip": 0.01138607,
      "auxiliary_loss_mlp": 0.01024107,
      "balance_loss_clip": 1.04597068,
      "balance_loss_mlp": 1.01667178,
      "epoch": 0.7539229243071004,
      "flos": 21104701482240.0,
      "grad_norm": 1.8454125871671823,
      "language_loss": 0.74463296,
      "learning_rate": 6.0212850461128e-07,
      "loss": 0.76626015,
      "num_input_tokens_seen": 134801170,
      "step": 6270,
      "time_per_iteration": 2.5406906604766846
    },
    {
      "auxiliary_loss_clip": 0.01138766,
      "auxiliary_loss_mlp": 0.01023863,
      "balance_loss_clip": 1.04377365,
      "balance_loss_mlp": 1.01648974,
      "epoch": 0.7540431671977395,
      "flos": 15158469340800.0,
      "grad_norm": 3.1456981106703363,
      "language_loss": 0.74921876,
      "learning_rate": 6.015715022316516e-07,
      "loss": 0.77084506,
      "num_input_tokens_seen": 134819150,
      "step": 6271,
      "time_per_iteration": 2.524465799331665
    },
    {
      "auxiliary_loss_clip": 0.01108246,
      "auxiliary_loss_mlp": 0.01020818,
      "balance_loss_clip": 1.04115987,
      "balance_loss_mlp": 1.01390946,
      "epoch": 0.7541634100883785,
      "flos": 18770579896320.0,
      "grad_norm": 2.54667103058873,
      "language_loss": 0.77206051,
      "learning_rate": 6.010147119925154e-07,
      "loss": 0.79335117,
      "num_input_tokens_seen": 134836905,
      "step": 6272,
      "time_per_iteration": 2.5998222827911377
    },
    {
      "auxiliary_loss_clip": 0.0111202,
      "auxiliary_loss_mlp": 0.01027199,
      "balance_loss_clip": 1.0422157,
      "balance_loss_mlp": 1.02020085,
      "epoch": 0.7542836529790176,
      "flos": 20594770053120.0,
      "grad_norm": 1.8524596076190376,
      "language_loss": 0.65988791,
      "learning_rate": 6.004581339783348e-07,
      "loss": 0.68128014,
      "num_input_tokens_seen": 134855225,
      "step": 6273,
      "time_per_iteration": 3.36283278465271
    },
    {
      "auxiliary_loss_clip": 0.01156403,
      "auxiliary_loss_mlp": 0.01032928,
      "balance_loss_clip": 1.04817319,
      "balance_loss_mlp": 1.02449954,
      "epoch": 0.7544038958696567,
      "flos": 19095104298240.0,
      "grad_norm": 4.1297571692097845,
      "language_loss": 0.68691266,
      "learning_rate": 5.999017682735425e-07,
      "loss": 0.70880598,
      "num_input_tokens_seen": 134871615,
      "step": 6274,
      "time_per_iteration": 2.4745991230010986
    },
    {
      "auxiliary_loss_clip": 0.01102501,
      "auxiliary_loss_mlp": 0.01027725,
      "balance_loss_clip": 1.04162598,
      "balance_loss_mlp": 1.02046561,
      "epoch": 0.7545241387602958,
      "flos": 31723306859520.0,
      "grad_norm": 1.898925435082557,
      "language_loss": 0.66423893,
      "learning_rate": 5.993456149625387e-07,
      "loss": 0.68554127,
      "num_input_tokens_seen": 134892765,
      "step": 6275,
      "time_per_iteration": 4.276025295257568
    },
    {
      "auxiliary_loss_clip": 0.01114641,
      "auxiliary_loss_mlp": 0.01028079,
      "balance_loss_clip": 1.04306865,
      "balance_loss_mlp": 1.02129936,
      "epoch": 0.7546443816509348,
      "flos": 20296495514880.0,
      "grad_norm": 1.734874114649532,
      "language_loss": 0.82523203,
      "learning_rate": 5.987896741296909e-07,
      "loss": 0.84665924,
      "num_input_tokens_seen": 134910505,
      "step": 6276,
      "time_per_iteration": 2.572920799255371
    },
    {
      "auxiliary_loss_clip": 0.01138407,
      "auxiliary_loss_mlp": 0.01026371,
      "balance_loss_clip": 1.04792833,
      "balance_loss_mlp": 1.01945651,
      "epoch": 0.754764624541574,
      "flos": 23696159080320.0,
      "grad_norm": 2.0298378408724207,
      "language_loss": 0.77934515,
      "learning_rate": 5.982339458593361e-07,
      "loss": 0.80099297,
      "num_input_tokens_seen": 134930445,
      "step": 6277,
      "time_per_iteration": 2.554483652114868
    },
    {
      "auxiliary_loss_clip": 0.01147542,
      "auxiliary_loss_mlp": 0.00760305,
      "balance_loss_clip": 1.04771543,
      "balance_loss_mlp": 1.00026536,
      "epoch": 0.7548848674322131,
      "flos": 25337204766720.0,
      "grad_norm": 1.6260373328700288,
      "language_loss": 0.84002388,
      "learning_rate": 5.976784302357767e-07,
      "loss": 0.85910237,
      "num_input_tokens_seen": 134951010,
      "step": 6278,
      "time_per_iteration": 2.559769630432129
    },
    {
      "auxiliary_loss_clip": 0.01154288,
      "auxiliary_loss_mlp": 0.01027358,
      "balance_loss_clip": 1.04864573,
      "balance_loss_mlp": 1.02060437,
      "epoch": 0.7550051103228521,
      "flos": 19573147428480.0,
      "grad_norm": 1.7530932220885231,
      "language_loss": 0.73919082,
      "learning_rate": 5.971231273432855e-07,
      "loss": 0.76100731,
      "num_input_tokens_seen": 134970495,
      "step": 6279,
      "time_per_iteration": 2.489847183227539
    },
    {
      "auxiliary_loss_clip": 0.01049335,
      "auxiliary_loss_mlp": 0.01002518,
      "balance_loss_clip": 1.01093113,
      "balance_loss_mlp": 1.00140309,
      "epoch": 0.7551253532134913,
      "flos": 64150068648960.0,
      "grad_norm": 0.8178661607288347,
      "language_loss": 0.54593796,
      "learning_rate": 5.965680372661e-07,
      "loss": 0.56645644,
      "num_input_tokens_seen": 135028060,
      "step": 6280,
      "time_per_iteration": 3.0285227298736572
    },
    {
      "auxiliary_loss_clip": 0.01138612,
      "auxiliary_loss_mlp": 0.01019769,
      "balance_loss_clip": 1.04817033,
      "balance_loss_mlp": 1.01345956,
      "epoch": 0.7552455961041303,
      "flos": 26067986968320.0,
      "grad_norm": 1.7293891092475546,
      "language_loss": 0.56356275,
      "learning_rate": 5.960131600884266e-07,
      "loss": 0.58514655,
      "num_input_tokens_seen": 135047330,
      "step": 6281,
      "time_per_iteration": 2.5642974376678467
    },
    {
      "auxiliary_loss_clip": 0.0112426,
      "auxiliary_loss_mlp": 0.01022194,
      "balance_loss_clip": 1.04350984,
      "balance_loss_mlp": 1.01607251,
      "epoch": 0.7553658389947694,
      "flos": 24498223822080.0,
      "grad_norm": 1.6271385526730684,
      "language_loss": 0.76204622,
      "learning_rate": 5.954584958944413e-07,
      "loss": 0.78351074,
      "num_input_tokens_seen": 135065995,
      "step": 6282,
      "time_per_iteration": 2.6135056018829346
    },
    {
      "auxiliary_loss_clip": 0.01124877,
      "auxiliary_loss_mlp": 0.00760481,
      "balance_loss_clip": 1.0423578,
      "balance_loss_mlp": 1.00022769,
      "epoch": 0.7554860818854086,
      "flos": 21799465320960.0,
      "grad_norm": 1.9333301126501194,
      "language_loss": 0.81627566,
      "learning_rate": 5.949040447682854e-07,
      "loss": 0.83512926,
      "num_input_tokens_seen": 135085820,
      "step": 6283,
      "time_per_iteration": 2.598297595977783
    },
    {
      "auxiliary_loss_clip": 0.01142553,
      "auxiliary_loss_mlp": 0.01024692,
      "balance_loss_clip": 1.04554546,
      "balance_loss_mlp": 1.01758111,
      "epoch": 0.7556063247760476,
      "flos": 16362123114240.0,
      "grad_norm": 2.6332056672184128,
      "language_loss": 0.6866976,
      "learning_rate": 5.943498067940686e-07,
      "loss": 0.70837009,
      "num_input_tokens_seen": 135102845,
      "step": 6284,
      "time_per_iteration": 2.5193393230438232
    },
    {
      "auxiliary_loss_clip": 0.01128036,
      "auxiliary_loss_mlp": 0.01030451,
      "balance_loss_clip": 1.04693544,
      "balance_loss_mlp": 1.02374232,
      "epoch": 0.7557265676666867,
      "flos": 27235155502080.0,
      "grad_norm": 1.794549547219913,
      "language_loss": 0.81447613,
      "learning_rate": 5.937957820558686e-07,
      "loss": 0.836061,
      "num_input_tokens_seen": 135122190,
      "step": 6285,
      "time_per_iteration": 2.567260265350342
    },
    {
      "auxiliary_loss_clip": 0.01033864,
      "auxiliary_loss_mlp": 0.01000616,
      "balance_loss_clip": 1.00808001,
      "balance_loss_mlp": 0.99950695,
      "epoch": 0.7558468105573258,
      "flos": 62189131415040.0,
      "grad_norm": 0.8472660789058853,
      "language_loss": 0.65450746,
      "learning_rate": 5.932419706377296e-07,
      "loss": 0.67485225,
      "num_input_tokens_seen": 135180495,
      "step": 6286,
      "time_per_iteration": 3.076275587081909
    },
    {
      "auxiliary_loss_clip": 0.01120809,
      "auxiliary_loss_mlp": 0.01027472,
      "balance_loss_clip": 1.04768693,
      "balance_loss_mlp": 1.02019691,
      "epoch": 0.7559670534479649,
      "flos": 33249078823680.0,
      "grad_norm": 1.9585203073361095,
      "language_loss": 0.74198288,
      "learning_rate": 5.92688372623666e-07,
      "loss": 0.76346564,
      "num_input_tokens_seen": 135199200,
      "step": 6287,
      "time_per_iteration": 3.4155259132385254
    },
    {
      "auxiliary_loss_clip": 0.0115019,
      "auxiliary_loss_mlp": 0.01023002,
      "balance_loss_clip": 1.04473758,
      "balance_loss_mlp": 1.01528311,
      "epoch": 0.7560872963386039,
      "flos": 14064379027200.0,
      "grad_norm": 1.9530731473118517,
      "language_loss": 0.74086165,
      "learning_rate": 5.921349880976574e-07,
      "loss": 0.76259363,
      "num_input_tokens_seen": 135217035,
      "step": 6288,
      "time_per_iteration": 2.473651170730591
    },
    {
      "auxiliary_loss_clip": 0.01138407,
      "auxiliary_loss_mlp": 0.00760438,
      "balance_loss_clip": 1.04275894,
      "balance_loss_mlp": 1.00024307,
      "epoch": 0.7562075392292431,
      "flos": 20412307941120.0,
      "grad_norm": 2.5936671517322982,
      "language_loss": 0.8199861,
      "learning_rate": 5.915818171436515e-07,
      "loss": 0.83897454,
      "num_input_tokens_seen": 135236370,
      "step": 6289,
      "time_per_iteration": 2.5446269512176514
    },
    {
      "auxiliary_loss_clip": 0.01133963,
      "auxiliary_loss_mlp": 0.01028733,
      "balance_loss_clip": 1.04004383,
      "balance_loss_mlp": 1.02143705,
      "epoch": 0.7563277821198822,
      "flos": 20376792368640.0,
      "grad_norm": 1.6621966362084493,
      "language_loss": 0.7503475,
      "learning_rate": 5.910288598455642e-07,
      "loss": 0.77197438,
      "num_input_tokens_seen": 135255720,
      "step": 6290,
      "time_per_iteration": 2.520799398422241
    },
    {
      "auxiliary_loss_clip": 0.01156047,
      "auxiliary_loss_mlp": 0.01025518,
      "balance_loss_clip": 1.04614234,
      "balance_loss_mlp": 1.01775742,
      "epoch": 0.7564480250105212,
      "flos": 18588261438720.0,
      "grad_norm": 2.2020485704627792,
      "language_loss": 0.74840701,
      "learning_rate": 5.90476116287278e-07,
      "loss": 0.77022266,
      "num_input_tokens_seen": 135273320,
      "step": 6291,
      "time_per_iteration": 2.4727063179016113
    },
    {
      "auxiliary_loss_clip": 0.0113926,
      "auxiliary_loss_mlp": 0.01026294,
      "balance_loss_clip": 1.04804504,
      "balance_loss_mlp": 1.01927876,
      "epoch": 0.7565682679011604,
      "flos": 21215521918080.0,
      "grad_norm": 1.8818925405281577,
      "language_loss": 0.68192637,
      "learning_rate": 5.899235865526456e-07,
      "loss": 0.70358193,
      "num_input_tokens_seen": 135292615,
      "step": 6292,
      "time_per_iteration": 2.5118982791900635
    },
    {
      "auxiliary_loss_clip": 0.01115276,
      "auxiliary_loss_mlp": 0.01022857,
      "balance_loss_clip": 1.04299402,
      "balance_loss_mlp": 1.01616025,
      "epoch": 0.7566885107917994,
      "flos": 20449008662400.0,
      "grad_norm": 1.6548226669149304,
      "language_loss": 0.82116491,
      "learning_rate": 5.893712707254825e-07,
      "loss": 0.84254622,
      "num_input_tokens_seen": 135310075,
      "step": 6293,
      "time_per_iteration": 2.5548954010009766
    },
    {
      "auxiliary_loss_clip": 0.0110315,
      "auxiliary_loss_mlp": 0.01026026,
      "balance_loss_clip": 1.0386796,
      "balance_loss_mlp": 1.01920748,
      "epoch": 0.7568087536824385,
      "flos": 19025832919680.0,
      "grad_norm": 2.8261807469211075,
      "language_loss": 0.66233253,
      "learning_rate": 5.888191688895769e-07,
      "loss": 0.68362433,
      "num_input_tokens_seen": 135327335,
      "step": 6294,
      "time_per_iteration": 2.5610146522521973
    },
    {
      "auxiliary_loss_clip": 0.01163521,
      "auxiliary_loss_mlp": 0.01021311,
      "balance_loss_clip": 1.04661083,
      "balance_loss_mlp": 1.01377964,
      "epoch": 0.7569289965730777,
      "flos": 15225442248960.0,
      "grad_norm": 1.962535535504141,
      "language_loss": 0.61988747,
      "learning_rate": 5.882672811286813e-07,
      "loss": 0.64173579,
      "num_input_tokens_seen": 135343615,
      "step": 6295,
      "time_per_iteration": 2.4576311111450195
    },
    {
      "auxiliary_loss_clip": 0.01165076,
      "auxiliary_loss_mlp": 0.01026138,
      "balance_loss_clip": 1.04708529,
      "balance_loss_mlp": 1.01926279,
      "epoch": 0.7570492394637167,
      "flos": 20769367086720.0,
      "grad_norm": 2.104774822748368,
      "language_loss": 0.69502455,
      "learning_rate": 5.877156075265166e-07,
      "loss": 0.71693665,
      "num_input_tokens_seen": 135359880,
      "step": 6296,
      "time_per_iteration": 2.4464166164398193
    },
    {
      "auxiliary_loss_clip": 0.01133816,
      "auxiliary_loss_mlp": 0.01024606,
      "balance_loss_clip": 1.04246712,
      "balance_loss_mlp": 1.01738501,
      "epoch": 0.7571694823543558,
      "flos": 15664091137920.0,
      "grad_norm": 2.508183112578675,
      "language_loss": 0.69937485,
      "learning_rate": 5.871641481667715e-07,
      "loss": 0.72095907,
      "num_input_tokens_seen": 135374325,
      "step": 6297,
      "time_per_iteration": 2.4712486267089844
    },
    {
      "auxiliary_loss_clip": 0.01113823,
      "auxiliary_loss_mlp": 0.01028131,
      "balance_loss_clip": 1.04342628,
      "balance_loss_mlp": 1.02138412,
      "epoch": 0.7572897252449949,
      "flos": 25409241492480.0,
      "grad_norm": 1.632630195196061,
      "language_loss": 0.84287155,
      "learning_rate": 5.866129031331011e-07,
      "loss": 0.86429107,
      "num_input_tokens_seen": 135393980,
      "step": 6298,
      "time_per_iteration": 2.606428384780884
    },
    {
      "auxiliary_loss_clip": 0.01139191,
      "auxiliary_loss_mlp": 0.010253,
      "balance_loss_clip": 1.04476643,
      "balance_loss_mlp": 1.01835346,
      "epoch": 0.757409968135634,
      "flos": 24279348297600.0,
      "grad_norm": 2.1545124764189643,
      "language_loss": 0.83329141,
      "learning_rate": 5.8606187250913e-07,
      "loss": 0.85493636,
      "num_input_tokens_seen": 135412030,
      "step": 6299,
      "time_per_iteration": 3.309645891189575
    },
    {
      "auxiliary_loss_clip": 0.01150203,
      "auxiliary_loss_mlp": 0.00760916,
      "balance_loss_clip": 1.04874969,
      "balance_loss_mlp": 1.00027537,
      "epoch": 0.757530211026273,
      "flos": 24133766474880.0,
      "grad_norm": 1.7278152983801374,
      "language_loss": 0.83810246,
      "learning_rate": 5.855110563784482e-07,
      "loss": 0.85721362,
      "num_input_tokens_seen": 135430565,
      "step": 6300,
      "time_per_iteration": 3.354339122772217
    },
    {
      "auxiliary_loss_clip": 0.01145205,
      "auxiliary_loss_mlp": 0.00760321,
      "balance_loss_clip": 1.044981,
      "balance_loss_mlp": 1.00023484,
      "epoch": 0.7576504539169122,
      "flos": 23951807153280.0,
      "grad_norm": 1.589729935439796,
      "language_loss": 0.6412127,
      "learning_rate": 5.849604548246156e-07,
      "loss": 0.66026807,
      "num_input_tokens_seen": 135451675,
      "step": 6301,
      "time_per_iteration": 3.2500948905944824
    },
    {
      "auxiliary_loss_clip": 0.01144256,
      "auxiliary_loss_mlp": 0.00760354,
      "balance_loss_clip": 1.04901731,
      "balance_loss_mlp": 1.00026512,
      "epoch": 0.7577706968075513,
      "flos": 21251360712960.0,
      "grad_norm": 2.8099834376029573,
      "language_loss": 0.80317807,
      "learning_rate": 5.844100679311565e-07,
      "loss": 0.82222414,
      "num_input_tokens_seen": 135470635,
      "step": 6302,
      "time_per_iteration": 2.5195209980010986
    },
    {
      "auxiliary_loss_clip": 0.01137837,
      "auxiliary_loss_mlp": 0.01026345,
      "balance_loss_clip": 1.04792738,
      "balance_loss_mlp": 1.01885259,
      "epoch": 0.7578909396981903,
      "flos": 18296595002880.0,
      "grad_norm": 1.9811602822511167,
      "language_loss": 0.76441038,
      "learning_rate": 5.838598957815637e-07,
      "loss": 0.78605223,
      "num_input_tokens_seen": 135487865,
      "step": 6303,
      "time_per_iteration": 2.485959053039551
    },
    {
      "auxiliary_loss_clip": 0.01131239,
      "auxiliary_loss_mlp": 0.01025293,
      "balance_loss_clip": 1.04603338,
      "balance_loss_mlp": 1.01806355,
      "epoch": 0.7580111825888295,
      "flos": 25373869574400.0,
      "grad_norm": 1.5234044348824602,
      "language_loss": 0.85359859,
      "learning_rate": 5.833099384592996e-07,
      "loss": 0.87516397,
      "num_input_tokens_seen": 135508440,
      "step": 6304,
      "time_per_iteration": 2.56187105178833
    },
    {
      "auxiliary_loss_clip": 0.01135168,
      "auxiliary_loss_mlp": 0.01023676,
      "balance_loss_clip": 1.04680026,
      "balance_loss_mlp": 1.01715231,
      "epoch": 0.7581314254794685,
      "flos": 23768662682880.0,
      "grad_norm": 2.4332544721032487,
      "language_loss": 0.71617883,
      "learning_rate": 5.827601960477913e-07,
      "loss": 0.73776728,
      "num_input_tokens_seen": 135526365,
      "step": 6305,
      "time_per_iteration": 2.515650987625122
    },
    {
      "auxiliary_loss_clip": 0.01147095,
      "auxiliary_loss_mlp": 0.01024441,
      "balance_loss_clip": 1.04476774,
      "balance_loss_mlp": 1.01767612,
      "epoch": 0.7582516683701076,
      "flos": 22054610603520.0,
      "grad_norm": 1.693623405581964,
      "language_loss": 0.70603848,
      "learning_rate": 5.822106686304344e-07,
      "loss": 0.72775388,
      "num_input_tokens_seen": 135545655,
      "step": 6306,
      "time_per_iteration": 2.5034329891204834
    },
    {
      "auxiliary_loss_clip": 0.01127908,
      "auxiliary_loss_mlp": 0.01028416,
      "balance_loss_clip": 1.04246116,
      "balance_loss_mlp": 1.02186823,
      "epoch": 0.7583719112607467,
      "flos": 31649725848960.0,
      "grad_norm": 1.592147231226397,
      "language_loss": 0.57802141,
      "learning_rate": 5.816613562905919e-07,
      "loss": 0.59958458,
      "num_input_tokens_seen": 135566840,
      "step": 6307,
      "time_per_iteration": 2.633861780166626
    },
    {
      "auxiliary_loss_clip": 0.01115066,
      "auxiliary_loss_mlp": 0.01023375,
      "balance_loss_clip": 1.04187584,
      "balance_loss_mlp": 1.0167979,
      "epoch": 0.7584921541513858,
      "flos": 33068376478080.0,
      "grad_norm": 1.5438072518692956,
      "language_loss": 0.69804072,
      "learning_rate": 5.811122591115933e-07,
      "loss": 0.7194252,
      "num_input_tokens_seen": 135587825,
      "step": 6308,
      "time_per_iteration": 2.6755828857421875
    },
    {
      "auxiliary_loss_clip": 0.01122856,
      "auxiliary_loss_mlp": 0.01030186,
      "balance_loss_clip": 1.04777932,
      "balance_loss_mlp": 1.02328658,
      "epoch": 0.7586123970420249,
      "flos": 23326350606720.0,
      "grad_norm": 2.5264297523187498,
      "language_loss": 0.71434706,
      "learning_rate": 5.805633771767376e-07,
      "loss": 0.73587751,
      "num_input_tokens_seen": 135605220,
      "step": 6309,
      "time_per_iteration": 2.5452022552490234
    },
    {
      "auxiliary_loss_clip": 0.01129412,
      "auxiliary_loss_mlp": 0.01025055,
      "balance_loss_clip": 1.0448513,
      "balance_loss_mlp": 1.01800096,
      "epoch": 0.7587326399326639,
      "flos": 18334229477760.0,
      "grad_norm": 1.6248246600704532,
      "language_loss": 0.77555394,
      "learning_rate": 5.800147105692888e-07,
      "loss": 0.79709864,
      "num_input_tokens_seen": 135624795,
      "step": 6310,
      "time_per_iteration": 2.545656681060791
    },
    {
      "auxiliary_loss_clip": 0.01149355,
      "auxiliary_loss_mlp": 0.01024289,
      "balance_loss_clip": 1.04283464,
      "balance_loss_mlp": 1.01746464,
      "epoch": 0.7588528828233031,
      "flos": 17275080119040.0,
      "grad_norm": 1.828367730074388,
      "language_loss": 0.79011232,
      "learning_rate": 5.794662593724795e-07,
      "loss": 0.81184876,
      "num_input_tokens_seen": 135643800,
      "step": 6311,
      "time_per_iteration": 2.4788119792938232
    },
    {
      "auxiliary_loss_clip": 0.01165493,
      "auxiliary_loss_mlp": 0.01028444,
      "balance_loss_clip": 1.04854155,
      "balance_loss_mlp": 1.02113903,
      "epoch": 0.7589731257139422,
      "flos": 17713621267200.0,
      "grad_norm": 1.8858461267105502,
      "language_loss": 0.74803972,
      "learning_rate": 5.789180236695091e-07,
      "loss": 0.76997912,
      "num_input_tokens_seen": 135660655,
      "step": 6312,
      "time_per_iteration": 2.450998306274414
    },
    {
      "auxiliary_loss_clip": 0.01145612,
      "auxiliary_loss_mlp": 0.01021866,
      "balance_loss_clip": 1.04615116,
      "balance_loss_mlp": 1.01557136,
      "epoch": 0.7590933686045812,
      "flos": 15961072786560.0,
      "grad_norm": 1.9491831132253836,
      "language_loss": 0.84868789,
      "learning_rate": 5.78370003543544e-07,
      "loss": 0.87036264,
      "num_input_tokens_seen": 135679410,
      "step": 6313,
      "time_per_iteration": 2.461963176727295
    },
    {
      "auxiliary_loss_clip": 0.01151503,
      "auxiliary_loss_mlp": 0.0076076,
      "balance_loss_clip": 1.04723406,
      "balance_loss_mlp": 1.00026166,
      "epoch": 0.7592136114952204,
      "flos": 21068072588160.0,
      "grad_norm": 1.9028909066499755,
      "language_loss": 0.8385517,
      "learning_rate": 5.778221990777203e-07,
      "loss": 0.85767436,
      "num_input_tokens_seen": 135697150,
      "step": 6314,
      "time_per_iteration": 3.25311541557312
    },
    {
      "auxiliary_loss_clip": 0.0113993,
      "auxiliary_loss_mlp": 0.0102681,
      "balance_loss_clip": 1.04780293,
      "balance_loss_mlp": 1.01988387,
      "epoch": 0.7593338543858594,
      "flos": 25297666871040.0,
      "grad_norm": 1.9740704353826688,
      "language_loss": 0.82804322,
      "learning_rate": 5.772746103551372e-07,
      "loss": 0.84971064,
      "num_input_tokens_seen": 135712545,
      "step": 6315,
      "time_per_iteration": 2.5311686992645264
    },
    {
      "auxiliary_loss_clip": 0.0113502,
      "auxiliary_loss_mlp": 0.01024409,
      "balance_loss_clip": 1.04699397,
      "balance_loss_mlp": 1.0178194,
      "epoch": 0.7594540972764985,
      "flos": 31832367528960.0,
      "grad_norm": 1.8355363712275725,
      "language_loss": 0.71621025,
      "learning_rate": 5.767272374588648e-07,
      "loss": 0.73780453,
      "num_input_tokens_seen": 135733950,
      "step": 6316,
      "time_per_iteration": 2.6082472801208496
    },
    {
      "auxiliary_loss_clip": 0.01148553,
      "auxiliary_loss_mlp": 0.01024153,
      "balance_loss_clip": 1.04734182,
      "balance_loss_mlp": 1.01753116,
      "epoch": 0.7595743401671377,
      "flos": 37597250880000.0,
      "grad_norm": 2.526937294712178,
      "language_loss": 0.77912152,
      "learning_rate": 5.76180080471939e-07,
      "loss": 0.8008486,
      "num_input_tokens_seen": 135757120,
      "step": 6317,
      "time_per_iteration": 2.638488292694092
    },
    {
      "auxiliary_loss_clip": 0.01167824,
      "auxiliary_loss_mlp": 0.01029037,
      "balance_loss_clip": 1.04903936,
      "balance_loss_mlp": 1.02145529,
      "epoch": 0.7596945830577767,
      "flos": 18287724343680.0,
      "grad_norm": 1.9725976441731259,
      "language_loss": 0.71721423,
      "learning_rate": 5.756331394773631e-07,
      "loss": 0.73918283,
      "num_input_tokens_seen": 135773335,
      "step": 6318,
      "time_per_iteration": 2.4152064323425293
    },
    {
      "auxiliary_loss_clip": 0.01093858,
      "auxiliary_loss_mlp": 0.00760429,
      "balance_loss_clip": 1.04124546,
      "balance_loss_mlp": 1.00022244,
      "epoch": 0.7598148259484158,
      "flos": 22233122219520.0,
      "grad_norm": 1.630465295060502,
      "language_loss": 0.75609148,
      "learning_rate": 5.750864145581071e-07,
      "loss": 0.77463436,
      "num_input_tokens_seen": 135792555,
      "step": 6319,
      "time_per_iteration": 2.6289210319519043
    },
    {
      "auxiliary_loss_clip": 0.0116485,
      "auxiliary_loss_mlp": 0.01031225,
      "balance_loss_clip": 1.0489186,
      "balance_loss_mlp": 1.0245254,
      "epoch": 0.7599350688390549,
      "flos": 27161718145920.0,
      "grad_norm": 1.8547230341334604,
      "language_loss": 0.86334121,
      "learning_rate": 5.745399057971085e-07,
      "loss": 0.88530195,
      "num_input_tokens_seen": 135813690,
      "step": 6320,
      "time_per_iteration": 2.516059637069702
    },
    {
      "auxiliary_loss_clip": 0.01153779,
      "auxiliary_loss_mlp": 0.01029138,
      "balance_loss_clip": 1.04607236,
      "balance_loss_mlp": 1.02174115,
      "epoch": 0.760055311729694,
      "flos": 15560704817280.0,
      "grad_norm": 2.046105799644153,
      "language_loss": 0.74953353,
      "learning_rate": 5.739936132772738e-07,
      "loss": 0.77136272,
      "num_input_tokens_seen": 135832255,
      "step": 6321,
      "time_per_iteration": 2.466679334640503
    },
    {
      "auxiliary_loss_clip": 0.01162807,
      "auxiliary_loss_mlp": 0.01027729,
      "balance_loss_clip": 1.04645646,
      "balance_loss_mlp": 1.01990259,
      "epoch": 0.760175554620333,
      "flos": 25155496840320.0,
      "grad_norm": 1.837311660323439,
      "language_loss": 0.74341613,
      "learning_rate": 5.734475370814733e-07,
      "loss": 0.76532149,
      "num_input_tokens_seen": 135851935,
      "step": 6322,
      "time_per_iteration": 2.4814727306365967
    },
    {
      "auxiliary_loss_clip": 0.01149471,
      "auxiliary_loss_mlp": 0.01025574,
      "balance_loss_clip": 1.04361558,
      "balance_loss_mlp": 1.01862383,
      "epoch": 0.7602957975109722,
      "flos": 24353791234560.0,
      "grad_norm": 1.5007692699439728,
      "language_loss": 0.78543818,
      "learning_rate": 5.729016772925483e-07,
      "loss": 0.80718863,
      "num_input_tokens_seen": 135873510,
      "step": 6323,
      "time_per_iteration": 2.5236868858337402
    },
    {
      "auxiliary_loss_clip": 0.01104179,
      "auxiliary_loss_mlp": 0.01030051,
      "balance_loss_clip": 1.04266405,
      "balance_loss_mlp": 1.02177191,
      "epoch": 0.7604160404016113,
      "flos": 25192664438400.0,
      "grad_norm": 1.7043596616014984,
      "language_loss": 0.70684195,
      "learning_rate": 5.723560339933038e-07,
      "loss": 0.72818422,
      "num_input_tokens_seen": 135893845,
      "step": 6324,
      "time_per_iteration": 2.6087496280670166
    },
    {
      "auxiliary_loss_clip": 0.01150247,
      "auxiliary_loss_mlp": 0.00760619,
      "balance_loss_clip": 1.04589593,
      "balance_loss_mlp": 1.00025296,
      "epoch": 0.7605362832922503,
      "flos": 29861841363840.0,
      "grad_norm": 2.193203560848772,
      "language_loss": 0.64761871,
      "learning_rate": 5.71810607266513e-07,
      "loss": 0.66672736,
      "num_input_tokens_seen": 135912430,
      "step": 6325,
      "time_per_iteration": 3.307513952255249
    },
    {
      "auxiliary_loss_clip": 0.0115149,
      "auxiliary_loss_mlp": 0.01027673,
      "balance_loss_clip": 1.04637814,
      "balance_loss_mlp": 1.02067232,
      "epoch": 0.7606565261828895,
      "flos": 13917935278080.0,
      "grad_norm": 1.810525429734524,
      "language_loss": 0.60192609,
      "learning_rate": 5.712653971949184e-07,
      "loss": 0.62371773,
      "num_input_tokens_seen": 135930550,
      "step": 6326,
      "time_per_iteration": 3.2753994464874268
    },
    {
      "auxiliary_loss_clip": 0.01145787,
      "auxiliary_loss_mlp": 0.01024434,
      "balance_loss_clip": 1.04491234,
      "balance_loss_mlp": 1.01679873,
      "epoch": 0.7607767690735285,
      "flos": 18551273408640.0,
      "grad_norm": 2.525304684920854,
      "language_loss": 0.75163108,
      "learning_rate": 5.707204038612268e-07,
      "loss": 0.77333331,
      "num_input_tokens_seen": 135947980,
      "step": 6327,
      "time_per_iteration": 3.2129251956939697
    },
    {
      "auxiliary_loss_clip": 0.01148692,
      "auxiliary_loss_mlp": 0.01033543,
      "balance_loss_clip": 1.05229187,
      "balance_loss_mlp": 1.02534127,
      "epoch": 0.7608970119641676,
      "flos": 20922993555840.0,
      "grad_norm": 2.737354611202029,
      "language_loss": 0.74038863,
      "learning_rate": 5.701756273481138e-07,
      "loss": 0.76221097,
      "num_input_tokens_seen": 135965400,
      "step": 6328,
      "time_per_iteration": 2.5178442001342773
    },
    {
      "auxiliary_loss_clip": 0.01135197,
      "auxiliary_loss_mlp": 0.01025799,
      "balance_loss_clip": 1.04357374,
      "balance_loss_mlp": 1.01859617,
      "epoch": 0.7610172548548068,
      "flos": 23807302738560.0,
      "grad_norm": 1.4728384510788137,
      "language_loss": 0.73859912,
      "learning_rate": 5.696310677382212e-07,
      "loss": 0.76020908,
      "num_input_tokens_seen": 135986795,
      "step": 6329,
      "time_per_iteration": 2.5391077995300293
    },
    {
      "auxiliary_loss_clip": 0.01023816,
      "auxiliary_loss_mlp": 0.01001389,
      "balance_loss_clip": 1.01029754,
      "balance_loss_mlp": 1.00031602,
      "epoch": 0.7611374977454458,
      "flos": 66496580426880.0,
      "grad_norm": 0.8852852776416007,
      "language_loss": 0.61770189,
      "learning_rate": 5.690867251141576e-07,
      "loss": 0.63795394,
      "num_input_tokens_seen": 136053450,
      "step": 6330,
      "time_per_iteration": 3.3051459789276123
    },
    {
      "auxiliary_loss_clip": 0.0115406,
      "auxiliary_loss_mlp": 0.01030604,
      "balance_loss_clip": 1.04642272,
      "balance_loss_mlp": 1.02287948,
      "epoch": 0.7612577406360849,
      "flos": 15633136592640.0,
      "grad_norm": 2.013891165056972,
      "language_loss": 0.91489697,
      "learning_rate": 5.685425995585013e-07,
      "loss": 0.93674356,
      "num_input_tokens_seen": 136071375,
      "step": 6331,
      "time_per_iteration": 2.5030243396759033
    },
    {
      "auxiliary_loss_clip": 0.01038565,
      "auxiliary_loss_mlp": 0.01004163,
      "balance_loss_clip": 1.00948453,
      "balance_loss_mlp": 1.00301254,
      "epoch": 0.761377983526724,
      "flos": 60526253237760.0,
      "grad_norm": 0.761871995392517,
      "language_loss": 0.59058797,
      "learning_rate": 5.679986911537935e-07,
      "loss": 0.6110152,
      "num_input_tokens_seen": 136138905,
      "step": 6332,
      "time_per_iteration": 3.2729945182800293
    },
    {
      "auxiliary_loss_clip": 0.01094266,
      "auxiliary_loss_mlp": 0.01028241,
      "balance_loss_clip": 1.03996325,
      "balance_loss_mlp": 1.02121639,
      "epoch": 0.7614982264173631,
      "flos": 35772522019200.0,
      "grad_norm": 1.7062527960777727,
      "language_loss": 0.67029548,
      "learning_rate": 5.674549999825462e-07,
      "loss": 0.69152051,
      "num_input_tokens_seen": 136161720,
      "step": 6333,
      "time_per_iteration": 2.6852099895477295
    },
    {
      "auxiliary_loss_clip": 0.01047657,
      "auxiliary_loss_mlp": 0.01000421,
      "balance_loss_clip": 1.00912833,
      "balance_loss_mlp": 0.99926436,
      "epoch": 0.7616184693080021,
      "flos": 67925502345600.0,
      "grad_norm": 0.9196568058139132,
      "language_loss": 0.71454853,
      "learning_rate": 5.669115261272363e-07,
      "loss": 0.73502934,
      "num_input_tokens_seen": 136222040,
      "step": 6334,
      "time_per_iteration": 3.0886387825012207
    },
    {
      "auxiliary_loss_clip": 0.01154618,
      "auxiliary_loss_mlp": 0.01028956,
      "balance_loss_clip": 1.04832792,
      "balance_loss_mlp": 1.02197337,
      "epoch": 0.7617387121986413,
      "flos": 20521979141760.0,
      "grad_norm": 2.22525732197616,
      "language_loss": 0.72213817,
      "learning_rate": 5.663682696703081e-07,
      "loss": 0.74397391,
      "num_input_tokens_seen": 136240305,
      "step": 6335,
      "time_per_iteration": 2.4654006958007812
    },
    {
      "auxiliary_loss_clip": 0.01163705,
      "auxiliary_loss_mlp": 0.0102863,
      "balance_loss_clip": 1.04815042,
      "balance_loss_mlp": 1.02183759,
      "epoch": 0.7618589550892804,
      "flos": 18624495283200.0,
      "grad_norm": 1.7134878725615768,
      "language_loss": 0.81967413,
      "learning_rate": 5.658252306941746e-07,
      "loss": 0.8415975,
      "num_input_tokens_seen": 136259625,
      "step": 6336,
      "time_per_iteration": 2.428879737854004
    },
    {
      "auxiliary_loss_clip": 0.01112562,
      "auxiliary_loss_mlp": 0.01024496,
      "balance_loss_clip": 1.04442656,
      "balance_loss_mlp": 1.01694393,
      "epoch": 0.7619791979799194,
      "flos": 17453735389440.0,
      "grad_norm": 2.211111597295149,
      "language_loss": 0.75053298,
      "learning_rate": 5.65282409281212e-07,
      "loss": 0.77190357,
      "num_input_tokens_seen": 136277090,
      "step": 6337,
      "time_per_iteration": 2.55572772026062
    },
    {
      "auxiliary_loss_clip": 0.01132741,
      "auxiliary_loss_mlp": 0.01031499,
      "balance_loss_clip": 1.04439962,
      "balance_loss_mlp": 1.02465641,
      "epoch": 0.7620994408705585,
      "flos": 14137421333760.0,
      "grad_norm": 2.4967145260602126,
      "language_loss": 0.69746745,
      "learning_rate": 5.64739805513768e-07,
      "loss": 0.71910977,
      "num_input_tokens_seen": 136294635,
      "step": 6338,
      "time_per_iteration": 2.503326654434204
    },
    {
      "auxiliary_loss_clip": 0.01045985,
      "auxiliary_loss_mlp": 0.00751267,
      "balance_loss_clip": 1.01040137,
      "balance_loss_mlp": 1.00007582,
      "epoch": 0.7622196837611976,
      "flos": 70708792527360.0,
      "grad_norm": 0.7880571559918096,
      "language_loss": 0.55737901,
      "learning_rate": 5.641974194741541e-07,
      "loss": 0.57535148,
      "num_input_tokens_seen": 136350320,
      "step": 6339,
      "time_per_iteration": 2.9996328353881836
    },
    {
      "auxiliary_loss_clip": 0.01027927,
      "auxiliary_loss_mlp": 0.01002374,
      "balance_loss_clip": 1.01209581,
      "balance_loss_mlp": 1.00121176,
      "epoch": 0.7623399266518367,
      "flos": 60684150447360.0,
      "grad_norm": 0.7751933052684012,
      "language_loss": 0.63717395,
      "learning_rate": 5.636552512446502e-07,
      "loss": 0.65747696,
      "num_input_tokens_seen": 136411375,
      "step": 6340,
      "time_per_iteration": 3.744497776031494
    },
    {
      "auxiliary_loss_clip": 0.01145351,
      "auxiliary_loss_mlp": 0.01030427,
      "balance_loss_clip": 1.04606199,
      "balance_loss_mlp": 1.02323604,
      "epoch": 0.7624601695424758,
      "flos": 26468893641600.0,
      "grad_norm": 1.672086263401076,
      "language_loss": 0.78005278,
      "learning_rate": 5.631133009075027e-07,
      "loss": 0.80181062,
      "num_input_tokens_seen": 136430560,
      "step": 6341,
      "time_per_iteration": 2.5267655849456787
    },
    {
      "auxiliary_loss_clip": 0.01150873,
      "auxiliary_loss_mlp": 0.00760188,
      "balance_loss_clip": 1.04708362,
      "balance_loss_mlp": 1.00023913,
      "epoch": 0.7625804124331149,
      "flos": 19135755515520.0,
      "grad_norm": 1.8070997533263713,
      "language_loss": 0.68233848,
      "learning_rate": 5.625715685449242e-07,
      "loss": 0.7014491,
      "num_input_tokens_seen": 136448665,
      "step": 6342,
      "time_per_iteration": 2.470076322555542
    },
    {
      "auxiliary_loss_clip": 0.01117611,
      "auxiliary_loss_mlp": 0.01030445,
      "balance_loss_clip": 1.04778266,
      "balance_loss_mlp": 1.02396631,
      "epoch": 0.762700655323754,
      "flos": 26213101914240.0,
      "grad_norm": 1.5351551921640936,
      "language_loss": 0.71685427,
      "learning_rate": 5.620300542390966e-07,
      "loss": 0.73833477,
      "num_input_tokens_seen": 136469710,
      "step": 6343,
      "time_per_iteration": 2.5825188159942627
    },
    {
      "auxiliary_loss_clip": 0.01130849,
      "auxiliary_loss_mlp": 0.01023019,
      "balance_loss_clip": 1.04455829,
      "balance_loss_mlp": 1.01618195,
      "epoch": 0.762820898214393,
      "flos": 22382582711040.0,
      "grad_norm": 1.8676049612498908,
      "language_loss": 0.85095334,
      "learning_rate": 5.614887580721659e-07,
      "loss": 0.87249207,
      "num_input_tokens_seen": 136489855,
      "step": 6344,
      "time_per_iteration": 2.525118827819824
    },
    {
      "auxiliary_loss_clip": 0.01119682,
      "auxiliary_loss_mlp": 0.01031538,
      "balance_loss_clip": 1.04714942,
      "balance_loss_mlp": 1.02476144,
      "epoch": 0.7629411411050322,
      "flos": 15700504550400.0,
      "grad_norm": 2.322787147383839,
      "language_loss": 0.73517859,
      "learning_rate": 5.609476801262481e-07,
      "loss": 0.75669074,
      "num_input_tokens_seen": 136504715,
      "step": 6345,
      "time_per_iteration": 2.5189685821533203
    },
    {
      "auxiliary_loss_clip": 0.01119906,
      "auxiliary_loss_mlp": 0.01025671,
      "balance_loss_clip": 1.04485798,
      "balance_loss_mlp": 1.0186646,
      "epoch": 0.7630613839956712,
      "flos": 13770342293760.0,
      "grad_norm": 2.1162797310752524,
      "language_loss": 0.64066082,
      "learning_rate": 5.604068204834223e-07,
      "loss": 0.66211659,
      "num_input_tokens_seen": 136521610,
      "step": 6346,
      "time_per_iteration": 2.4946060180664062
    },
    {
      "auxiliary_loss_clip": 0.01107,
      "auxiliary_loss_mlp": 0.00761224,
      "balance_loss_clip": 1.04369164,
      "balance_loss_mlp": 1.00022507,
      "epoch": 0.7631816268863103,
      "flos": 14569569861120.0,
      "grad_norm": 1.99258828786321,
      "language_loss": 0.76383305,
      "learning_rate": 5.598661792257367e-07,
      "loss": 0.78251529,
      "num_input_tokens_seen": 136538655,
      "step": 6347,
      "time_per_iteration": 2.5658271312713623
    },
    {
      "auxiliary_loss_clip": 0.01148331,
      "auxiliary_loss_mlp": 0.01025539,
      "balance_loss_clip": 1.04543245,
      "balance_loss_mlp": 1.01850557,
      "epoch": 0.7633018697769495,
      "flos": 19062210418560.0,
      "grad_norm": 1.8836035393487447,
      "language_loss": 0.75830019,
      "learning_rate": 5.593257564352071e-07,
      "loss": 0.78003883,
      "num_input_tokens_seen": 136557095,
      "step": 6348,
      "time_per_iteration": 2.4694929122924805
    },
    {
      "auxiliary_loss_clip": 0.01148368,
      "auxiliary_loss_mlp": 0.01028685,
      "balance_loss_clip": 1.04680407,
      "balance_loss_mlp": 1.02203321,
      "epoch": 0.7634221126675885,
      "flos": 22052958577920.0,
      "grad_norm": 1.8500280001909801,
      "language_loss": 0.75650716,
      "learning_rate": 5.58785552193815e-07,
      "loss": 0.77827764,
      "num_input_tokens_seen": 136577340,
      "step": 6349,
      "time_per_iteration": 2.4983203411102295
    },
    {
      "auxiliary_loss_clip": 0.01162843,
      "auxiliary_loss_mlp": 0.0102377,
      "balance_loss_clip": 1.04646146,
      "balance_loss_mlp": 1.01725793,
      "epoch": 0.7635423555582276,
      "flos": 29382720825600.0,
      "grad_norm": 1.7762796062646888,
      "language_loss": 0.75394315,
      "learning_rate": 5.582455665835086e-07,
      "loss": 0.77580929,
      "num_input_tokens_seen": 136597635,
      "step": 6350,
      "time_per_iteration": 2.498189687728882
    },
    {
      "auxiliary_loss_clip": 0.01143102,
      "auxiliary_loss_mlp": 0.01029628,
      "balance_loss_clip": 1.04329622,
      "balance_loss_mlp": 1.02202868,
      "epoch": 0.7636625984488667,
      "flos": 17784903807360.0,
      "grad_norm": 3.0252774209669013,
      "language_loss": 0.7263214,
      "learning_rate": 5.577057996862036e-07,
      "loss": 0.74804866,
      "num_input_tokens_seen": 136615260,
      "step": 6351,
      "time_per_iteration": 3.277927875518799
    },
    {
      "auxiliary_loss_clip": 0.01160441,
      "auxiliary_loss_mlp": 0.01023693,
      "balance_loss_clip": 1.04765308,
      "balance_loss_mlp": 1.01713037,
      "epoch": 0.7637828413395058,
      "flos": 23734583654400.0,
      "grad_norm": 1.5014443500598753,
      "language_loss": 0.7594313,
      "learning_rate": 5.571662515837814e-07,
      "loss": 0.78127265,
      "num_input_tokens_seen": 136637220,
      "step": 6352,
      "time_per_iteration": 3.310213327407837
    },
    {
      "auxiliary_loss_clip": 0.01135589,
      "auxiliary_loss_mlp": 0.01022655,
      "balance_loss_clip": 1.04500389,
      "balance_loss_mlp": 1.01610172,
      "epoch": 0.7639030842301449,
      "flos": 36283279461120.0,
      "grad_norm": 1.5560494706296821,
      "language_loss": 0.83729237,
      "learning_rate": 5.566269223580926e-07,
      "loss": 0.8588748,
      "num_input_tokens_seen": 136658930,
      "step": 6353,
      "time_per_iteration": 2.6533987522125244
    },
    {
      "auxiliary_loss_clip": 0.01151741,
      "auxiliary_loss_mlp": 0.010273,
      "balance_loss_clip": 1.04584289,
      "balance_loss_mlp": 1.02080607,
      "epoch": 0.764023327120784,
      "flos": 28878104609280.0,
      "grad_norm": 1.595168008154554,
      "language_loss": 0.75391185,
      "learning_rate": 5.560878120909511e-07,
      "loss": 0.7757023,
      "num_input_tokens_seen": 136681530,
      "step": 6354,
      "time_per_iteration": 2.546997308731079
    },
    {
      "auxiliary_loss_clip": 0.0104791,
      "auxiliary_loss_mlp": 0.01004474,
      "balance_loss_clip": 1.0091083,
      "balance_loss_mlp": 1.00334167,
      "epoch": 0.7641435700114231,
      "flos": 64789711067520.0,
      "grad_norm": 0.8410175480921163,
      "language_loss": 0.58633167,
      "learning_rate": 5.55548920864141e-07,
      "loss": 0.60685551,
      "num_input_tokens_seen": 136742185,
      "step": 6355,
      "time_per_iteration": 3.1071739196777344
    },
    {
      "auxiliary_loss_clip": 0.01151129,
      "auxiliary_loss_mlp": 0.01028929,
      "balance_loss_clip": 1.05033982,
      "balance_loss_mlp": 1.02227736,
      "epoch": 0.7642638129020621,
      "flos": 16835784785280.0,
      "grad_norm": 1.798471801209253,
      "language_loss": 0.78148651,
      "learning_rate": 5.550102487594113e-07,
      "loss": 0.80328715,
      "num_input_tokens_seen": 136760855,
      "step": 6356,
      "time_per_iteration": 2.4796981811523438
    },
    {
      "auxiliary_loss_clip": 0.01110068,
      "auxiliary_loss_mlp": 0.00760653,
      "balance_loss_clip": 1.04043889,
      "balance_loss_mlp": 1.00023746,
      "epoch": 0.7643840557927013,
      "flos": 30408940391040.0,
      "grad_norm": 1.5610727290281945,
      "language_loss": 0.71611345,
      "learning_rate": 5.54471795858477e-07,
      "loss": 0.73482066,
      "num_input_tokens_seen": 136780925,
      "step": 6357,
      "time_per_iteration": 2.64902925491333
    },
    {
      "auxiliary_loss_clip": 0.01120297,
      "auxiliary_loss_mlp": 0.0102601,
      "balance_loss_clip": 1.03970838,
      "balance_loss_mlp": 1.01913714,
      "epoch": 0.7645042986833404,
      "flos": 16983234115200.0,
      "grad_norm": 2.244122700963788,
      "language_loss": 0.82912391,
      "learning_rate": 5.539335622430235e-07,
      "loss": 0.85058701,
      "num_input_tokens_seen": 136799545,
      "step": 6358,
      "time_per_iteration": 2.552835702896118
    },
    {
      "auxiliary_loss_clip": 0.01144076,
      "auxiliary_loss_mlp": 0.01024331,
      "balance_loss_clip": 1.04372573,
      "balance_loss_mlp": 1.01736903,
      "epoch": 0.7646245415739794,
      "flos": 17311493531520.0,
      "grad_norm": 1.9392747431724207,
      "language_loss": 0.74648976,
      "learning_rate": 5.533955479946975e-07,
      "loss": 0.76817387,
      "num_input_tokens_seen": 136818325,
      "step": 6359,
      "time_per_iteration": 2.4674482345581055
    },
    {
      "auxiliary_loss_clip": 0.0101505,
      "auxiliary_loss_mlp": 0.00750892,
      "balance_loss_clip": 1.01246977,
      "balance_loss_mlp": 1.0000124,
      "epoch": 0.7647447844646186,
      "flos": 70402332666240.0,
      "grad_norm": 0.8524677021695252,
      "language_loss": 0.65775782,
      "learning_rate": 5.528577531951173e-07,
      "loss": 0.67541724,
      "num_input_tokens_seen": 136878730,
      "step": 6360,
      "time_per_iteration": 3.115774393081665
    },
    {
      "auxiliary_loss_clip": 0.01137955,
      "auxiliary_loss_mlp": 0.01026071,
      "balance_loss_clip": 1.04453433,
      "balance_loss_mlp": 1.01964545,
      "epoch": 0.7648650273552576,
      "flos": 17675914965120.0,
      "grad_norm": 1.9828307627069413,
      "language_loss": 0.74076986,
      "learning_rate": 5.523201779258653e-07,
      "loss": 0.7624101,
      "num_input_tokens_seen": 136897705,
      "step": 6361,
      "time_per_iteration": 2.4971768856048584
    },
    {
      "auxiliary_loss_clip": 0.0116476,
      "auxiliary_loss_mlp": 0.01023252,
      "balance_loss_clip": 1.04731965,
      "balance_loss_mlp": 1.01609647,
      "epoch": 0.7649852702458967,
      "flos": 22162019247360.0,
      "grad_norm": 1.7824519658894622,
      "language_loss": 0.83907866,
      "learning_rate": 5.517828222684912e-07,
      "loss": 0.86095876,
      "num_input_tokens_seen": 136918360,
      "step": 6362,
      "time_per_iteration": 2.4656870365142822
    },
    {
      "auxiliary_loss_clip": 0.01037882,
      "auxiliary_loss_mlp": 0.01003384,
      "balance_loss_clip": 1.01162744,
      "balance_loss_mlp": 1.00228143,
      "epoch": 0.7651055131365359,
      "flos": 69848338227840.0,
      "grad_norm": 0.7737078642485606,
      "language_loss": 0.59101796,
      "learning_rate": 5.512456863045117e-07,
      "loss": 0.61143064,
      "num_input_tokens_seen": 136979050,
      "step": 6363,
      "time_per_iteration": 3.1221446990966797
    },
    {
      "auxiliary_loss_clip": 0.01162854,
      "auxiliary_loss_mlp": 0.01026218,
      "balance_loss_clip": 1.04639876,
      "balance_loss_mlp": 1.01883268,
      "epoch": 0.7652257560271749,
      "flos": 19464014931840.0,
      "grad_norm": 1.6368106367265947,
      "language_loss": 0.73981816,
      "learning_rate": 5.507087701154089e-07,
      "loss": 0.76170886,
      "num_input_tokens_seen": 136998970,
      "step": 6364,
      "time_per_iteration": 2.471106767654419
    },
    {
      "auxiliary_loss_clip": 0.01111203,
      "auxiliary_loss_mlp": 0.01030989,
      "balance_loss_clip": 1.04173875,
      "balance_loss_mlp": 1.02418804,
      "epoch": 0.765345998917814,
      "flos": 15961108700160.0,
      "grad_norm": 2.8690555840516625,
      "language_loss": 0.75353587,
      "learning_rate": 5.50172073782634e-07,
      "loss": 0.77495778,
      "num_input_tokens_seen": 137016950,
      "step": 6365,
      "time_per_iteration": 2.5637259483337402
    },
    {
      "auxiliary_loss_clip": 0.01120088,
      "auxiliary_loss_mlp": 0.01027972,
      "balance_loss_clip": 1.04548073,
      "balance_loss_mlp": 1.02132261,
      "epoch": 0.7654662418084531,
      "flos": 23659853408640.0,
      "grad_norm": 1.7348416832205575,
      "language_loss": 0.87720394,
      "learning_rate": 5.496355973876023e-07,
      "loss": 0.89868456,
      "num_input_tokens_seen": 137036205,
      "step": 6366,
      "time_per_iteration": 3.344924211502075
    },
    {
      "auxiliary_loss_clip": 0.0111876,
      "auxiliary_loss_mlp": 0.00761424,
      "balance_loss_clip": 1.0413965,
      "balance_loss_mlp": 1.00020552,
      "epoch": 0.7655864846990922,
      "flos": 41463608878080.0,
      "grad_norm": 1.7847356064734612,
      "language_loss": 0.7109251,
      "learning_rate": 5.490993410116984e-07,
      "loss": 0.72972703,
      "num_input_tokens_seen": 137059195,
      "step": 6367,
      "time_per_iteration": 2.739509344100952
    },
    {
      "auxiliary_loss_clip": 0.01113681,
      "auxiliary_loss_mlp": 0.01024978,
      "balance_loss_clip": 1.04092908,
      "balance_loss_mlp": 1.01827812,
      "epoch": 0.7657067275897312,
      "flos": 43142684088960.0,
      "grad_norm": 1.598759116198481,
      "language_loss": 0.70075858,
      "learning_rate": 5.485633047362704e-07,
      "loss": 0.7221452,
      "num_input_tokens_seen": 137081200,
      "step": 6368,
      "time_per_iteration": 2.7446553707122803
    },
    {
      "auxiliary_loss_clip": 0.01169234,
      "auxiliary_loss_mlp": 0.01033419,
      "balance_loss_clip": 1.05067754,
      "balance_loss_mlp": 1.02575648,
      "epoch": 0.7658269704803703,
      "flos": 17311780840320.0,
      "grad_norm": 1.8301820493381575,
      "language_loss": 0.78455609,
      "learning_rate": 5.480274886426341e-07,
      "loss": 0.80658257,
      "num_input_tokens_seen": 137097840,
      "step": 6369,
      "time_per_iteration": 2.435459613800049
    },
    {
      "auxiliary_loss_clip": 0.01145436,
      "auxiliary_loss_mlp": 0.0102683,
      "balance_loss_clip": 1.04728246,
      "balance_loss_mlp": 1.01974916,
      "epoch": 0.7659472133710095,
      "flos": 12568160977920.0,
      "grad_norm": 2.07059247950138,
      "language_loss": 0.77812505,
      "learning_rate": 5.474918928120744e-07,
      "loss": 0.79984772,
      "num_input_tokens_seen": 137114335,
      "step": 6370,
      "time_per_iteration": 2.472855806350708
    },
    {
      "auxiliary_loss_clip": 0.01148611,
      "auxiliary_loss_mlp": 0.01024105,
      "balance_loss_clip": 1.04701924,
      "balance_loss_mlp": 1.01788199,
      "epoch": 0.7660674562616485,
      "flos": 22707430335360.0,
      "grad_norm": 2.513685171015111,
      "language_loss": 0.87263155,
      "learning_rate": 5.469565173258392e-07,
      "loss": 0.89435875,
      "num_input_tokens_seen": 137132850,
      "step": 6371,
      "time_per_iteration": 2.5022947788238525
    },
    {
      "auxiliary_loss_clip": 0.01168067,
      "auxiliary_loss_mlp": 0.01025572,
      "balance_loss_clip": 1.04753184,
      "balance_loss_mlp": 1.01813006,
      "epoch": 0.7661876991522876,
      "flos": 17056455989760.0,
      "grad_norm": 2.0985176289185237,
      "language_loss": 0.6343652,
      "learning_rate": 5.464213622651454e-07,
      "loss": 0.65630156,
      "num_input_tokens_seen": 137150665,
      "step": 6372,
      "time_per_iteration": 2.421651840209961
    },
    {
      "auxiliary_loss_clip": 0.01132184,
      "auxiliary_loss_mlp": 0.01028128,
      "balance_loss_clip": 1.0474087,
      "balance_loss_mlp": 1.02133036,
      "epoch": 0.7663079420429267,
      "flos": 20084228092800.0,
      "grad_norm": 1.6100169399190218,
      "language_loss": 0.84142804,
      "learning_rate": 5.458864277111753e-07,
      "loss": 0.86303127,
      "num_input_tokens_seen": 137168500,
      "step": 6373,
      "time_per_iteration": 2.566838026046753
    },
    {
      "auxiliary_loss_clip": 0.0113041,
      "auxiliary_loss_mlp": 0.00760415,
      "balance_loss_clip": 1.0428375,
      "balance_loss_mlp": 1.0002085,
      "epoch": 0.7664281849335658,
      "flos": 12677473042560.0,
      "grad_norm": 2.1421963015460372,
      "language_loss": 0.68665141,
      "learning_rate": 5.453517137450769e-07,
      "loss": 0.70555967,
      "num_input_tokens_seen": 137185075,
      "step": 6374,
      "time_per_iteration": 2.493882656097412
    },
    {
      "auxiliary_loss_clip": 0.01153817,
      "auxiliary_loss_mlp": 0.01022856,
      "balance_loss_clip": 1.04901934,
      "balance_loss_mlp": 1.01569784,
      "epoch": 0.7665484278242048,
      "flos": 22345271458560.0,
      "grad_norm": 1.6087903778414263,
      "language_loss": 0.75858778,
      "learning_rate": 5.448172204479684e-07,
      "loss": 0.7803545,
      "num_input_tokens_seen": 137204355,
      "step": 6375,
      "time_per_iteration": 2.502561330795288
    },
    {
      "auxiliary_loss_clip": 0.01162091,
      "auxiliary_loss_mlp": 0.01023018,
      "balance_loss_clip": 1.04639006,
      "balance_loss_mlp": 1.01638389,
      "epoch": 0.766668670714844,
      "flos": 23617909301760.0,
      "grad_norm": 1.5765141797159277,
      "language_loss": 0.74712133,
      "learning_rate": 5.442829479009294e-07,
      "loss": 0.76897246,
      "num_input_tokens_seen": 137223135,
      "step": 6376,
      "time_per_iteration": 3.257929563522339
    },
    {
      "auxiliary_loss_clip": 0.01153073,
      "auxiliary_loss_mlp": 0.01026591,
      "balance_loss_clip": 1.04702163,
      "balance_loss_mlp": 1.01892543,
      "epoch": 0.7667889136054831,
      "flos": 19427134642560.0,
      "grad_norm": 2.150556942665171,
      "language_loss": 0.71551037,
      "learning_rate": 5.437488961850103e-07,
      "loss": 0.73730707,
      "num_input_tokens_seen": 137242935,
      "step": 6377,
      "time_per_iteration": 2.4704620838165283
    },
    {
      "auxiliary_loss_clip": 0.01107528,
      "auxiliary_loss_mlp": 0.01027276,
      "balance_loss_clip": 1.04262471,
      "balance_loss_mlp": 1.02087176,
      "epoch": 0.7669091564961221,
      "flos": 26866352609280.0,
      "grad_norm": 1.6842195202731018,
      "language_loss": 0.75392795,
      "learning_rate": 5.432150653812258e-07,
      "loss": 0.77527606,
      "num_input_tokens_seen": 137262970,
      "step": 6378,
      "time_per_iteration": 4.146621227264404
    },
    {
      "auxiliary_loss_clip": 0.01145726,
      "auxiliary_loss_mlp": 0.01025816,
      "balance_loss_clip": 1.04627621,
      "balance_loss_mlp": 1.01906872,
      "epoch": 0.7670293993867613,
      "flos": 12385303816320.0,
      "grad_norm": 1.9391776451888283,
      "language_loss": 0.82509184,
      "learning_rate": 5.42681455570557e-07,
      "loss": 0.84680718,
      "num_input_tokens_seen": 137279500,
      "step": 6379,
      "time_per_iteration": 2.4746603965759277
    },
    {
      "auxiliary_loss_clip": 0.01158146,
      "auxiliary_loss_mlp": 0.01023705,
      "balance_loss_clip": 1.04458928,
      "balance_loss_mlp": 1.0167253,
      "epoch": 0.7671496422774003,
      "flos": 21762944167680.0,
      "grad_norm": 2.1733985760190393,
      "language_loss": 0.64911819,
      "learning_rate": 5.42148066833954e-07,
      "loss": 0.6709367,
      "num_input_tokens_seen": 137298745,
      "step": 6380,
      "time_per_iteration": 2.4844486713409424
    },
    {
      "auxiliary_loss_clip": 0.0116121,
      "auxiliary_loss_mlp": 0.01023623,
      "balance_loss_clip": 1.04753733,
      "balance_loss_mlp": 1.01701593,
      "epoch": 0.7672698851680394,
      "flos": 21069221823360.0,
      "grad_norm": 1.8734702093360256,
      "language_loss": 0.7527163,
      "learning_rate": 5.416148992523289e-07,
      "loss": 0.77456456,
      "num_input_tokens_seen": 137317320,
      "step": 6381,
      "time_per_iteration": 2.444093704223633
    },
    {
      "auxiliary_loss_clip": 0.01081235,
      "auxiliary_loss_mlp": 0.01024634,
      "balance_loss_clip": 1.0426532,
      "balance_loss_mlp": 1.01805699,
      "epoch": 0.7673901280586786,
      "flos": 16976697840000.0,
      "grad_norm": 1.6488511126466572,
      "language_loss": 0.78342968,
      "learning_rate": 5.410819529065644e-07,
      "loss": 0.80448842,
      "num_input_tokens_seen": 137335275,
      "step": 6382,
      "time_per_iteration": 2.6573266983032227
    },
    {
      "auxiliary_loss_clip": 0.01107249,
      "auxiliary_loss_mlp": 0.01028106,
      "balance_loss_clip": 1.04062986,
      "balance_loss_mlp": 1.02103972,
      "epoch": 0.7675103709493176,
      "flos": 29242669697280.0,
      "grad_norm": 1.7749751680167105,
      "language_loss": 0.65039605,
      "learning_rate": 5.405492278775079e-07,
      "loss": 0.67174959,
      "num_input_tokens_seen": 137355055,
      "step": 6383,
      "time_per_iteration": 2.6509013175964355
    },
    {
      "auxiliary_loss_clip": 0.01136538,
      "auxiliary_loss_mlp": 0.01024178,
      "balance_loss_clip": 1.04371452,
      "balance_loss_mlp": 1.01688874,
      "epoch": 0.7676306138399567,
      "flos": 29023004073600.0,
      "grad_norm": 1.9638934708340456,
      "language_loss": 0.80159479,
      "learning_rate": 5.400167242459732e-07,
      "loss": 0.82320195,
      "num_input_tokens_seen": 137374015,
      "step": 6384,
      "time_per_iteration": 2.58976149559021
    },
    {
      "auxiliary_loss_clip": 0.01145794,
      "auxiliary_loss_mlp": 0.01024361,
      "balance_loss_clip": 1.04491985,
      "balance_loss_mlp": 1.01779866,
      "epoch": 0.7677508567305958,
      "flos": 22565116650240.0,
      "grad_norm": 1.655398959915776,
      "language_loss": 0.80510634,
      "learning_rate": 5.394844420927405e-07,
      "loss": 0.82680792,
      "num_input_tokens_seen": 137393625,
      "step": 6385,
      "time_per_iteration": 2.508486032485962
    },
    {
      "auxiliary_loss_clip": 0.01159465,
      "auxiliary_loss_mlp": 0.01024423,
      "balance_loss_clip": 1.04529858,
      "balance_loss_mlp": 1.01728237,
      "epoch": 0.7678710996212349,
      "flos": 25411432222080.0,
      "grad_norm": 2.046150424050049,
      "language_loss": 0.7341246,
      "learning_rate": 5.389523814985562e-07,
      "loss": 0.75596344,
      "num_input_tokens_seen": 137413045,
      "step": 6386,
      "time_per_iteration": 2.51172137260437
    },
    {
      "auxiliary_loss_clip": 0.01104347,
      "auxiliary_loss_mlp": 0.01024016,
      "balance_loss_clip": 1.040097,
      "balance_loss_mlp": 1.01717091,
      "epoch": 0.767991342511874,
      "flos": 26756825063040.0,
      "grad_norm": 1.9497956210470164,
      "language_loss": 0.76289546,
      "learning_rate": 5.384205425441344e-07,
      "loss": 0.78417915,
      "num_input_tokens_seen": 137433955,
      "step": 6387,
      "time_per_iteration": 2.6410443782806396
    },
    {
      "auxiliary_loss_clip": 0.0113843,
      "auxiliary_loss_mlp": 0.01022627,
      "balance_loss_clip": 1.04460013,
      "balance_loss_mlp": 1.01595163,
      "epoch": 0.7681115854025131,
      "flos": 26359509749760.0,
      "grad_norm": 1.641148196280358,
      "language_loss": 0.84357321,
      "learning_rate": 5.378889253101537e-07,
      "loss": 0.86518371,
      "num_input_tokens_seen": 137454510,
      "step": 6388,
      "time_per_iteration": 2.5982351303100586
    },
    {
      "auxiliary_loss_clip": 0.01148324,
      "auxiliary_loss_mlp": 0.01023014,
      "balance_loss_clip": 1.04364121,
      "balance_loss_mlp": 1.0164305,
      "epoch": 0.7682318282931522,
      "flos": 23257043314560.0,
      "grad_norm": 1.720035926421481,
      "language_loss": 0.808752,
      "learning_rate": 5.373575298772617e-07,
      "loss": 0.83046538,
      "num_input_tokens_seen": 137473630,
      "step": 6389,
      "time_per_iteration": 2.508378267288208
    },
    {
      "auxiliary_loss_clip": 0.01047613,
      "auxiliary_loss_mlp": 0.01000535,
      "balance_loss_clip": 1.00859785,
      "balance_loss_mlp": 0.99942034,
      "epoch": 0.7683520711837912,
      "flos": 70072457137920.0,
      "grad_norm": 0.7552759285509741,
      "language_loss": 0.61352092,
      "learning_rate": 5.368263563260689e-07,
      "loss": 0.63400239,
      "num_input_tokens_seen": 137538765,
      "step": 6390,
      "time_per_iteration": 3.1872360706329346
    },
    {
      "auxiliary_loss_clip": 0.01148478,
      "auxiliary_loss_mlp": 0.010237,
      "balance_loss_clip": 1.04422522,
      "balance_loss_mlp": 1.01702976,
      "epoch": 0.7684723140744304,
      "flos": 18624890332800.0,
      "grad_norm": 1.4949110052754477,
      "language_loss": 0.64542645,
      "learning_rate": 5.362954047371537e-07,
      "loss": 0.66714823,
      "num_input_tokens_seen": 137557875,
      "step": 6391,
      "time_per_iteration": 2.482348918914795
    },
    {
      "auxiliary_loss_clip": 0.01121485,
      "auxiliary_loss_mlp": 0.01024868,
      "balance_loss_clip": 1.04714584,
      "balance_loss_mlp": 1.01733446,
      "epoch": 0.7685925569650695,
      "flos": 27452989532160.0,
      "grad_norm": 1.8230751769463514,
      "language_loss": 0.71946347,
      "learning_rate": 5.357646751910627e-07,
      "loss": 0.74092698,
      "num_input_tokens_seen": 137579055,
      "step": 6392,
      "time_per_iteration": 3.3394582271575928
    },
    {
      "auxiliary_loss_clip": 0.01131201,
      "auxiliary_loss_mlp": 0.01022773,
      "balance_loss_clip": 1.0419805,
      "balance_loss_mlp": 1.01556396,
      "epoch": 0.7687127998557085,
      "flos": 24535714642560.0,
      "grad_norm": 2.0285053753777924,
      "language_loss": 0.79819858,
      "learning_rate": 5.352341677683061e-07,
      "loss": 0.81973827,
      "num_input_tokens_seen": 137600355,
      "step": 6393,
      "time_per_iteration": 2.549882411956787
    },
    {
      "auxiliary_loss_clip": 0.01128285,
      "auxiliary_loss_mlp": 0.01027412,
      "balance_loss_clip": 1.04348516,
      "balance_loss_mlp": 1.02056003,
      "epoch": 0.7688330427463477,
      "flos": 25155963717120.0,
      "grad_norm": 1.8587208202714947,
      "language_loss": 0.78894484,
      "learning_rate": 5.347038825493617e-07,
      "loss": 0.81050175,
      "num_input_tokens_seen": 137621885,
      "step": 6394,
      "time_per_iteration": 2.5968708992004395
    },
    {
      "auxiliary_loss_clip": 0.0113332,
      "auxiliary_loss_mlp": 0.01025156,
      "balance_loss_clip": 1.04653466,
      "balance_loss_mlp": 1.01850414,
      "epoch": 0.7689532856369867,
      "flos": 21211284113280.0,
      "grad_norm": 2.089386710532059,
      "language_loss": 0.6866132,
      "learning_rate": 5.341738196146732e-07,
      "loss": 0.70819795,
      "num_input_tokens_seen": 137640230,
      "step": 6395,
      "time_per_iteration": 2.532658338546753
    },
    {
      "auxiliary_loss_clip": 0.01144332,
      "auxiliary_loss_mlp": 0.01022543,
      "balance_loss_clip": 1.04212439,
      "balance_loss_mlp": 1.01548553,
      "epoch": 0.7690735285276258,
      "flos": 25119083427840.0,
      "grad_norm": 2.105038495461312,
      "language_loss": 0.73342848,
      "learning_rate": 5.336439790446503e-07,
      "loss": 0.75509727,
      "num_input_tokens_seen": 137659330,
      "step": 6396,
      "time_per_iteration": 2.65083384513855
    },
    {
      "auxiliary_loss_clip": 0.01117601,
      "auxiliary_loss_mlp": 0.01023711,
      "balance_loss_clip": 1.04098105,
      "balance_loss_mlp": 1.01704383,
      "epoch": 0.769193771418265,
      "flos": 54744020640000.0,
      "grad_norm": 1.8633060771115304,
      "language_loss": 0.63061708,
      "learning_rate": 5.331143609196711e-07,
      "loss": 0.65203017,
      "num_input_tokens_seen": 137683145,
      "step": 6397,
      "time_per_iteration": 2.9449517726898193
    },
    {
      "auxiliary_loss_clip": 0.01150394,
      "auxiliary_loss_mlp": 0.01025051,
      "balance_loss_clip": 1.04847085,
      "balance_loss_mlp": 1.01824379,
      "epoch": 0.769314014308904,
      "flos": 37341890115840.0,
      "grad_norm": 1.9527778582427457,
      "language_loss": 0.77427357,
      "learning_rate": 5.325849653200758e-07,
      "loss": 0.79602802,
      "num_input_tokens_seen": 137707095,
      "step": 6398,
      "time_per_iteration": 2.621087074279785
    },
    {
      "auxiliary_loss_clip": 0.01164048,
      "auxiliary_loss_mlp": 0.01027977,
      "balance_loss_clip": 1.04842067,
      "balance_loss_mlp": 1.02089012,
      "epoch": 0.7694342571995431,
      "flos": 20631686256000.0,
      "grad_norm": 1.6423741286904339,
      "language_loss": 0.76109594,
      "learning_rate": 5.32055792326175e-07,
      "loss": 0.7830162,
      "num_input_tokens_seen": 137725520,
      "step": 6399,
      "time_per_iteration": 2.490973949432373
    },
    {
      "auxiliary_loss_clip": 0.01138871,
      "auxiliary_loss_mlp": 0.01025488,
      "balance_loss_clip": 1.0470525,
      "balance_loss_mlp": 1.01831794,
      "epoch": 0.7695545000901821,
      "flos": 24207706621440.0,
      "grad_norm": 1.8466481262118786,
      "language_loss": 0.72717899,
      "learning_rate": 5.315268420182437e-07,
      "loss": 0.74882263,
      "num_input_tokens_seen": 137744195,
      "step": 6400,
      "time_per_iteration": 2.544527530670166
    },
    {
      "auxiliary_loss_clip": 0.01129093,
      "auxiliary_loss_mlp": 0.00760376,
      "balance_loss_clip": 1.04442906,
      "balance_loss_mlp": 1.00022554,
      "epoch": 0.7696747429808213,
      "flos": 28001273708160.0,
      "grad_norm": 2.232961975106676,
      "language_loss": 0.76262093,
      "learning_rate": 5.309981144765221e-07,
      "loss": 0.7815156,
      "num_input_tokens_seen": 137764340,
      "step": 6401,
      "time_per_iteration": 2.627885580062866
    },
    {
      "auxiliary_loss_clip": 0.01114566,
      "auxiliary_loss_mlp": 0.01027527,
      "balance_loss_clip": 1.04159391,
      "balance_loss_mlp": 1.02029145,
      "epoch": 0.7697949858714603,
      "flos": 11509550323200.0,
      "grad_norm": 2.0256091307156545,
      "language_loss": 0.75303411,
      "learning_rate": 5.304696097812196e-07,
      "loss": 0.77445507,
      "num_input_tokens_seen": 137780940,
      "step": 6402,
      "time_per_iteration": 3.3409650325775146
    },
    {
      "auxiliary_loss_clip": 0.01134182,
      "auxiliary_loss_mlp": 0.01032328,
      "balance_loss_clip": 1.04534316,
      "balance_loss_mlp": 1.02459764,
      "epoch": 0.7699152287620994,
      "flos": 26688271956480.0,
      "grad_norm": 2.735328713060162,
      "language_loss": 0.60040563,
      "learning_rate": 5.299413280125078e-07,
      "loss": 0.62207079,
      "num_input_tokens_seen": 137799250,
      "step": 6403,
      "time_per_iteration": 2.5599632263183594
    },
    {
      "auxiliary_loss_clip": 0.01136219,
      "auxiliary_loss_mlp": 0.01023727,
      "balance_loss_clip": 1.04343772,
      "balance_loss_mlp": 1.01704824,
      "epoch": 0.7700354716527386,
      "flos": 16544944362240.0,
      "grad_norm": 1.8590350838044205,
      "language_loss": 0.72395968,
      "learning_rate": 5.294132692505284e-07,
      "loss": 0.74555922,
      "num_input_tokens_seen": 137817660,
      "step": 6404,
      "time_per_iteration": 4.044933795928955
    },
    {
      "auxiliary_loss_clip": 0.01097687,
      "auxiliary_loss_mlp": 0.01029957,
      "balance_loss_clip": 1.03825092,
      "balance_loss_mlp": 1.02313209,
      "epoch": 0.7701557145433776,
      "flos": 19242733196160.0,
      "grad_norm": 1.8326223960576682,
      "language_loss": 0.79349428,
      "learning_rate": 5.288854335753861e-07,
      "loss": 0.8147707,
      "num_input_tokens_seen": 137835920,
      "step": 6405,
      "time_per_iteration": 2.559601306915283
    },
    {
      "auxiliary_loss_clip": 0.01152628,
      "auxiliary_loss_mlp": 0.01020785,
      "balance_loss_clip": 1.04664242,
      "balance_loss_mlp": 1.01414835,
      "epoch": 0.7702759574340167,
      "flos": 31685744211840.0,
      "grad_norm": 1.5924572349572663,
      "language_loss": 0.75516945,
      "learning_rate": 5.283578210671551e-07,
      "loss": 0.77690351,
      "num_input_tokens_seen": 137858160,
      "step": 6406,
      "time_per_iteration": 2.58896803855896
    },
    {
      "auxiliary_loss_clip": 0.01140248,
      "auxiliary_loss_mlp": 0.01021549,
      "balance_loss_clip": 1.04620063,
      "balance_loss_mlp": 1.01546907,
      "epoch": 0.7703962003246558,
      "flos": 16800089644800.0,
      "grad_norm": 1.901077472047307,
      "language_loss": 0.76306444,
      "learning_rate": 5.278304318058719e-07,
      "loss": 0.78468239,
      "num_input_tokens_seen": 137876015,
      "step": 6407,
      "time_per_iteration": 2.5181946754455566
    },
    {
      "auxiliary_loss_clip": 0.01091703,
      "auxiliary_loss_mlp": 0.01026798,
      "balance_loss_clip": 1.03870416,
      "balance_loss_mlp": 1.01963329,
      "epoch": 0.7705164432152949,
      "flos": 35736072693120.0,
      "grad_norm": 1.843010964209276,
      "language_loss": 0.78982174,
      "learning_rate": 5.273032658715411e-07,
      "loss": 0.81100667,
      "num_input_tokens_seen": 137898825,
      "step": 6408,
      "time_per_iteration": 2.7308335304260254
    },
    {
      "auxiliary_loss_clip": 0.01105775,
      "auxiliary_loss_mlp": 0.01028235,
      "balance_loss_clip": 1.04184008,
      "balance_loss_mlp": 1.02145231,
      "epoch": 0.7706366861059339,
      "flos": 23365960329600.0,
      "grad_norm": 2.0453285610750833,
      "language_loss": 0.76972187,
      "learning_rate": 5.267763233441347e-07,
      "loss": 0.791062,
      "num_input_tokens_seen": 137919455,
      "step": 6409,
      "time_per_iteration": 2.65564227104187
    },
    {
      "auxiliary_loss_clip": 0.01153403,
      "auxiliary_loss_mlp": 0.01029903,
      "balance_loss_clip": 1.04801106,
      "balance_loss_mlp": 1.02208638,
      "epoch": 0.7707569289965731,
      "flos": 22929897219840.0,
      "grad_norm": 2.0551105671964325,
      "language_loss": 0.69387698,
      "learning_rate": 5.26249604303588e-07,
      "loss": 0.71571004,
      "num_input_tokens_seen": 137937960,
      "step": 6410,
      "time_per_iteration": 2.5191636085510254
    },
    {
      "auxiliary_loss_clip": 0.01164509,
      "auxiliary_loss_mlp": 0.01025869,
      "balance_loss_clip": 1.04845619,
      "balance_loss_mlp": 1.01937795,
      "epoch": 0.7708771718872122,
      "flos": 17420661941760.0,
      "grad_norm": 2.0448114769913928,
      "language_loss": 0.78078723,
      "learning_rate": 5.257231088298057e-07,
      "loss": 0.80269098,
      "num_input_tokens_seen": 137956370,
      "step": 6411,
      "time_per_iteration": 2.4659805297851562
    },
    {
      "auxiliary_loss_clip": 0.0102258,
      "auxiliary_loss_mlp": 0.01003078,
      "balance_loss_clip": 1.00798464,
      "balance_loss_mlp": 1.00195754,
      "epoch": 0.7709974147778512,
      "flos": 72241316248320.0,
      "grad_norm": 0.8182572661422302,
      "language_loss": 0.54007554,
      "learning_rate": 5.25196837002655e-07,
      "loss": 0.56033218,
      "num_input_tokens_seen": 138016080,
      "step": 6412,
      "time_per_iteration": 3.186418294906616
    },
    {
      "auxiliary_loss_clip": 0.01132906,
      "auxiliary_loss_mlp": 0.0103217,
      "balance_loss_clip": 1.04477417,
      "balance_loss_mlp": 1.02481794,
      "epoch": 0.7711176576684904,
      "flos": 39859694876160.0,
      "grad_norm": 2.1684231501459577,
      "language_loss": 0.68583512,
      "learning_rate": 5.24670788901971e-07,
      "loss": 0.7074858,
      "num_input_tokens_seen": 138039170,
      "step": 6413,
      "time_per_iteration": 2.699737787246704
    },
    {
      "auxiliary_loss_clip": 0.01136456,
      "auxiliary_loss_mlp": 0.01032307,
      "balance_loss_clip": 1.04768336,
      "balance_loss_mlp": 1.02450204,
      "epoch": 0.7712379005591294,
      "flos": 36976391274240.0,
      "grad_norm": 2.066212159095324,
      "language_loss": 0.6813333,
      "learning_rate": 5.241449646075557e-07,
      "loss": 0.70302093,
      "num_input_tokens_seen": 138062395,
      "step": 6414,
      "time_per_iteration": 2.651027202606201
    },
    {
      "auxiliary_loss_clip": 0.01156198,
      "auxiliary_loss_mlp": 0.01028671,
      "balance_loss_clip": 1.04643333,
      "balance_loss_mlp": 1.02149129,
      "epoch": 0.7713581434497685,
      "flos": 22776773541120.0,
      "grad_norm": 2.0726864616829515,
      "language_loss": 0.72804213,
      "learning_rate": 5.236193641991762e-07,
      "loss": 0.7498908,
      "num_input_tokens_seen": 138080325,
      "step": 6415,
      "time_per_iteration": 2.5300230979919434
    },
    {
      "auxiliary_loss_clip": 0.0113531,
      "auxiliary_loss_mlp": 0.01025471,
      "balance_loss_clip": 1.04646349,
      "balance_loss_mlp": 1.01851773,
      "epoch": 0.7714783863404077,
      "flos": 24097460803200.0,
      "grad_norm": 1.7511401401440978,
      "language_loss": 0.69727266,
      "learning_rate": 5.23093987756565e-07,
      "loss": 0.71888053,
      "num_input_tokens_seen": 138099020,
      "step": 6416,
      "time_per_iteration": 2.557650566101074
    },
    {
      "auxiliary_loss_clip": 0.01126206,
      "auxiliary_loss_mlp": 0.01032733,
      "balance_loss_clip": 1.03991628,
      "balance_loss_mlp": 1.02537763,
      "epoch": 0.7715986292310467,
      "flos": 21063655215360.0,
      "grad_norm": 1.658660304565217,
      "language_loss": 0.75392139,
      "learning_rate": 5.225688353594217e-07,
      "loss": 0.77551079,
      "num_input_tokens_seen": 138118650,
      "step": 6417,
      "time_per_iteration": 2.6053831577301025
    },
    {
      "auxiliary_loss_clip": 0.01143042,
      "auxiliary_loss_mlp": 0.00760549,
      "balance_loss_clip": 1.04745722,
      "balance_loss_mlp": 1.00021398,
      "epoch": 0.7717188721216858,
      "flos": 20594877793920.0,
      "grad_norm": 2.103647854230917,
      "language_loss": 0.77558756,
      "learning_rate": 5.220439070874108e-07,
      "loss": 0.79462343,
      "num_input_tokens_seen": 138137890,
      "step": 6418,
      "time_per_iteration": 2.558582067489624
    },
    {
      "auxiliary_loss_clip": 0.0115341,
      "auxiliary_loss_mlp": 0.01022986,
      "balance_loss_clip": 1.04994786,
      "balance_loss_mlp": 1.01620269,
      "epoch": 0.7718391150123249,
      "flos": 26250951870720.0,
      "grad_norm": 2.7691769409386744,
      "language_loss": 0.7079711,
      "learning_rate": 5.215192030201652e-07,
      "loss": 0.72973508,
      "num_input_tokens_seen": 138158880,
      "step": 6419,
      "time_per_iteration": 3.346935987472534
    },
    {
      "auxiliary_loss_clip": 0.01109081,
      "auxiliary_loss_mlp": 0.01027726,
      "balance_loss_clip": 1.03940547,
      "balance_loss_mlp": 1.02097869,
      "epoch": 0.771959357902964,
      "flos": 22049762267520.0,
      "grad_norm": 1.7082350388563132,
      "language_loss": 0.86181515,
      "learning_rate": 5.209947232372798e-07,
      "loss": 0.88318324,
      "num_input_tokens_seen": 138176370,
      "step": 6420,
      "time_per_iteration": 2.585092306137085
    },
    {
      "auxiliary_loss_clip": 0.01153634,
      "auxiliary_loss_mlp": 0.00760601,
      "balance_loss_clip": 1.04614663,
      "balance_loss_mlp": 1.00023079,
      "epoch": 0.772079600793603,
      "flos": 30446000248320.0,
      "grad_norm": 1.6230721827542016,
      "language_loss": 0.81024241,
      "learning_rate": 5.204704678183196e-07,
      "loss": 0.82938474,
      "num_input_tokens_seen": 138195105,
      "step": 6421,
      "time_per_iteration": 2.582650899887085
    },
    {
      "auxiliary_loss_clip": 0.01166626,
      "auxiliary_loss_mlp": 0.01022434,
      "balance_loss_clip": 1.04990768,
      "balance_loss_mlp": 1.01556134,
      "epoch": 0.7721998436842422,
      "flos": 12969857750400.0,
      "grad_norm": 1.7361202300143974,
      "language_loss": 0.84869528,
      "learning_rate": 5.19946436842813e-07,
      "loss": 0.87058592,
      "num_input_tokens_seen": 138212235,
      "step": 6422,
      "time_per_iteration": 2.447450876235962
    },
    {
      "auxiliary_loss_clip": 0.01119401,
      "auxiliary_loss_mlp": 0.01023831,
      "balance_loss_clip": 1.04255271,
      "balance_loss_mlp": 1.01756048,
      "epoch": 0.7723200865748813,
      "flos": 32635509678720.0,
      "grad_norm": 1.648709464697266,
      "language_loss": 0.68407154,
      "learning_rate": 5.194226303902546e-07,
      "loss": 0.70550382,
      "num_input_tokens_seen": 138231970,
      "step": 6423,
      "time_per_iteration": 2.706016778945923
    },
    {
      "auxiliary_loss_clip": 0.01134844,
      "auxiliary_loss_mlp": 0.01023108,
      "balance_loss_clip": 1.04401445,
      "balance_loss_mlp": 1.01582432,
      "epoch": 0.7724403294655203,
      "flos": 21105707063040.0,
      "grad_norm": 1.7802655526290958,
      "language_loss": 0.7099728,
      "learning_rate": 5.188990485401072e-07,
      "loss": 0.73155236,
      "num_input_tokens_seen": 138251175,
      "step": 6424,
      "time_per_iteration": 2.542248249053955
    },
    {
      "auxiliary_loss_clip": 0.01151473,
      "auxiliary_loss_mlp": 0.01023799,
      "balance_loss_clip": 1.04828477,
      "balance_loss_mlp": 1.0168308,
      "epoch": 0.7725605723561595,
      "flos": 22090736707200.0,
      "grad_norm": 1.745704902969292,
      "language_loss": 0.86208785,
      "learning_rate": 5.183756913717954e-07,
      "loss": 0.88384056,
      "num_input_tokens_seen": 138270950,
      "step": 6425,
      "time_per_iteration": 2.5610287189483643
    },
    {
      "auxiliary_loss_clip": 0.01133684,
      "auxiliary_loss_mlp": 0.01029883,
      "balance_loss_clip": 1.04556513,
      "balance_loss_mlp": 1.02280152,
      "epoch": 0.7726808152467985,
      "flos": 34495610457600.0,
      "grad_norm": 1.8344887933971807,
      "language_loss": 0.72737086,
      "learning_rate": 5.178525589647136e-07,
      "loss": 0.74900651,
      "num_input_tokens_seen": 138292590,
      "step": 6426,
      "time_per_iteration": 2.6384189128875732
    },
    {
      "auxiliary_loss_clip": 0.01140859,
      "auxiliary_loss_mlp": 0.01025843,
      "balance_loss_clip": 1.0448662,
      "balance_loss_mlp": 1.0193758,
      "epoch": 0.7728010581374376,
      "flos": 22306344094080.0,
      "grad_norm": 1.9843517447337005,
      "language_loss": 0.78756529,
      "learning_rate": 5.173296513982197e-07,
      "loss": 0.80923229,
      "num_input_tokens_seen": 138311115,
      "step": 6427,
      "time_per_iteration": 2.556851625442505
    },
    {
      "auxiliary_loss_clip": 0.01132659,
      "auxiliary_loss_mlp": 0.01027297,
      "balance_loss_clip": 1.04557729,
      "balance_loss_mlp": 1.01999497,
      "epoch": 0.7729213010280768,
      "flos": 27126453968640.0,
      "grad_norm": 1.8882902076952024,
      "language_loss": 0.64927369,
      "learning_rate": 5.168069687516398e-07,
      "loss": 0.67087322,
      "num_input_tokens_seen": 138330885,
      "step": 6428,
      "time_per_iteration": 3.4023709297180176
    },
    {
      "auxiliary_loss_clip": 0.01138508,
      "auxiliary_loss_mlp": 0.01026906,
      "balance_loss_clip": 1.04870272,
      "balance_loss_mlp": 1.01983047,
      "epoch": 0.7730415439187158,
      "flos": 18150223080960.0,
      "grad_norm": 2.74337789418084,
      "language_loss": 0.7161448,
      "learning_rate": 5.16284511104263e-07,
      "loss": 0.73779905,
      "num_input_tokens_seen": 138350020,
      "step": 6429,
      "time_per_iteration": 2.5161678791046143
    },
    {
      "auxiliary_loss_clip": 0.01134852,
      "auxiliary_loss_mlp": 0.0103215,
      "balance_loss_clip": 1.0449475,
      "balance_loss_mlp": 1.02482438,
      "epoch": 0.7731617868093549,
      "flos": 11947480940160.0,
      "grad_norm": 3.0284045482112756,
      "language_loss": 0.80839634,
      "learning_rate": 5.157622785353457e-07,
      "loss": 0.83006638,
      "num_input_tokens_seen": 138368135,
      "step": 6430,
      "time_per_iteration": 4.072638034820557
    },
    {
      "auxiliary_loss_clip": 0.01047942,
      "auxiliary_loss_mlp": 0.01000345,
      "balance_loss_clip": 1.00930727,
      "balance_loss_mlp": 0.99923021,
      "epoch": 0.7732820296999939,
      "flos": 64201027069440.0,
      "grad_norm": 0.6432247736178279,
      "language_loss": 0.60389096,
      "learning_rate": 5.152402711241113e-07,
      "loss": 0.62437385,
      "num_input_tokens_seen": 138436040,
      "step": 6431,
      "time_per_iteration": 3.1827900409698486
    },
    {
      "auxiliary_loss_clip": 0.01118437,
      "auxiliary_loss_mlp": 0.01028451,
      "balance_loss_clip": 1.04126477,
      "balance_loss_mlp": 1.02193618,
      "epoch": 0.7734022725906331,
      "flos": 25302191984640.0,
      "grad_norm": 1.5558118230906002,
      "language_loss": 0.82793123,
      "learning_rate": 5.147184889497465e-07,
      "loss": 0.84940004,
      "num_input_tokens_seen": 138455510,
      "step": 6432,
      "time_per_iteration": 2.604698419570923
    },
    {
      "auxiliary_loss_clip": 0.01114735,
      "auxiliary_loss_mlp": 0.01025324,
      "balance_loss_clip": 1.04282355,
      "balance_loss_mlp": 1.0177896,
      "epoch": 0.7735225154812722,
      "flos": 17347440067200.0,
      "grad_norm": 2.3935628858707276,
      "language_loss": 0.80139041,
      "learning_rate": 5.141969320914072e-07,
      "loss": 0.82279098,
      "num_input_tokens_seen": 138473015,
      "step": 6433,
      "time_per_iteration": 2.531937837600708
    },
    {
      "auxiliary_loss_clip": 0.01165549,
      "auxiliary_loss_mlp": 0.01028293,
      "balance_loss_clip": 1.04681587,
      "balance_loss_mlp": 1.02007949,
      "epoch": 0.7736427583719112,
      "flos": 32630086725120.0,
      "grad_norm": 6.152047815374636,
      "language_loss": 0.62326634,
      "learning_rate": 5.136756006282113e-07,
      "loss": 0.64520478,
      "num_input_tokens_seen": 138491680,
      "step": 6434,
      "time_per_iteration": 2.569178342819214
    },
    {
      "auxiliary_loss_clip": 0.0116895,
      "auxiliary_loss_mlp": 0.01038238,
      "balance_loss_clip": 1.05102611,
      "balance_loss_mlp": 1.03064442,
      "epoch": 0.7737630012625504,
      "flos": 19860073269120.0,
      "grad_norm": 2.2641667986546175,
      "language_loss": 0.85234588,
      "learning_rate": 5.131544946392446e-07,
      "loss": 0.87441778,
      "num_input_tokens_seen": 138506960,
      "step": 6435,
      "time_per_iteration": 2.438101291656494
    },
    {
      "auxiliary_loss_clip": 0.01134354,
      "auxiliary_loss_mlp": 0.01028188,
      "balance_loss_clip": 1.04618645,
      "balance_loss_mlp": 1.02090132,
      "epoch": 0.7738832441531894,
      "flos": 36022639397760.0,
      "grad_norm": 1.9214723098600501,
      "language_loss": 0.64094543,
      "learning_rate": 5.126336142035592e-07,
      "loss": 0.66257083,
      "num_input_tokens_seen": 138526995,
      "step": 6436,
      "time_per_iteration": 2.626328229904175
    },
    {
      "auxiliary_loss_clip": 0.01135347,
      "auxiliary_loss_mlp": 0.01024595,
      "balance_loss_clip": 1.04242635,
      "balance_loss_mlp": 1.0176096,
      "epoch": 0.7740034870438285,
      "flos": 13405274415360.0,
      "grad_norm": 2.643458051517162,
      "language_loss": 0.72056687,
      "learning_rate": 5.121129594001721e-07,
      "loss": 0.74216628,
      "num_input_tokens_seen": 138541260,
      "step": 6437,
      "time_per_iteration": 2.4923996925354004
    },
    {
      "auxiliary_loss_clip": 0.01151308,
      "auxiliary_loss_mlp": 0.01031448,
      "balance_loss_clip": 1.04861772,
      "balance_loss_mlp": 1.02452517,
      "epoch": 0.7741237299344677,
      "flos": 22086714384000.0,
      "grad_norm": 1.5540147291959931,
      "language_loss": 0.81087315,
      "learning_rate": 5.115925303080661e-07,
      "loss": 0.83270067,
      "num_input_tokens_seen": 138560970,
      "step": 6438,
      "time_per_iteration": 2.5081732273101807
    },
    {
      "auxiliary_loss_clip": 0.01134267,
      "auxiliary_loss_mlp": 0.01021012,
      "balance_loss_clip": 1.04573989,
      "balance_loss_mlp": 1.01431572,
      "epoch": 0.7742439728251067,
      "flos": 19864777950720.0,
      "grad_norm": 1.8593351752264253,
      "language_loss": 0.79330182,
      "learning_rate": 5.110723270061899e-07,
      "loss": 0.81485462,
      "num_input_tokens_seen": 138577460,
      "step": 6439,
      "time_per_iteration": 2.489534854888916
    },
    {
      "auxiliary_loss_clip": 0.01161588,
      "auxiliary_loss_mlp": 0.01021291,
      "balance_loss_clip": 1.04709148,
      "balance_loss_mlp": 1.01478815,
      "epoch": 0.7743642157157458,
      "flos": 16690167048960.0,
      "grad_norm": 3.0232928796060428,
      "language_loss": 0.79402894,
      "learning_rate": 5.105523495734572e-07,
      "loss": 0.81585777,
      "num_input_tokens_seen": 138594860,
      "step": 6440,
      "time_per_iteration": 2.412870407104492
    },
    {
      "auxiliary_loss_clip": 0.01165723,
      "auxiliary_loss_mlp": 0.01028723,
      "balance_loss_clip": 1.04735756,
      "balance_loss_mlp": 1.02139211,
      "epoch": 0.7744844586063849,
      "flos": 20304360593280.0,
      "grad_norm": 1.5748189080194266,
      "language_loss": 0.75033569,
      "learning_rate": 5.100325980887499e-07,
      "loss": 0.77228022,
      "num_input_tokens_seen": 138614785,
      "step": 6441,
      "time_per_iteration": 2.46769380569458
    },
    {
      "auxiliary_loss_clip": 0.01139224,
      "auxiliary_loss_mlp": 0.01028942,
      "balance_loss_clip": 1.04587317,
      "balance_loss_mlp": 1.02186716,
      "epoch": 0.774604701497024,
      "flos": 22966705681920.0,
      "grad_norm": 1.788238122188804,
      "language_loss": 0.82951927,
      "learning_rate": 5.095130726309116e-07,
      "loss": 0.85120094,
      "num_input_tokens_seen": 138634960,
      "step": 6442,
      "time_per_iteration": 2.532679557800293
    },
    {
      "auxiliary_loss_clip": 0.01056431,
      "auxiliary_loss_mlp": 0.01000065,
      "balance_loss_clip": 1.00878119,
      "balance_loss_mlp": 0.99903405,
      "epoch": 0.774724944387663,
      "flos": 60288523073280.0,
      "grad_norm": 0.8222596346774548,
      "language_loss": 0.59054834,
      "learning_rate": 5.089937732787559e-07,
      "loss": 0.61111331,
      "num_input_tokens_seen": 138699520,
      "step": 6443,
      "time_per_iteration": 3.130455255508423
    },
    {
      "auxiliary_loss_clip": 0.01122391,
      "auxiliary_loss_mlp": 0.01025891,
      "balance_loss_clip": 1.04310584,
      "balance_loss_mlp": 1.01872015,
      "epoch": 0.7748451872783022,
      "flos": 26761026954240.0,
      "grad_norm": 2.13058300491253,
      "language_loss": 0.66755062,
      "learning_rate": 5.084747001110592e-07,
      "loss": 0.68903345,
      "num_input_tokens_seen": 138719145,
      "step": 6444,
      "time_per_iteration": 2.6195971965789795
    },
    {
      "auxiliary_loss_clip": 0.01143521,
      "auxiliary_loss_mlp": 0.00760612,
      "balance_loss_clip": 1.04632807,
      "balance_loss_mlp": 1.00024664,
      "epoch": 0.7749654301689413,
      "flos": 30338627518080.0,
      "grad_norm": 1.673585102441682,
      "language_loss": 0.70054698,
      "learning_rate": 5.07955853206564e-07,
      "loss": 0.71958828,
      "num_input_tokens_seen": 138743850,
      "step": 6445,
      "time_per_iteration": 3.3109371662139893
    },
    {
      "auxiliary_loss_clip": 0.01155202,
      "auxiliary_loss_mlp": 0.01025571,
      "balance_loss_clip": 1.04847431,
      "balance_loss_mlp": 1.01861846,
      "epoch": 0.7750856730595803,
      "flos": 43179851687040.0,
      "grad_norm": 2.2186866867521227,
      "language_loss": 0.71046883,
      "learning_rate": 5.074372326439807e-07,
      "loss": 0.73227656,
      "num_input_tokens_seen": 138766860,
      "step": 6446,
      "time_per_iteration": 2.687319278717041
    },
    {
      "auxiliary_loss_clip": 0.01122976,
      "auxiliary_loss_mlp": 0.01026665,
      "balance_loss_clip": 1.04376221,
      "balance_loss_mlp": 1.01953578,
      "epoch": 0.7752059159502195,
      "flos": 17640040256640.0,
      "grad_norm": 2.2778185383403025,
      "language_loss": 0.73152232,
      "learning_rate": 5.069188385019814e-07,
      "loss": 0.75301862,
      "num_input_tokens_seen": 138784560,
      "step": 6447,
      "time_per_iteration": 2.5269172191619873
    },
    {
      "auxiliary_loss_clip": 0.01115192,
      "auxiliary_loss_mlp": 0.01025683,
      "balance_loss_clip": 1.04189515,
      "balance_loss_mlp": 1.01873064,
      "epoch": 0.7753261588408585,
      "flos": 12677688524160.0,
      "grad_norm": 2.9163157008205243,
      "language_loss": 0.61134666,
      "learning_rate": 5.064006708592077e-07,
      "loss": 0.6327554,
      "num_input_tokens_seen": 138800805,
      "step": 6448,
      "time_per_iteration": 2.560739040374756
    },
    {
      "auxiliary_loss_clip": 0.0113244,
      "auxiliary_loss_mlp": 0.01024112,
      "balance_loss_clip": 1.0486325,
      "balance_loss_mlp": 1.01734102,
      "epoch": 0.7754464017314976,
      "flos": 16690741666560.0,
      "grad_norm": 2.6838110903232724,
      "language_loss": 0.75556868,
      "learning_rate": 5.058827297942641e-07,
      "loss": 0.77713418,
      "num_input_tokens_seen": 138815910,
      "step": 6449,
      "time_per_iteration": 2.4854915142059326
    },
    {
      "auxiliary_loss_clip": 0.01140167,
      "auxiliary_loss_mlp": 0.01025136,
      "balance_loss_clip": 1.04684615,
      "balance_loss_mlp": 1.01867521,
      "epoch": 0.7755666446221368,
      "flos": 19718944732800.0,
      "grad_norm": 1.8342279071406626,
      "language_loss": 0.75224912,
      "learning_rate": 5.053650153857237e-07,
      "loss": 0.77390218,
      "num_input_tokens_seen": 138834920,
      "step": 6450,
      "time_per_iteration": 2.524054765701294
    },
    {
      "auxiliary_loss_clip": 0.01149736,
      "auxiliary_loss_mlp": 0.01026742,
      "balance_loss_clip": 1.0473392,
      "balance_loss_mlp": 1.0192976,
      "epoch": 0.7756868875127758,
      "flos": 18693623007360.0,
      "grad_norm": 1.5923475685214625,
      "language_loss": 0.69798863,
      "learning_rate": 5.048475277121214e-07,
      "loss": 0.71975344,
      "num_input_tokens_seen": 138852135,
      "step": 6451,
      "time_per_iteration": 2.4632203578948975
    },
    {
      "auxiliary_loss_clip": 0.01152298,
      "auxiliary_loss_mlp": 0.01026285,
      "balance_loss_clip": 1.04740262,
      "balance_loss_mlp": 1.01887035,
      "epoch": 0.7758071304034149,
      "flos": 28404191543040.0,
      "grad_norm": 2.5577606357876497,
      "language_loss": 0.77165943,
      "learning_rate": 5.043302668519598e-07,
      "loss": 0.79344523,
      "num_input_tokens_seen": 138871470,
      "step": 6452,
      "time_per_iteration": 2.5650269985198975
    },
    {
      "auxiliary_loss_clip": 0.01152302,
      "auxiliary_loss_mlp": 0.01027062,
      "balance_loss_clip": 1.04604924,
      "balance_loss_mlp": 1.01990342,
      "epoch": 0.775927373294054,
      "flos": 20595344670720.0,
      "grad_norm": 1.9637669686093535,
      "language_loss": 0.72143769,
      "learning_rate": 5.038132328837079e-07,
      "loss": 0.7432313,
      "num_input_tokens_seen": 138889860,
      "step": 6453,
      "time_per_iteration": 2.4759116172790527
    },
    {
      "auxiliary_loss_clip": 0.01150506,
      "auxiliary_loss_mlp": 0.01022115,
      "balance_loss_clip": 1.04582405,
      "balance_loss_mlp": 1.01563597,
      "epoch": 0.7760476161846931,
      "flos": 22526368853760.0,
      "grad_norm": 1.9868925439199594,
      "language_loss": 0.736148,
      "learning_rate": 5.032964258857993e-07,
      "loss": 0.75787413,
      "num_input_tokens_seen": 138909955,
      "step": 6454,
      "time_per_iteration": 3.3098433017730713
    },
    {
      "auxiliary_loss_clip": 0.01148161,
      "auxiliary_loss_mlp": 0.01025139,
      "balance_loss_clip": 1.04271984,
      "balance_loss_mlp": 1.01811719,
      "epoch": 0.7761678590753321,
      "flos": 48651488403840.0,
      "grad_norm": 1.4834152017900126,
      "language_loss": 0.68058598,
      "learning_rate": 5.027798459366329e-07,
      "loss": 0.70231897,
      "num_input_tokens_seen": 138935320,
      "step": 6455,
      "time_per_iteration": 3.574043035507202
    },
    {
      "auxiliary_loss_clip": 0.01153371,
      "auxiliary_loss_mlp": 0.01026914,
      "balance_loss_clip": 1.04562211,
      "balance_loss_mlp": 1.0199194,
      "epoch": 0.7762881019659713,
      "flos": 26177047637760.0,
      "grad_norm": 1.5264554480203447,
      "language_loss": 0.63518417,
      "learning_rate": 5.02263493114573e-07,
      "loss": 0.65698701,
      "num_input_tokens_seen": 138957115,
      "step": 6456,
      "time_per_iteration": 2.5310778617858887
    },
    {
      "auxiliary_loss_clip": 0.01162072,
      "auxiliary_loss_mlp": 0.01021934,
      "balance_loss_clip": 1.04678488,
      "balance_loss_mlp": 1.01501691,
      "epoch": 0.7764083448566104,
      "flos": 20588341518720.0,
      "grad_norm": 2.2524415144245755,
      "language_loss": 0.77292264,
      "learning_rate": 5.017473674979502e-07,
      "loss": 0.79476267,
      "num_input_tokens_seen": 138973140,
      "step": 6457,
      "time_per_iteration": 3.2398018836975098
    },
    {
      "auxiliary_loss_clip": 0.01026852,
      "auxiliary_loss_mlp": 0.01003774,
      "balance_loss_clip": 1.01798153,
      "balance_loss_mlp": 1.00278425,
      "epoch": 0.7765285877472494,
      "flos": 67293078560640.0,
      "grad_norm": 0.7440096850543245,
      "language_loss": 0.58359283,
      "learning_rate": 5.01231469165061e-07,
      "loss": 0.60389912,
      "num_input_tokens_seen": 139028965,
      "step": 6458,
      "time_per_iteration": 3.0386314392089844
    },
    {
      "auxiliary_loss_clip": 0.01047829,
      "auxiliary_loss_mlp": 0.01000677,
      "balance_loss_clip": 1.00922894,
      "balance_loss_mlp": 0.99959797,
      "epoch": 0.7766488306378886,
      "flos": 61344476121600.0,
      "grad_norm": 0.8319243547413446,
      "language_loss": 0.56872571,
      "learning_rate": 5.007157981941663e-07,
      "loss": 0.58921081,
      "num_input_tokens_seen": 139094325,
      "step": 6459,
      "time_per_iteration": 3.192441701889038
    },
    {
      "auxiliary_loss_clip": 0.01037368,
      "auxiliary_loss_mlp": 0.01000069,
      "balance_loss_clip": 1.00914693,
      "balance_loss_mlp": 0.9989962,
      "epoch": 0.7767690735285276,
      "flos": 62946199393920.0,
      "grad_norm": 0.8774775426335493,
      "language_loss": 0.67527282,
      "learning_rate": 5.002003546634928e-07,
      "loss": 0.69564724,
      "num_input_tokens_seen": 139150425,
      "step": 6460,
      "time_per_iteration": 3.057231903076172
    },
    {
      "auxiliary_loss_clip": 0.01104617,
      "auxiliary_loss_mlp": 0.01030448,
      "balance_loss_clip": 1.04276097,
      "balance_loss_mlp": 1.02366793,
      "epoch": 0.7768893164191667,
      "flos": 20886400575360.0,
      "grad_norm": 1.700649900150202,
      "language_loss": 0.76038468,
      "learning_rate": 4.996851386512331e-07,
      "loss": 0.78173536,
      "num_input_tokens_seen": 139169130,
      "step": 6461,
      "time_per_iteration": 2.616244077682495
    },
    {
      "auxiliary_loss_clip": 0.01139873,
      "auxiliary_loss_mlp": 0.01026168,
      "balance_loss_clip": 1.04706526,
      "balance_loss_mlp": 1.01874769,
      "epoch": 0.7770095593098058,
      "flos": 20704584908160.0,
      "grad_norm": 2.3100777882831904,
      "language_loss": 0.83089393,
      "learning_rate": 4.991701502355444e-07,
      "loss": 0.85255432,
      "num_input_tokens_seen": 139189595,
      "step": 6462,
      "time_per_iteration": 2.5306806564331055
    },
    {
      "auxiliary_loss_clip": 0.01153385,
      "auxiliary_loss_mlp": 0.01021507,
      "balance_loss_clip": 1.04579258,
      "balance_loss_mlp": 1.01485848,
      "epoch": 0.7771298022004449,
      "flos": 24717709877760.0,
      "grad_norm": 1.4963032764702198,
      "language_loss": 0.75891149,
      "learning_rate": 4.986553894945518e-07,
      "loss": 0.78066039,
      "num_input_tokens_seen": 139210805,
      "step": 6463,
      "time_per_iteration": 2.51418399810791
    },
    {
      "auxiliary_loss_clip": 0.01106868,
      "auxiliary_loss_mlp": 0.01026468,
      "balance_loss_clip": 1.0425036,
      "balance_loss_mlp": 1.01943731,
      "epoch": 0.777250045091084,
      "flos": 25009232659200.0,
      "grad_norm": 1.8613418117524225,
      "language_loss": 0.8595984,
      "learning_rate": 4.981408565063416e-07,
      "loss": 0.88093174,
      "num_input_tokens_seen": 139230750,
      "step": 6464,
      "time_per_iteration": 2.625422954559326
    },
    {
      "auxiliary_loss_clip": 0.01166157,
      "auxiliary_loss_mlp": 0.01024454,
      "balance_loss_clip": 1.04825187,
      "balance_loss_mlp": 1.01739943,
      "epoch": 0.777370287981723,
      "flos": 20119887319680.0,
      "grad_norm": 1.9144009612590764,
      "language_loss": 0.7635411,
      "learning_rate": 4.976265513489701e-07,
      "loss": 0.78544724,
      "num_input_tokens_seen": 139250720,
      "step": 6465,
      "time_per_iteration": 2.472801923751831
    },
    {
      "auxiliary_loss_clip": 0.01150905,
      "auxiliary_loss_mlp": 0.01025317,
      "balance_loss_clip": 1.04550076,
      "balance_loss_mlp": 1.01811635,
      "epoch": 0.7774905308723622,
      "flos": 21718809331200.0,
      "grad_norm": 1.6857516502403274,
      "language_loss": 0.80172974,
      "learning_rate": 4.971124741004562e-07,
      "loss": 0.82349193,
      "num_input_tokens_seen": 139269720,
      "step": 6466,
      "time_per_iteration": 2.5214762687683105
    },
    {
      "auxiliary_loss_clip": 0.01148178,
      "auxiliary_loss_mlp": 0.01023648,
      "balance_loss_clip": 1.04461098,
      "balance_loss_mlp": 1.0169518,
      "epoch": 0.7776107737630013,
      "flos": 16034115093120.0,
      "grad_norm": 1.732008655719451,
      "language_loss": 0.76075304,
      "learning_rate": 4.965986248387846e-07,
      "loss": 0.7824713,
      "num_input_tokens_seen": 139288035,
      "step": 6467,
      "time_per_iteration": 2.4609971046447754
    },
    {
      "auxiliary_loss_clip": 0.01138162,
      "auxiliary_loss_mlp": 0.01025954,
      "balance_loss_clip": 1.04367948,
      "balance_loss_mlp": 1.0193584,
      "epoch": 0.7777310166536403,
      "flos": 24790895838720.0,
      "grad_norm": 1.7664616536605562,
      "language_loss": 0.76964509,
      "learning_rate": 4.960850036419073e-07,
      "loss": 0.79128623,
      "num_input_tokens_seen": 139307135,
      "step": 6468,
      "time_per_iteration": 2.551643133163452
    },
    {
      "auxiliary_loss_clip": 0.01132998,
      "auxiliary_loss_mlp": 0.01028477,
      "balance_loss_clip": 1.04431021,
      "balance_loss_mlp": 1.02121437,
      "epoch": 0.7778512595442795,
      "flos": 17272530253440.0,
      "grad_norm": 1.8255918376108202,
      "language_loss": 0.78555775,
      "learning_rate": 4.955716105877378e-07,
      "loss": 0.80717242,
      "num_input_tokens_seen": 139325905,
      "step": 6469,
      "time_per_iteration": 2.483099937438965
    },
    {
      "auxiliary_loss_clip": 0.01153621,
      "auxiliary_loss_mlp": 0.00760229,
      "balance_loss_clip": 1.04591572,
      "balance_loss_mlp": 1.00027823,
      "epoch": 0.7779715024349185,
      "flos": 17748418567680.0,
      "grad_norm": 1.6075458413453387,
      "language_loss": 0.82742143,
      "learning_rate": 4.950584457541598e-07,
      "loss": 0.84656,
      "num_input_tokens_seen": 139344370,
      "step": 6470,
      "time_per_iteration": 2.4863955974578857
    },
    {
      "auxiliary_loss_clip": 0.01152686,
      "auxiliary_loss_mlp": 0.01030816,
      "balance_loss_clip": 1.0468657,
      "balance_loss_mlp": 1.02428889,
      "epoch": 0.7780917453255576,
      "flos": 24316875031680.0,
      "grad_norm": 1.3959176170436078,
      "language_loss": 0.82025301,
      "learning_rate": 4.945455092190183e-07,
      "loss": 0.84208798,
      "num_input_tokens_seen": 139365625,
      "step": 6471,
      "time_per_iteration": 3.3291244506835938
    },
    {
      "auxiliary_loss_clip": 0.01056385,
      "auxiliary_loss_mlp": 0.01001529,
      "balance_loss_clip": 1.00851607,
      "balance_loss_mlp": 1.00045574,
      "epoch": 0.7782119882161967,
      "flos": 56364601530240.0,
      "grad_norm": 0.6769566640865966,
      "language_loss": 0.55970228,
      "learning_rate": 4.940328010601271e-07,
      "loss": 0.58028144,
      "num_input_tokens_seen": 139430540,
      "step": 6472,
      "time_per_iteration": 3.0740015506744385
    },
    {
      "auxiliary_loss_clip": 0.01145105,
      "auxiliary_loss_mlp": 0.01033912,
      "balance_loss_clip": 1.05157852,
      "balance_loss_mlp": 1.02642262,
      "epoch": 0.7783322311068358,
      "flos": 46789986994560.0,
      "grad_norm": 1.7784367275674553,
      "language_loss": 0.76671809,
      "learning_rate": 4.935203213552621e-07,
      "loss": 0.7885083,
      "num_input_tokens_seen": 139454280,
      "step": 6473,
      "time_per_iteration": 2.732941150665283
    },
    {
      "auxiliary_loss_clip": 0.01134981,
      "auxiliary_loss_mlp": 0.01024802,
      "balance_loss_clip": 1.04382741,
      "balance_loss_mlp": 1.01744699,
      "epoch": 0.7784524739974749,
      "flos": 19057864872960.0,
      "grad_norm": 2.0778088142655906,
      "language_loss": 0.66937625,
      "learning_rate": 4.930080701821662e-07,
      "loss": 0.69097406,
      "num_input_tokens_seen": 139471745,
      "step": 6474,
      "time_per_iteration": 2.5188586711883545
    },
    {
      "auxiliary_loss_clip": 0.01137662,
      "auxiliary_loss_mlp": 0.01027282,
      "balance_loss_clip": 1.04541707,
      "balance_loss_mlp": 1.01996541,
      "epoch": 0.778572716888114,
      "flos": 24791111320320.0,
      "grad_norm": 2.0099915191027966,
      "language_loss": 0.77219188,
      "learning_rate": 4.92496047618548e-07,
      "loss": 0.79384124,
      "num_input_tokens_seen": 139491505,
      "step": 6475,
      "time_per_iteration": 2.556781053543091
    },
    {
      "auxiliary_loss_clip": 0.01153962,
      "auxiliary_loss_mlp": 0.01030929,
      "balance_loss_clip": 1.04895759,
      "balance_loss_mlp": 1.02360916,
      "epoch": 0.7786929597787531,
      "flos": 20078086867200.0,
      "grad_norm": 1.8916079996294195,
      "language_loss": 0.77801442,
      "learning_rate": 4.919842537420811e-07,
      "loss": 0.79986334,
      "num_input_tokens_seen": 139508620,
      "step": 6476,
      "time_per_iteration": 2.5134615898132324
    },
    {
      "auxiliary_loss_clip": 0.01140752,
      "auxiliary_loss_mlp": 0.01032064,
      "balance_loss_clip": 1.04838943,
      "balance_loss_mlp": 1.02549255,
      "epoch": 0.7788132026693921,
      "flos": 21872220318720.0,
      "grad_norm": 1.8584801829312216,
      "language_loss": 0.78945428,
      "learning_rate": 4.91472688630404e-07,
      "loss": 0.8111825,
      "num_input_tokens_seen": 139529360,
      "step": 6477,
      "time_per_iteration": 2.5466794967651367
    },
    {
      "auxiliary_loss_clip": 0.01161121,
      "auxiliary_loss_mlp": 0.01029277,
      "balance_loss_clip": 1.04620206,
      "balance_loss_mlp": 1.02265167,
      "epoch": 0.7789334455600313,
      "flos": 11181937351680.0,
      "grad_norm": 1.5805027394379267,
      "language_loss": 0.73780751,
      "learning_rate": 4.909613523611202e-07,
      "loss": 0.7597115,
      "num_input_tokens_seen": 139546240,
      "step": 6478,
      "time_per_iteration": 2.469123601913452
    },
    {
      "auxiliary_loss_clip": 0.01099968,
      "auxiliary_loss_mlp": 0.0076102,
      "balance_loss_clip": 1.03942752,
      "balance_loss_mlp": 1.00025129,
      "epoch": 0.7790536884506704,
      "flos": 28695427015680.0,
      "grad_norm": 1.8155462318572995,
      "language_loss": 0.74473429,
      "learning_rate": 4.904502450117991e-07,
      "loss": 0.76334417,
      "num_input_tokens_seen": 139567200,
      "step": 6479,
      "time_per_iteration": 2.6692214012145996
    },
    {
      "auxiliary_loss_clip": 0.01131121,
      "auxiliary_loss_mlp": 0.01028179,
      "balance_loss_clip": 1.04614115,
      "balance_loss_mlp": 1.02145278,
      "epoch": 0.7791739313413094,
      "flos": 11072302064640.0,
      "grad_norm": 2.411159898019545,
      "language_loss": 0.72551942,
      "learning_rate": 4.899393666599762e-07,
      "loss": 0.74711239,
      "num_input_tokens_seen": 139583775,
      "step": 6480,
      "time_per_iteration": 3.3149571418762207
    },
    {
      "auxiliary_loss_clip": 0.01162186,
      "auxiliary_loss_mlp": 0.01024214,
      "balance_loss_clip": 1.0451926,
      "balance_loss_mlp": 1.0173502,
      "epoch": 0.7792941742319486,
      "flos": 14679276975360.0,
      "grad_norm": 1.9580919513513324,
      "language_loss": 0.72624886,
      "learning_rate": 4.894287173831506e-07,
      "loss": 0.74811286,
      "num_input_tokens_seen": 139599735,
      "step": 6481,
      "time_per_iteration": 3.145453691482544
    },
    {
      "auxiliary_loss_clip": 0.01135848,
      "auxiliary_loss_mlp": 0.01026356,
      "balance_loss_clip": 1.04243279,
      "balance_loss_mlp": 1.01921821,
      "epoch": 0.7794144171225876,
      "flos": 23258874908160.0,
      "grad_norm": 1.8932223203062266,
      "language_loss": 0.84375733,
      "learning_rate": 4.889182972587877e-07,
      "loss": 0.86537939,
      "num_input_tokens_seen": 139619030,
      "step": 6482,
      "time_per_iteration": 2.523501396179199
    },
    {
      "auxiliary_loss_clip": 0.01130948,
      "auxiliary_loss_mlp": 0.01027832,
      "balance_loss_clip": 1.04473662,
      "balance_loss_mlp": 1.02107906,
      "epoch": 0.7795346600132267,
      "flos": 21507080613120.0,
      "grad_norm": 9.710303848078075,
      "language_loss": 0.66234672,
      "learning_rate": 4.884081063643177e-07,
      "loss": 0.68393451,
      "num_input_tokens_seen": 139637690,
      "step": 6483,
      "time_per_iteration": 3.2897355556488037
    },
    {
      "auxiliary_loss_clip": 0.0103226,
      "auxiliary_loss_mlp": 0.00999927,
      "balance_loss_clip": 1.01141191,
      "balance_loss_mlp": 0.99902087,
      "epoch": 0.7796549029038659,
      "flos": 70052273694720.0,
      "grad_norm": 0.8621934649572839,
      "language_loss": 0.52566767,
      "learning_rate": 4.878981447771353e-07,
      "loss": 0.54598957,
      "num_input_tokens_seen": 139692070,
      "step": 6484,
      "time_per_iteration": 3.115163564682007
    },
    {
      "auxiliary_loss_clip": 0.01120947,
      "auxiliary_loss_mlp": 0.01029551,
      "balance_loss_clip": 1.04456258,
      "balance_loss_mlp": 1.02232361,
      "epoch": 0.7797751457945049,
      "flos": 23989405714560.0,
      "grad_norm": 1.6316992261830194,
      "language_loss": 0.72827148,
      "learning_rate": 4.873884125746035e-07,
      "loss": 0.74977648,
      "num_input_tokens_seen": 139713745,
      "step": 6485,
      "time_per_iteration": 2.581382989883423
    },
    {
      "auxiliary_loss_clip": 0.0113265,
      "auxiliary_loss_mlp": 0.01023939,
      "balance_loss_clip": 1.04523301,
      "balance_loss_mlp": 1.01695383,
      "epoch": 0.779895388685144,
      "flos": 22674751937280.0,
      "grad_norm": 2.126933165693386,
      "language_loss": 0.72381508,
      "learning_rate": 4.868789098340456e-07,
      "loss": 0.745381,
      "num_input_tokens_seen": 139731650,
      "step": 6486,
      "time_per_iteration": 2.523045778274536
    },
    {
      "auxiliary_loss_clip": 0.01124488,
      "auxiliary_loss_mlp": 0.01025102,
      "balance_loss_clip": 1.04337788,
      "balance_loss_mlp": 1.01789618,
      "epoch": 0.7800156315757831,
      "flos": 23768698596480.0,
      "grad_norm": 2.3194994017288977,
      "language_loss": 0.73251039,
      "learning_rate": 4.863696366327543e-07,
      "loss": 0.75400627,
      "num_input_tokens_seen": 139750820,
      "step": 6487,
      "time_per_iteration": 2.563617706298828
    },
    {
      "auxiliary_loss_clip": 0.01150123,
      "auxiliary_loss_mlp": 0.01027835,
      "balance_loss_clip": 1.04295731,
      "balance_loss_mlp": 1.0208106,
      "epoch": 0.7801358744664222,
      "flos": 26429714881920.0,
      "grad_norm": 1.7333609652875686,
      "language_loss": 0.77841908,
      "learning_rate": 4.85860593047986e-07,
      "loss": 0.80019861,
      "num_input_tokens_seen": 139770885,
      "step": 6488,
      "time_per_iteration": 2.5248007774353027
    },
    {
      "auxiliary_loss_clip": 0.01114353,
      "auxiliary_loss_mlp": 0.01027662,
      "balance_loss_clip": 1.03809643,
      "balance_loss_mlp": 1.02055681,
      "epoch": 0.7802561173570612,
      "flos": 26322162583680.0,
      "grad_norm": 1.4696677239899198,
      "language_loss": 0.74806821,
      "learning_rate": 4.853517791569613e-07,
      "loss": 0.76948833,
      "num_input_tokens_seen": 139793065,
      "step": 6489,
      "time_per_iteration": 2.5916953086853027
    },
    {
      "auxiliary_loss_clip": 0.01140195,
      "auxiliary_loss_mlp": 0.0076095,
      "balance_loss_clip": 1.04340172,
      "balance_loss_mlp": 1.00026381,
      "epoch": 0.7803763602477004,
      "flos": 40333751596800.0,
      "grad_norm": 1.92825875798555,
      "language_loss": 0.66068298,
      "learning_rate": 4.848431950368684e-07,
      "loss": 0.67969441,
      "num_input_tokens_seen": 139815625,
      "step": 6490,
      "time_per_iteration": 2.6829874515533447
    },
    {
      "auxiliary_loss_clip": 0.01056035,
      "auxiliary_loss_mlp": 0.00751187,
      "balance_loss_clip": 1.00844383,
      "balance_loss_mlp": 1.00003803,
      "epoch": 0.7804966031383395,
      "flos": 67001448038400.0,
      "grad_norm": 0.7139268136616196,
      "language_loss": 0.55773413,
      "learning_rate": 4.843348407648569e-07,
      "loss": 0.57580638,
      "num_input_tokens_seen": 139876905,
      "step": 6491,
      "time_per_iteration": 3.0175840854644775
    },
    {
      "auxiliary_loss_clip": 0.01149116,
      "auxiliary_loss_mlp": 0.01025248,
      "balance_loss_clip": 1.04084373,
      "balance_loss_mlp": 1.01769602,
      "epoch": 0.7806168460289785,
      "flos": 17740733057280.0,
      "grad_norm": 2.4361989538748676,
      "language_loss": 0.82596505,
      "learning_rate": 4.838267164180457e-07,
      "loss": 0.8477087,
      "num_input_tokens_seen": 139892575,
      "step": 6492,
      "time_per_iteration": 2.4603161811828613
    },
    {
      "auxiliary_loss_clip": 0.0116711,
      "auxiliary_loss_mlp": 0.01027178,
      "balance_loss_clip": 1.04793692,
      "balance_loss_mlp": 1.02002573,
      "epoch": 0.7807370889196176,
      "flos": 23946240545280.0,
      "grad_norm": 1.7451426401907058,
      "language_loss": 0.83712161,
      "learning_rate": 4.833188220735156e-07,
      "loss": 0.85906446,
      "num_input_tokens_seen": 139912245,
      "step": 6493,
      "time_per_iteration": 2.46107816696167
    },
    {
      "auxiliary_loss_clip": 0.01150253,
      "auxiliary_loss_mlp": 0.01021587,
      "balance_loss_clip": 1.0466907,
      "balance_loss_mlp": 1.01471806,
      "epoch": 0.7808573318102567,
      "flos": 18989024457600.0,
      "grad_norm": 2.0679413148494663,
      "language_loss": 0.74675047,
      "learning_rate": 4.828111578083152e-07,
      "loss": 0.76846886,
      "num_input_tokens_seen": 139929150,
      "step": 6494,
      "time_per_iteration": 2.495389461517334
    },
    {
      "auxiliary_loss_clip": 0.01135213,
      "auxiliary_loss_mlp": 0.01026048,
      "balance_loss_clip": 1.04497671,
      "balance_loss_mlp": 1.01893067,
      "epoch": 0.7809775747008958,
      "flos": 23980750536960.0,
      "grad_norm": 1.832012049869721,
      "language_loss": 0.81131995,
      "learning_rate": 4.823037236994556e-07,
      "loss": 0.83293253,
      "num_input_tokens_seen": 139947315,
      "step": 6495,
      "time_per_iteration": 2.5544991493225098
    },
    {
      "auxiliary_loss_clip": 0.01046751,
      "auxiliary_loss_mlp": 0.01000446,
      "balance_loss_clip": 1.00799012,
      "balance_loss_mlp": 0.99933761,
      "epoch": 0.7810978175915348,
      "flos": 68535875180160.0,
      "grad_norm": 0.7753786716000538,
      "language_loss": 0.56360388,
      "learning_rate": 4.817965198239136e-07,
      "loss": 0.58407581,
      "num_input_tokens_seen": 140013775,
      "step": 6496,
      "time_per_iteration": 3.0868732929229736
    },
    {
      "auxiliary_loss_clip": 0.01119831,
      "auxiliary_loss_mlp": 0.01027211,
      "balance_loss_clip": 1.04198766,
      "balance_loss_mlp": 1.02034771,
      "epoch": 0.781218060482174,
      "flos": 19642131498240.0,
      "grad_norm": 2.074702896653299,
      "language_loss": 0.74391162,
      "learning_rate": 4.812895462586331e-07,
      "loss": 0.76538205,
      "num_input_tokens_seen": 140031600,
      "step": 6497,
      "time_per_iteration": 3.2819712162017822
    },
    {
      "auxiliary_loss_clip": 0.01126745,
      "auxiliary_loss_mlp": 0.01027677,
      "balance_loss_clip": 1.04486251,
      "balance_loss_mlp": 1.0209682,
      "epoch": 0.7813383033728131,
      "flos": 25627865621760.0,
      "grad_norm": 1.608433596862151,
      "language_loss": 0.82066858,
      "learning_rate": 4.807828030805207e-07,
      "loss": 0.8422128,
      "num_input_tokens_seen": 140050590,
      "step": 6498,
      "time_per_iteration": 2.615826368331909
    },
    {
      "auxiliary_loss_clip": 0.01146696,
      "auxiliary_loss_mlp": 0.01028316,
      "balance_loss_clip": 1.04606175,
      "balance_loss_mlp": 1.02128553,
      "epoch": 0.7814585462634521,
      "flos": 20485924865280.0,
      "grad_norm": 1.8173695644406185,
      "language_loss": 0.67633998,
      "learning_rate": 4.802762903664495e-07,
      "loss": 0.69809008,
      "num_input_tokens_seen": 140069770,
      "step": 6499,
      "time_per_iteration": 2.4908089637756348
    },
    {
      "auxiliary_loss_clip": 0.01140653,
      "auxiliary_loss_mlp": 0.01027806,
      "balance_loss_clip": 1.04590511,
      "balance_loss_mlp": 1.02043843,
      "epoch": 0.7815787891540913,
      "flos": 22304297018880.0,
      "grad_norm": 9.070221819689818,
      "language_loss": 0.73735523,
      "learning_rate": 4.797700081932565e-07,
      "loss": 0.75903982,
      "num_input_tokens_seen": 140087635,
      "step": 6500,
      "time_per_iteration": 2.51689076423645
    },
    {
      "auxiliary_loss_clip": 0.0109077,
      "auxiliary_loss_mlp": 0.01030228,
      "balance_loss_clip": 1.03769064,
      "balance_loss_mlp": 1.02315879,
      "epoch": 0.7816990320447303,
      "flos": 22600668136320.0,
      "grad_norm": 2.342391542116195,
      "language_loss": 0.82162005,
      "learning_rate": 4.792639566377442e-07,
      "loss": 0.84283,
      "num_input_tokens_seen": 140105045,
      "step": 6501,
      "time_per_iteration": 2.6163601875305176
    },
    {
      "auxiliary_loss_clip": 0.01143956,
      "auxiliary_loss_mlp": 0.01027231,
      "balance_loss_clip": 1.04297376,
      "balance_loss_mlp": 1.02003407,
      "epoch": 0.7818192749353694,
      "flos": 24935974871040.0,
      "grad_norm": 1.5957907069921615,
      "language_loss": 0.77157086,
      "learning_rate": 4.78758135776681e-07,
      "loss": 0.79328275,
      "num_input_tokens_seen": 140124900,
      "step": 6502,
      "time_per_iteration": 2.5160863399505615
    },
    {
      "auxiliary_loss_clip": 0.01138158,
      "auxiliary_loss_mlp": 0.01029135,
      "balance_loss_clip": 1.04600811,
      "balance_loss_mlp": 1.02146113,
      "epoch": 0.7819395178260086,
      "flos": 23733039369600.0,
      "grad_norm": 1.9421097432815195,
      "language_loss": 0.78911799,
      "learning_rate": 4.782525456867989e-07,
      "loss": 0.8107909,
      "num_input_tokens_seen": 140143755,
      "step": 6503,
      "time_per_iteration": 2.5491530895233154
    },
    {
      "auxiliary_loss_clip": 0.01122356,
      "auxiliary_loss_mlp": 0.01023816,
      "balance_loss_clip": 1.04519081,
      "balance_loss_mlp": 1.01619208,
      "epoch": 0.7820597607166476,
      "flos": 23221671396480.0,
      "grad_norm": 1.5288338747309502,
      "language_loss": 0.83341682,
      "learning_rate": 4.777471864447959e-07,
      "loss": 0.85487854,
      "num_input_tokens_seen": 140164495,
      "step": 6504,
      "time_per_iteration": 2.5561225414276123
    },
    {
      "auxiliary_loss_clip": 0.01137227,
      "auxiliary_loss_mlp": 0.01024811,
      "balance_loss_clip": 1.04304075,
      "balance_loss_mlp": 1.01822782,
      "epoch": 0.7821800036072867,
      "flos": 22309540404480.0,
      "grad_norm": 2.02836879675236,
      "language_loss": 0.80618054,
      "learning_rate": 4.772420581273344e-07,
      "loss": 0.82780087,
      "num_input_tokens_seen": 140181980,
      "step": 6505,
      "time_per_iteration": 2.5175833702087402
    },
    {
      "auxiliary_loss_clip": 0.01146086,
      "auxiliary_loss_mlp": 0.01022365,
      "balance_loss_clip": 1.04630637,
      "balance_loss_mlp": 1.01572752,
      "epoch": 0.7823002464979258,
      "flos": 21544176384000.0,
      "grad_norm": 1.8524589084194576,
      "language_loss": 0.76136214,
      "learning_rate": 4.7673716081104134e-07,
      "loss": 0.78304666,
      "num_input_tokens_seen": 140202155,
      "step": 6506,
      "time_per_iteration": 3.2811100482940674
    },
    {
      "auxiliary_loss_clip": 0.01150063,
      "auxiliary_loss_mlp": 0.01029061,
      "balance_loss_clip": 1.04857183,
      "balance_loss_mlp": 1.02241552,
      "epoch": 0.7824204893885649,
      "flos": 24535642815360.0,
      "grad_norm": 1.68563867933467,
      "language_loss": 0.84693635,
      "learning_rate": 4.762324945725109e-07,
      "loss": 0.86872756,
      "num_input_tokens_seen": 140221600,
      "step": 6507,
      "time_per_iteration": 3.3656256198883057
    },
    {
      "auxiliary_loss_clip": 0.0113383,
      "auxiliary_loss_mlp": 0.01029417,
      "balance_loss_clip": 1.04780316,
      "balance_loss_mlp": 1.02289665,
      "epoch": 0.782540732279204,
      "flos": 27415211402880.0,
      "grad_norm": 1.7168648052556679,
      "language_loss": 0.75630844,
      "learning_rate": 4.7572805948829844e-07,
      "loss": 0.77794099,
      "num_input_tokens_seen": 140241860,
      "step": 6508,
      "time_per_iteration": 2.5579142570495605
    },
    {
      "auxiliary_loss_clip": 0.01115301,
      "auxiliary_loss_mlp": 0.01025269,
      "balance_loss_clip": 1.0436002,
      "balance_loss_mlp": 1.01873589,
      "epoch": 0.7826609751698431,
      "flos": 24353216616960.0,
      "grad_norm": 1.665977734085383,
      "language_loss": 0.71022904,
      "learning_rate": 4.7522385563492795e-07,
      "loss": 0.73163474,
      "num_input_tokens_seen": 140262160,
      "step": 6509,
      "time_per_iteration": 3.344266891479492
    },
    {
      "auxiliary_loss_clip": 0.01125531,
      "auxiliary_loss_mlp": 0.01036689,
      "balance_loss_clip": 1.04485643,
      "balance_loss_mlp": 1.02979887,
      "epoch": 0.7827812180604822,
      "flos": 23988543788160.0,
      "grad_norm": 1.7755482925981065,
      "language_loss": 0.70295185,
      "learning_rate": 4.747198830888863e-07,
      "loss": 0.72457403,
      "num_input_tokens_seen": 140282030,
      "step": 6510,
      "time_per_iteration": 2.5800187587738037
    },
    {
      "auxiliary_loss_clip": 0.0113382,
      "auxiliary_loss_mlp": 0.01026823,
      "balance_loss_clip": 1.04430735,
      "balance_loss_mlp": 1.01976025,
      "epoch": 0.7829014609511212,
      "flos": 27454318335360.0,
      "grad_norm": 1.9406361521144146,
      "language_loss": 0.68176603,
      "learning_rate": 4.742161419266251e-07,
      "loss": 0.70337248,
      "num_input_tokens_seen": 140301190,
      "step": 6511,
      "time_per_iteration": 2.5673305988311768
    },
    {
      "auxiliary_loss_clip": 0.01154505,
      "auxiliary_loss_mlp": 0.01031483,
      "balance_loss_clip": 1.04847836,
      "balance_loss_mlp": 1.02435446,
      "epoch": 0.7830217038417604,
      "flos": 29204532432000.0,
      "grad_norm": 2.643026042730957,
      "language_loss": 0.64946854,
      "learning_rate": 4.7371263222456304e-07,
      "loss": 0.67132843,
      "num_input_tokens_seen": 140318510,
      "step": 6512,
      "time_per_iteration": 2.5491061210632324
    },
    {
      "auxiliary_loss_clip": 0.0104402,
      "auxiliary_loss_mlp": 0.01002374,
      "balance_loss_clip": 1.00843453,
      "balance_loss_mlp": 1.00134313,
      "epoch": 0.7831419467323995,
      "flos": 60950895822720.0,
      "grad_norm": 0.7989762737519451,
      "language_loss": 0.61480826,
      "learning_rate": 4.7320935405908004e-07,
      "loss": 0.63527215,
      "num_input_tokens_seen": 140379380,
      "step": 6513,
      "time_per_iteration": 3.0544650554656982
    },
    {
      "auxiliary_loss_clip": 0.01166419,
      "auxiliary_loss_mlp": 0.01026991,
      "balance_loss_clip": 1.04766107,
      "balance_loss_mlp": 1.01904011,
      "epoch": 0.7832621896230385,
      "flos": 19682531320320.0,
      "grad_norm": 2.256425534944916,
      "language_loss": 0.84065819,
      "learning_rate": 4.7270630750652475e-07,
      "loss": 0.86259234,
      "num_input_tokens_seen": 140395335,
      "step": 6514,
      "time_per_iteration": 2.453723192214966
    },
    {
      "auxiliary_loss_clip": 0.01147882,
      "auxiliary_loss_mlp": 0.01022702,
      "balance_loss_clip": 1.04525208,
      "balance_loss_mlp": 1.01590443,
      "epoch": 0.7833824325136777,
      "flos": 25009232659200.0,
      "grad_norm": 1.648343507656554,
      "language_loss": 0.80012298,
      "learning_rate": 4.7220349264320746e-07,
      "loss": 0.82182884,
      "num_input_tokens_seen": 140414420,
      "step": 6515,
      "time_per_iteration": 2.5175106525421143
    },
    {
      "auxiliary_loss_clip": 0.01045229,
      "auxiliary_loss_mlp": 0.01003005,
      "balance_loss_clip": 1.00824189,
      "balance_loss_mlp": 1.00199175,
      "epoch": 0.7835026754043167,
      "flos": 68800142517120.0,
      "grad_norm": 0.7359818475499356,
      "language_loss": 0.54951131,
      "learning_rate": 4.71700909545407e-07,
      "loss": 0.56999362,
      "num_input_tokens_seen": 140477365,
      "step": 6516,
      "time_per_iteration": 3.1136608123779297
    },
    {
      "auxiliary_loss_clip": 0.01150534,
      "auxiliary_loss_mlp": 0.01025974,
      "balance_loss_clip": 1.04532814,
      "balance_loss_mlp": 1.01911056,
      "epoch": 0.7836229182949558,
      "flos": 19864598382720.0,
      "grad_norm": 3.3863844594218278,
      "language_loss": 0.7700336,
      "learning_rate": 4.711985582893627e-07,
      "loss": 0.79179859,
      "num_input_tokens_seen": 140495885,
      "step": 6517,
      "time_per_iteration": 2.477668285369873
    },
    {
      "auxiliary_loss_clip": 0.0110966,
      "auxiliary_loss_mlp": 0.0102479,
      "balance_loss_clip": 1.03935885,
      "balance_loss_mlp": 1.01767015,
      "epoch": 0.783743161185595,
      "flos": 22965843755520.0,
      "grad_norm": 1.8056698794777117,
      "language_loss": 0.72069395,
      "learning_rate": 4.706964389512811e-07,
      "loss": 0.74203843,
      "num_input_tokens_seen": 140515920,
      "step": 6518,
      "time_per_iteration": 2.6027333736419678
    },
    {
      "auxiliary_loss_clip": 0.01163318,
      "auxiliary_loss_mlp": 0.01030735,
      "balance_loss_clip": 1.04851174,
      "balance_loss_mlp": 1.02392817,
      "epoch": 0.783863404076234,
      "flos": 12458489777280.0,
      "grad_norm": 1.740197051847246,
      "language_loss": 0.87240738,
      "learning_rate": 4.701945516073345e-07,
      "loss": 0.89434791,
      "num_input_tokens_seen": 140533395,
      "step": 6519,
      "time_per_iteration": 2.4272100925445557
    },
    {
      "auxiliary_loss_clip": 0.01122233,
      "auxiliary_loss_mlp": 0.01025345,
      "balance_loss_clip": 1.04602253,
      "balance_loss_mlp": 1.01861262,
      "epoch": 0.7839836469668731,
      "flos": 24243940465920.0,
      "grad_norm": 1.8068563700734452,
      "language_loss": 0.75192785,
      "learning_rate": 4.696928963336577e-07,
      "loss": 0.77340364,
      "num_input_tokens_seen": 140552825,
      "step": 6520,
      "time_per_iteration": 2.595992088317871
    },
    {
      "auxiliary_loss_clip": 0.01043598,
      "auxiliary_loss_mlp": 0.01001169,
      "balance_loss_clip": 1.00823975,
      "balance_loss_mlp": 1.00016117,
      "epoch": 0.7841038898575122,
      "flos": 62121978938880.0,
      "grad_norm": 0.9335578673965258,
      "language_loss": 0.61044312,
      "learning_rate": 4.6919147320635224e-07,
      "loss": 0.63089079,
      "num_input_tokens_seen": 140615535,
      "step": 6521,
      "time_per_iteration": 3.054169178009033
    },
    {
      "auxiliary_loss_clip": 0.0115192,
      "auxiliary_loss_mlp": 0.01028701,
      "balance_loss_clip": 1.04476976,
      "balance_loss_mlp": 1.02207017,
      "epoch": 0.7842241327481513,
      "flos": 20193899293440.0,
      "grad_norm": 2.094490430636886,
      "language_loss": 0.7321499,
      "learning_rate": 4.6869028230148286e-07,
      "loss": 0.75395614,
      "num_input_tokens_seen": 140633330,
      "step": 6522,
      "time_per_iteration": 2.4916365146636963
    },
    {
      "auxiliary_loss_clip": 0.01116727,
      "auxiliary_loss_mlp": 0.01029236,
      "balance_loss_clip": 1.04089975,
      "balance_loss_mlp": 1.02218485,
      "epoch": 0.7843443756387903,
      "flos": 28074531496320.0,
      "grad_norm": 2.8972258883940922,
      "language_loss": 0.60075814,
      "learning_rate": 4.6818932369507957e-07,
      "loss": 0.62221777,
      "num_input_tokens_seen": 140652830,
      "step": 6523,
      "time_per_iteration": 3.525413990020752
    },
    {
      "auxiliary_loss_clip": 0.01149054,
      "auxiliary_loss_mlp": 0.01027395,
      "balance_loss_clip": 1.0466485,
      "balance_loss_mlp": 1.02007818,
      "epoch": 0.7844646185294295,
      "flos": 21323397438720.0,
      "grad_norm": 1.9103103071980514,
      "language_loss": 0.88785923,
      "learning_rate": 4.676885974631386e-07,
      "loss": 0.90962374,
      "num_input_tokens_seen": 140671190,
      "step": 6524,
      "time_per_iteration": 2.4865715503692627
    },
    {
      "auxiliary_loss_clip": 0.01152563,
      "auxiliary_loss_mlp": 0.01022132,
      "balance_loss_clip": 1.04787815,
      "balance_loss_mlp": 1.01439524,
      "epoch": 0.7845848614200686,
      "flos": 23656585271040.0,
      "grad_norm": 1.978844594948399,
      "language_loss": 0.81014836,
      "learning_rate": 4.67188103681619e-07,
      "loss": 0.83189529,
      "num_input_tokens_seen": 140690975,
      "step": 6525,
      "time_per_iteration": 2.5077309608459473
    },
    {
      "auxiliary_loss_clip": 0.01144333,
      "auxiliary_loss_mlp": 0.00760462,
      "balance_loss_clip": 1.04652739,
      "balance_loss_mlp": 1.00025547,
      "epoch": 0.7847051043107076,
      "flos": 23402194174080.0,
      "grad_norm": 2.185276750032085,
      "language_loss": 0.68891984,
      "learning_rate": 4.666878424264453e-07,
      "loss": 0.70796782,
      "num_input_tokens_seen": 140710930,
      "step": 6526,
      "time_per_iteration": 2.4974687099456787
    },
    {
      "auxiliary_loss_clip": 0.01125917,
      "auxiliary_loss_mlp": 0.01023975,
      "balance_loss_clip": 1.04238415,
      "balance_loss_mlp": 1.01735306,
      "epoch": 0.7848253472013467,
      "flos": 19022277473280.0,
      "grad_norm": 1.5982554999067324,
      "language_loss": 0.73814362,
      "learning_rate": 4.661878137735069e-07,
      "loss": 0.75964254,
      "num_input_tokens_seen": 140729120,
      "step": 6527,
      "time_per_iteration": 2.513636589050293
    },
    {
      "auxiliary_loss_clip": 0.01137448,
      "auxiliary_loss_mlp": 0.01026137,
      "balance_loss_clip": 1.04696262,
      "balance_loss_mlp": 1.01919007,
      "epoch": 0.7849455900919858,
      "flos": 21179180332800.0,
      "grad_norm": 2.0820400886071733,
      "language_loss": 0.75428712,
      "learning_rate": 4.656880177986571e-07,
      "loss": 0.77592289,
      "num_input_tokens_seen": 140747665,
      "step": 6528,
      "time_per_iteration": 2.523752212524414
    },
    {
      "auxiliary_loss_clip": 0.01139679,
      "auxiliary_loss_mlp": 0.01030509,
      "balance_loss_clip": 1.04399037,
      "balance_loss_mlp": 1.02271914,
      "epoch": 0.7850658329826249,
      "flos": 19536482620800.0,
      "grad_norm": 1.9234823988459748,
      "language_loss": 0.81236011,
      "learning_rate": 4.6518845457771607e-07,
      "loss": 0.83406198,
      "num_input_tokens_seen": 140766525,
      "step": 6529,
      "time_per_iteration": 2.5233147144317627
    },
    {
      "auxiliary_loss_clip": 0.0114149,
      "auxiliary_loss_mlp": 0.00760804,
      "balance_loss_clip": 1.04370761,
      "balance_loss_mlp": 1.00022781,
      "epoch": 0.7851860758732639,
      "flos": 12495334152960.0,
      "grad_norm": 1.7286957845191866,
      "language_loss": 0.79030704,
      "learning_rate": 4.646891241864652e-07,
      "loss": 0.80932999,
      "num_input_tokens_seen": 140785090,
      "step": 6530,
      "time_per_iteration": 2.4663784503936768
    },
    {
      "auxiliary_loss_clip": 0.01147091,
      "auxiliary_loss_mlp": 0.010294,
      "balance_loss_clip": 1.04395819,
      "balance_loss_mlp": 1.02183616,
      "epoch": 0.7853063187639031,
      "flos": 22960959505920.0,
      "grad_norm": 1.8159386985430017,
      "language_loss": 0.73232996,
      "learning_rate": 4.6419002670065397e-07,
      "loss": 0.75409484,
      "num_input_tokens_seen": 140804670,
      "step": 6531,
      "time_per_iteration": 2.4960975646972656
    },
    {
      "auxiliary_loss_clip": 0.01129495,
      "auxiliary_loss_mlp": 0.010277,
      "balance_loss_clip": 1.04682517,
      "balance_loss_mlp": 1.01969266,
      "epoch": 0.7854265616545422,
      "flos": 17347260499200.0,
      "grad_norm": 1.9496859553763313,
      "language_loss": 0.86550546,
      "learning_rate": 4.6369116219599445e-07,
      "loss": 0.88707733,
      "num_input_tokens_seen": 140820655,
      "step": 6532,
      "time_per_iteration": 3.318070888519287
    },
    {
      "auxiliary_loss_clip": 0.01121618,
      "auxiliary_loss_mlp": 0.01024942,
      "balance_loss_clip": 1.04257298,
      "balance_loss_mlp": 1.01805151,
      "epoch": 0.7855468045451812,
      "flos": 23838293197440.0,
      "grad_norm": 1.573388239489022,
      "language_loss": 0.79023468,
      "learning_rate": 4.631925307481637e-07,
      "loss": 0.81170028,
      "num_input_tokens_seen": 140840470,
      "step": 6533,
      "time_per_iteration": 2.587639808654785
    },
    {
      "auxiliary_loss_clip": 0.01137508,
      "auxiliary_loss_mlp": 0.01023173,
      "balance_loss_clip": 1.04732299,
      "balance_loss_mlp": 1.01652133,
      "epoch": 0.7856670474358204,
      "flos": 25666792986240.0,
      "grad_norm": 2.6418177358372152,
      "language_loss": 0.75856054,
      "learning_rate": 4.6269413243280533e-07,
      "loss": 0.78016734,
      "num_input_tokens_seen": 140859890,
      "step": 6534,
      "time_per_iteration": 3.357423782348633
    },
    {
      "auxiliary_loss_clip": 0.01140069,
      "auxiliary_loss_mlp": 0.01029258,
      "balance_loss_clip": 1.04670608,
      "balance_loss_mlp": 1.02194428,
      "epoch": 0.7857872903264594,
      "flos": 18144656472960.0,
      "grad_norm": 3.0411660101213913,
      "language_loss": 0.74391806,
      "learning_rate": 4.621959673255236e-07,
      "loss": 0.76561129,
      "num_input_tokens_seen": 140876190,
      "step": 6535,
      "time_per_iteration": 3.234729290008545
    },
    {
      "auxiliary_loss_clip": 0.01105967,
      "auxiliary_loss_mlp": 0.01028056,
      "balance_loss_clip": 1.04244089,
      "balance_loss_mlp": 1.02121031,
      "epoch": 0.7859075332170985,
      "flos": 14386138081920.0,
      "grad_norm": 3.5425540028963582,
      "language_loss": 0.90295386,
      "learning_rate": 4.6169803550189135e-07,
      "loss": 0.92429411,
      "num_input_tokens_seen": 140891885,
      "step": 6536,
      "time_per_iteration": 2.5640954971313477
    },
    {
      "auxiliary_loss_clip": 0.0110041,
      "auxiliary_loss_mlp": 0.01030551,
      "balance_loss_clip": 1.04259932,
      "balance_loss_mlp": 1.02293992,
      "epoch": 0.7860277761077377,
      "flos": 19864059678720.0,
      "grad_norm": 2.0251254314030955,
      "language_loss": 0.77240431,
      "learning_rate": 4.6120033703744355e-07,
      "loss": 0.79371393,
      "num_input_tokens_seen": 140910780,
      "step": 6537,
      "time_per_iteration": 2.567315101623535
    },
    {
      "auxiliary_loss_clip": 0.01126092,
      "auxiliary_loss_mlp": 0.01029239,
      "balance_loss_clip": 1.04243743,
      "balance_loss_mlp": 1.02270365,
      "epoch": 0.7861480189983767,
      "flos": 26396174557440.0,
      "grad_norm": 1.9274258473459358,
      "language_loss": 0.78228951,
      "learning_rate": 4.607028720076822e-07,
      "loss": 0.80384278,
      "num_input_tokens_seen": 140927460,
      "step": 6538,
      "time_per_iteration": 2.5812530517578125
    },
    {
      "auxiliary_loss_clip": 0.01149466,
      "auxiliary_loss_mlp": 0.01021332,
      "balance_loss_clip": 1.0464673,
      "balance_loss_mlp": 1.01427197,
      "epoch": 0.7862682618890158,
      "flos": 24236578177920.0,
      "grad_norm": 2.0833905949592673,
      "language_loss": 0.73407787,
      "learning_rate": 4.6020564048807074e-07,
      "loss": 0.75578582,
      "num_input_tokens_seen": 140945135,
      "step": 6539,
      "time_per_iteration": 2.532745599746704
    },
    {
      "auxiliary_loss_clip": 0.0115324,
      "auxiliary_loss_mlp": 0.01027105,
      "balance_loss_clip": 1.04783583,
      "balance_loss_mlp": 1.01963389,
      "epoch": 0.7863885047796549,
      "flos": 47551508259840.0,
      "grad_norm": 1.8784879344032177,
      "language_loss": 0.72231752,
      "learning_rate": 4.5970864255403883e-07,
      "loss": 0.74412096,
      "num_input_tokens_seen": 140966660,
      "step": 6540,
      "time_per_iteration": 2.747436046600342
    },
    {
      "auxiliary_loss_clip": 0.01140596,
      "auxiliary_loss_mlp": 0.01028799,
      "balance_loss_clip": 1.04464507,
      "balance_loss_mlp": 1.02213526,
      "epoch": 0.786508747670294,
      "flos": 24389234979840.0,
      "grad_norm": 1.758276010690744,
      "language_loss": 0.81901526,
      "learning_rate": 4.59211878280982e-07,
      "loss": 0.84070915,
      "num_input_tokens_seen": 140986175,
      "step": 6541,
      "time_per_iteration": 2.5154309272766113
    },
    {
      "auxiliary_loss_clip": 0.01138051,
      "auxiliary_loss_mlp": 0.01029672,
      "balance_loss_clip": 1.04500759,
      "balance_loss_mlp": 1.02257061,
      "epoch": 0.786628990560933,
      "flos": 18041234238720.0,
      "grad_norm": 2.698983340937698,
      "language_loss": 0.70154226,
      "learning_rate": 4.587153477442578e-07,
      "loss": 0.72321951,
      "num_input_tokens_seen": 141002490,
      "step": 6542,
      "time_per_iteration": 2.548576831817627
    },
    {
      "auxiliary_loss_clip": 0.01169555,
      "auxiliary_loss_mlp": 0.01028119,
      "balance_loss_clip": 1.05047679,
      "balance_loss_mlp": 1.02082324,
      "epoch": 0.7867492334515722,
      "flos": 25848860048640.0,
      "grad_norm": 2.7516088189161865,
      "language_loss": 0.81445944,
      "learning_rate": 4.582190510191899e-07,
      "loss": 0.83643615,
      "num_input_tokens_seen": 141021150,
      "step": 6543,
      "time_per_iteration": 2.5127110481262207
    },
    {
      "auxiliary_loss_clip": 0.01121064,
      "auxiliary_loss_mlp": 0.01028139,
      "balance_loss_clip": 1.044873,
      "balance_loss_mlp": 1.0215168,
      "epoch": 0.7868694763422113,
      "flos": 16580819070720.0,
      "grad_norm": 1.903996707130796,
      "language_loss": 0.86997771,
      "learning_rate": 4.5772298818106625e-07,
      "loss": 0.89146972,
      "num_input_tokens_seen": 141036940,
      "step": 6544,
      "time_per_iteration": 2.5808732509613037
    },
    {
      "auxiliary_loss_clip": 0.01125687,
      "auxiliary_loss_mlp": 0.01028039,
      "balance_loss_clip": 1.04611373,
      "balance_loss_mlp": 1.02117276,
      "epoch": 0.7869897192328503,
      "flos": 29386276272000.0,
      "grad_norm": 2.2335611829875086,
      "language_loss": 0.71969426,
      "learning_rate": 4.572271593051384e-07,
      "loss": 0.7412315,
      "num_input_tokens_seen": 141054295,
      "step": 6545,
      "time_per_iteration": 2.619264841079712
    },
    {
      "auxiliary_loss_clip": 0.01102593,
      "auxiliary_loss_mlp": 0.01025183,
      "balance_loss_clip": 1.04297161,
      "balance_loss_mlp": 1.01833165,
      "epoch": 0.7871099621234895,
      "flos": 17128923678720.0,
      "grad_norm": 1.677518204750608,
      "language_loss": 0.77856576,
      "learning_rate": 4.567315644666245e-07,
      "loss": 0.79984355,
      "num_input_tokens_seen": 141073090,
      "step": 6546,
      "time_per_iteration": 2.5928032398223877
    },
    {
      "auxiliary_loss_clip": 0.01118346,
      "auxiliary_loss_mlp": 0.01025194,
      "balance_loss_clip": 1.04551029,
      "balance_loss_mlp": 1.0188911,
      "epoch": 0.7872302050141285,
      "flos": 23440187784960.0,
      "grad_norm": 2.1818669333368708,
      "language_loss": 0.84390366,
      "learning_rate": 4.5623620374070507e-07,
      "loss": 0.86533904,
      "num_input_tokens_seen": 141092405,
      "step": 6547,
      "time_per_iteration": 2.6027333736419678
    },
    {
      "auxiliary_loss_clip": 0.01025392,
      "auxiliary_loss_mlp": 0.01001416,
      "balance_loss_clip": 1.00803733,
      "balance_loss_mlp": 1.00024784,
      "epoch": 0.7873504479047676,
      "flos": 65959752689280.0,
      "grad_norm": 0.760864912295606,
      "language_loss": 0.58434469,
      "learning_rate": 4.557410772025263e-07,
      "loss": 0.60461277,
      "num_input_tokens_seen": 141154355,
      "step": 6548,
      "time_per_iteration": 4.089269638061523
    },
    {
      "auxiliary_loss_clip": 0.01133769,
      "auxiliary_loss_mlp": 0.0102461,
      "balance_loss_clip": 1.0441153,
      "balance_loss_mlp": 1.01791906,
      "epoch": 0.7874706907954068,
      "flos": 23258336204160.0,
      "grad_norm": 1.7212458642125015,
      "language_loss": 0.66086709,
      "learning_rate": 4.5524618492719803e-07,
      "loss": 0.68245089,
      "num_input_tokens_seen": 141173575,
      "step": 6549,
      "time_per_iteration": 2.5578396320343018
    },
    {
      "auxiliary_loss_clip": 0.01151476,
      "auxiliary_loss_mlp": 0.01020555,
      "balance_loss_clip": 1.04602611,
      "balance_loss_mlp": 1.0138526,
      "epoch": 0.7875909336860458,
      "flos": 28767786963840.0,
      "grad_norm": 1.6066700830234217,
      "language_loss": 0.78869569,
      "learning_rate": 4.54751526989795e-07,
      "loss": 0.81041598,
      "num_input_tokens_seen": 141195415,
      "step": 6550,
      "time_per_iteration": 2.578993082046509
    },
    {
      "auxiliary_loss_clip": 0.01153138,
      "auxiliary_loss_mlp": 0.01025565,
      "balance_loss_clip": 1.04659474,
      "balance_loss_mlp": 1.01871371,
      "epoch": 0.7877111765766849,
      "flos": 18697286194560.0,
      "grad_norm": 1.9014530708026893,
      "language_loss": 0.79132909,
      "learning_rate": 4.5425710346535775e-07,
      "loss": 0.81311613,
      "num_input_tokens_seen": 141213360,
      "step": 6551,
      "time_per_iteration": 2.4730305671691895
    },
    {
      "auxiliary_loss_clip": 0.01153076,
      "auxiliary_loss_mlp": 0.01024037,
      "balance_loss_clip": 1.04679251,
      "balance_loss_mlp": 1.0171293,
      "epoch": 0.787831419467324,
      "flos": 27592968833280.0,
      "grad_norm": 2.118484432448108,
      "language_loss": 0.81577694,
      "learning_rate": 4.537629144288877e-07,
      "loss": 0.83754808,
      "num_input_tokens_seen": 141230815,
      "step": 6552,
      "time_per_iteration": 2.5583908557891846
    },
    {
      "auxiliary_loss_clip": 0.01110483,
      "auxiliary_loss_mlp": 0.01028897,
      "balance_loss_clip": 1.03993082,
      "balance_loss_mlp": 1.02196765,
      "epoch": 0.7879516623579631,
      "flos": 18150187167360.0,
      "grad_norm": 1.8989647464209403,
      "language_loss": 0.74988788,
      "learning_rate": 4.5326895995535477e-07,
      "loss": 0.77128166,
      "num_input_tokens_seen": 141249715,
      "step": 6553,
      "time_per_iteration": 2.61181640625
    },
    {
      "auxiliary_loss_clip": 0.01148921,
      "auxiliary_loss_mlp": 0.01026616,
      "balance_loss_clip": 1.0456171,
      "balance_loss_mlp": 1.01975572,
      "epoch": 0.7880719052486022,
      "flos": 20339193807360.0,
      "grad_norm": 2.323660118224841,
      "language_loss": 0.84239048,
      "learning_rate": 4.527752401196907e-07,
      "loss": 0.86414587,
      "num_input_tokens_seen": 141267730,
      "step": 6554,
      "time_per_iteration": 2.5769009590148926
    },
    {
      "auxiliary_loss_clip": 0.01130983,
      "auxiliary_loss_mlp": 0.01025734,
      "balance_loss_clip": 1.04427826,
      "balance_loss_mlp": 1.01846278,
      "epoch": 0.7881921481392413,
      "flos": 21653237053440.0,
      "grad_norm": 2.2979060673229177,
      "language_loss": 0.66559875,
      "learning_rate": 4.5228175499679254e-07,
      "loss": 0.68716598,
      "num_input_tokens_seen": 141287315,
      "step": 6555,
      "time_per_iteration": 2.5301008224487305
    },
    {
      "auxiliary_loss_clip": 0.0104637,
      "auxiliary_loss_mlp": 0.01002452,
      "balance_loss_clip": 1.00863194,
      "balance_loss_mlp": 1.00132525,
      "epoch": 0.7883123910298804,
      "flos": 68565860058240.0,
      "grad_norm": 0.8304992459385439,
      "language_loss": 0.54563928,
      "learning_rate": 4.5178850466152174e-07,
      "loss": 0.56612748,
      "num_input_tokens_seen": 141346145,
      "step": 6556,
      "time_per_iteration": 3.185530662536621
    },
    {
      "auxiliary_loss_clip": 0.01129949,
      "auxiliary_loss_mlp": 0.01024084,
      "balance_loss_clip": 1.04131067,
      "balance_loss_mlp": 1.01744676,
      "epoch": 0.7884326339205194,
      "flos": 19318217627520.0,
      "grad_norm": 1.9204364126700941,
      "language_loss": 0.8186959,
      "learning_rate": 4.512954891887031e-07,
      "loss": 0.84023625,
      "num_input_tokens_seen": 141364445,
      "step": 6557,
      "time_per_iteration": 2.5286951065063477
    },
    {
      "auxiliary_loss_clip": 0.01127546,
      "auxiliary_loss_mlp": 0.01028037,
      "balance_loss_clip": 1.043082,
      "balance_loss_mlp": 1.02078021,
      "epoch": 0.7885528768111585,
      "flos": 17784903807360.0,
      "grad_norm": 1.971820357221249,
      "language_loss": 0.83584082,
      "learning_rate": 4.5080270865312806e-07,
      "loss": 0.85739666,
      "num_input_tokens_seen": 141381640,
      "step": 6558,
      "time_per_iteration": 2.5132339000701904
    },
    {
      "auxiliary_loss_clip": 0.01149324,
      "auxiliary_loss_mlp": 0.01022626,
      "balance_loss_clip": 1.04500246,
      "balance_loss_mlp": 1.01602471,
      "epoch": 0.7886731197017977,
      "flos": 18807639753600.0,
      "grad_norm": 1.915480435031715,
      "language_loss": 0.70946866,
      "learning_rate": 4.5031016312954985e-07,
      "loss": 0.73118818,
      "num_input_tokens_seen": 141399955,
      "step": 6559,
      "time_per_iteration": 3.246853828430176
    },
    {
      "auxiliary_loss_clip": 0.0115966,
      "auxiliary_loss_mlp": 0.01031793,
      "balance_loss_clip": 1.05003786,
      "balance_loss_mlp": 1.02417588,
      "epoch": 0.7887933625924367,
      "flos": 33365358126720.0,
      "grad_norm": 1.8208343131140012,
      "language_loss": 0.74703884,
      "learning_rate": 4.498178526926886e-07,
      "loss": 0.76895332,
      "num_input_tokens_seen": 141420820,
      "step": 6560,
      "time_per_iteration": 4.182219505310059
    },
    {
      "auxiliary_loss_clip": 0.0116473,
      "auxiliary_loss_mlp": 0.01028724,
      "balance_loss_clip": 1.04904723,
      "balance_loss_mlp": 1.02232265,
      "epoch": 0.7889136054830758,
      "flos": 17019360218880.0,
      "grad_norm": 2.013164464309595,
      "language_loss": 0.7244612,
      "learning_rate": 4.4932577741722635e-07,
      "loss": 0.74639571,
      "num_input_tokens_seen": 141439350,
      "step": 6561,
      "time_per_iteration": 2.4558377265930176
    },
    {
      "auxiliary_loss_clip": 0.0113276,
      "auxiliary_loss_mlp": 0.01027071,
      "balance_loss_clip": 1.04368091,
      "balance_loss_mlp": 1.01996577,
      "epoch": 0.7890338483737149,
      "flos": 29424629018880.0,
      "grad_norm": 1.9177824332027407,
      "language_loss": 0.74020386,
      "learning_rate": 4.4883393737780985e-07,
      "loss": 0.76180214,
      "num_input_tokens_seen": 141460300,
      "step": 6562,
      "time_per_iteration": 2.591503381729126
    },
    {
      "auxiliary_loss_clip": 0.01143647,
      "auxiliary_loss_mlp": 0.01024116,
      "balance_loss_clip": 1.04405928,
      "balance_loss_mlp": 1.01748478,
      "epoch": 0.789154091264354,
      "flos": 19971576063360.0,
      "grad_norm": 2.0470126435630758,
      "language_loss": 0.78473544,
      "learning_rate": 4.4834233264905254e-07,
      "loss": 0.80641305,
      "num_input_tokens_seen": 141477315,
      "step": 6563,
      "time_per_iteration": 2.4825618267059326
    },
    {
      "auxiliary_loss_clip": 0.01115775,
      "auxiliary_loss_mlp": 0.01028815,
      "balance_loss_clip": 1.04064918,
      "balance_loss_mlp": 1.02101254,
      "epoch": 0.789274334154993,
      "flos": 14537825216640.0,
      "grad_norm": 2.2413293605828377,
      "language_loss": 0.71740973,
      "learning_rate": 4.478509633055294e-07,
      "loss": 0.7388556,
      "num_input_tokens_seen": 141495025,
      "step": 6564,
      "time_per_iteration": 2.5364372730255127
    },
    {
      "auxiliary_loss_clip": 0.01140433,
      "auxiliary_loss_mlp": 0.01024015,
      "balance_loss_clip": 1.04634404,
      "balance_loss_mlp": 1.01606417,
      "epoch": 0.7893945770456322,
      "flos": 21827403123840.0,
      "grad_norm": 2.377196273341085,
      "language_loss": 0.80166364,
      "learning_rate": 4.473598294217813e-07,
      "loss": 0.82330811,
      "num_input_tokens_seen": 141510450,
      "step": 6565,
      "time_per_iteration": 2.524785041809082
    },
    {
      "auxiliary_loss_clip": 0.01149623,
      "auxiliary_loss_mlp": 0.01027304,
      "balance_loss_clip": 1.04732227,
      "balance_loss_mlp": 1.02039027,
      "epoch": 0.7895148199362713,
      "flos": 20740639184640.0,
      "grad_norm": 2.221040590571018,
      "language_loss": 0.71823525,
      "learning_rate": 4.468689310723124e-07,
      "loss": 0.74000454,
      "num_input_tokens_seen": 141528265,
      "step": 6566,
      "time_per_iteration": 2.5010581016540527
    },
    {
      "auxiliary_loss_clip": 0.01126227,
      "auxiliary_loss_mlp": 0.01030386,
      "balance_loss_clip": 1.04329622,
      "balance_loss_mlp": 1.023314,
      "epoch": 0.7896350628269103,
      "flos": 16690669839360.0,
      "grad_norm": 1.6673561926721985,
      "language_loss": 0.78807998,
      "learning_rate": 4.463782683315913e-07,
      "loss": 0.80964607,
      "num_input_tokens_seen": 141547270,
      "step": 6567,
      "time_per_iteration": 2.554352283477783
    },
    {
      "auxiliary_loss_clip": 0.01162033,
      "auxiliary_loss_mlp": 0.01025166,
      "balance_loss_clip": 1.04737687,
      "balance_loss_mlp": 1.01849651,
      "epoch": 0.7897553057175495,
      "flos": 22638374438400.0,
      "grad_norm": 1.793289201887468,
      "language_loss": 0.73334587,
      "learning_rate": 4.458878412740523e-07,
      "loss": 0.75521785,
      "num_input_tokens_seen": 141566050,
      "step": 6568,
      "time_per_iteration": 2.482306957244873
    },
    {
      "auxiliary_loss_clip": 0.01145805,
      "auxiliary_loss_mlp": 0.01024995,
      "balance_loss_clip": 1.04653108,
      "balance_loss_mlp": 1.01815283,
      "epoch": 0.7898755486081885,
      "flos": 14537573821440.0,
      "grad_norm": 2.4210298186030292,
      "language_loss": 0.77454484,
      "learning_rate": 4.453976499740919e-07,
      "loss": 0.79625279,
      "num_input_tokens_seen": 141583695,
      "step": 6569,
      "time_per_iteration": 2.470735549926758
    },
    {
      "auxiliary_loss_clip": 0.01148835,
      "auxiliary_loss_mlp": 0.01025658,
      "balance_loss_clip": 1.04876828,
      "balance_loss_mlp": 1.01876128,
      "epoch": 0.7899957914988276,
      "flos": 17238487138560.0,
      "grad_norm": 2.005765110692006,
      "language_loss": 0.77781439,
      "learning_rate": 4.4490769450607215e-07,
      "loss": 0.79955935,
      "num_input_tokens_seen": 141601320,
      "step": 6570,
      "time_per_iteration": 2.47442626953125
    },
    {
      "auxiliary_loss_clip": 0.01118526,
      "auxiliary_loss_mlp": 0.0102595,
      "balance_loss_clip": 1.03976846,
      "balance_loss_mlp": 1.01885104,
      "epoch": 0.7901160343894668,
      "flos": 41279351086080.0,
      "grad_norm": 1.824234249984581,
      "language_loss": 0.72494495,
      "learning_rate": 4.4441797494431845e-07,
      "loss": 0.74638963,
      "num_input_tokens_seen": 141623125,
      "step": 6571,
      "time_per_iteration": 2.741027355194092
    },
    {
      "auxiliary_loss_clip": 0.01150599,
      "auxiliary_loss_mlp": 0.01033876,
      "balance_loss_clip": 1.0481739,
      "balance_loss_mlp": 1.02648211,
      "epoch": 0.7902362772801058,
      "flos": 16837005847680.0,
      "grad_norm": 4.696847151844976,
      "language_loss": 0.77913916,
      "learning_rate": 4.439284913631207e-07,
      "loss": 0.80098391,
      "num_input_tokens_seen": 141640335,
      "step": 6572,
      "time_per_iteration": 2.6032888889312744
    },
    {
      "auxiliary_loss_clip": 0.0112317,
      "auxiliary_loss_mlp": 0.01028905,
      "balance_loss_clip": 1.04580104,
      "balance_loss_mlp": 1.02128744,
      "epoch": 0.7903565201707449,
      "flos": 27125987091840.0,
      "grad_norm": 1.8216273393638742,
      "language_loss": 0.8386941,
      "learning_rate": 4.434392438367347e-07,
      "loss": 0.86021483,
      "num_input_tokens_seen": 141659760,
      "step": 6573,
      "time_per_iteration": 2.601834297180176
    },
    {
      "auxiliary_loss_clip": 0.01150152,
      "auxiliary_loss_mlp": 0.01027108,
      "balance_loss_clip": 1.04529345,
      "balance_loss_mlp": 1.02045596,
      "epoch": 0.790476763061384,
      "flos": 31025167142400.0,
      "grad_norm": 1.8248478663365675,
      "language_loss": 0.73779821,
      "learning_rate": 4.4295023243937677e-07,
      "loss": 0.75957084,
      "num_input_tokens_seen": 141679965,
      "step": 6574,
      "time_per_iteration": 2.563201665878296
    },
    {
      "auxiliary_loss_clip": 0.01149883,
      "auxiliary_loss_mlp": 0.01026429,
      "balance_loss_clip": 1.04789782,
      "balance_loss_mlp": 1.01863241,
      "epoch": 0.7905970059520231,
      "flos": 22089084681600.0,
      "grad_norm": 1.6342362322240132,
      "language_loss": 0.80083525,
      "learning_rate": 4.4246145724523123e-07,
      "loss": 0.82259834,
      "num_input_tokens_seen": 141697710,
      "step": 6575,
      "time_per_iteration": 3.3016839027404785
    },
    {
      "auxiliary_loss_clip": 0.01122774,
      "auxiliary_loss_mlp": 0.01024249,
      "balance_loss_clip": 1.04568315,
      "balance_loss_mlp": 1.01694143,
      "epoch": 0.7907172488426621,
      "flos": 20558141159040.0,
      "grad_norm": 2.133116550740585,
      "language_loss": 0.77217096,
      "learning_rate": 4.41972918328444e-07,
      "loss": 0.79364121,
      "num_input_tokens_seen": 141715145,
      "step": 6576,
      "time_per_iteration": 2.5421082973480225
    },
    {
      "auxiliary_loss_clip": 0.01151223,
      "auxiliary_loss_mlp": 0.01034327,
      "balance_loss_clip": 1.04860938,
      "balance_loss_mlp": 1.02734733,
      "epoch": 0.7908374917333013,
      "flos": 30081542901120.0,
      "grad_norm": 2.0517758680319895,
      "language_loss": 0.77630448,
      "learning_rate": 4.4148461576312646e-07,
      "loss": 0.79815996,
      "num_input_tokens_seen": 141734810,
      "step": 6577,
      "time_per_iteration": 2.5662777423858643
    },
    {
      "auxiliary_loss_clip": 0.01151506,
      "auxiliary_loss_mlp": 0.01024916,
      "balance_loss_clip": 1.0480001,
      "balance_loss_mlp": 1.01815128,
      "epoch": 0.7909577346239404,
      "flos": 20996359084800.0,
      "grad_norm": 1.522207665126103,
      "language_loss": 0.74633276,
      "learning_rate": 4.4099654962335343e-07,
      "loss": 0.76809698,
      "num_input_tokens_seen": 141755260,
      "step": 6578,
      "time_per_iteration": 2.52166748046875
    },
    {
      "auxiliary_loss_clip": 0.01145975,
      "auxiliary_loss_mlp": 0.01028701,
      "balance_loss_clip": 1.0477432,
      "balance_loss_mlp": 1.02118754,
      "epoch": 0.7910779775145794,
      "flos": 26247935128320.0,
      "grad_norm": 1.7162982234902366,
      "language_loss": 0.75044149,
      "learning_rate": 4.405087199831636e-07,
      "loss": 0.77218825,
      "num_input_tokens_seen": 141775500,
      "step": 6579,
      "time_per_iteration": 2.57381534576416
    },
    {
      "auxiliary_loss_clip": 0.011375,
      "auxiliary_loss_mlp": 0.00760505,
      "balance_loss_clip": 1.04330015,
      "balance_loss_mlp": 1.00023937,
      "epoch": 0.7911982204052186,
      "flos": 22564434291840.0,
      "grad_norm": 1.9595734191099345,
      "language_loss": 0.6692102,
      "learning_rate": 4.400211269165619e-07,
      "loss": 0.68819028,
      "num_input_tokens_seen": 141791955,
      "step": 6580,
      "time_per_iteration": 2.5479655265808105
    },
    {
      "auxiliary_loss_clip": 0.01167803,
      "auxiliary_loss_mlp": 0.0102889,
      "balance_loss_clip": 1.05225348,
      "balance_loss_mlp": 1.02178812,
      "epoch": 0.7913184632958576,
      "flos": 23112538899840.0,
      "grad_norm": 1.743187620008219,
      "language_loss": 0.76746804,
      "learning_rate": 4.3953377049751416e-07,
      "loss": 0.78943497,
      "num_input_tokens_seen": 141812380,
      "step": 6581,
      "time_per_iteration": 2.4746673107147217
    },
    {
      "auxiliary_loss_clip": 0.01141219,
      "auxiliary_loss_mlp": 0.01025047,
      "balance_loss_clip": 1.04764462,
      "balance_loss_mlp": 1.0185709,
      "epoch": 0.7914387061864967,
      "flos": 12311758719360.0,
      "grad_norm": 2.147341565890573,
      "language_loss": 0.77872324,
      "learning_rate": 4.390466507999537e-07,
      "loss": 0.80038595,
      "num_input_tokens_seen": 141828130,
      "step": 6582,
      "time_per_iteration": 2.495028018951416
    },
    {
      "auxiliary_loss_clip": 0.01117184,
      "auxiliary_loss_mlp": 0.01026257,
      "balance_loss_clip": 1.04293323,
      "balance_loss_mlp": 1.01939678,
      "epoch": 0.7915589490771359,
      "flos": 17603267708160.0,
      "grad_norm": 1.9929535220813694,
      "language_loss": 0.75689065,
      "learning_rate": 4.385597678977748e-07,
      "loss": 0.77832508,
      "num_input_tokens_seen": 141846965,
      "step": 6583,
      "time_per_iteration": 2.5318000316619873
    },
    {
      "auxiliary_loss_clip": 0.01131249,
      "auxiliary_loss_mlp": 0.01028411,
      "balance_loss_clip": 1.04184687,
      "balance_loss_mlp": 1.02078772,
      "epoch": 0.7916791919677749,
      "flos": 25591272641280.0,
      "grad_norm": 1.5531971008921983,
      "language_loss": 0.7540316,
      "learning_rate": 4.3807312186483726e-07,
      "loss": 0.77562821,
      "num_input_tokens_seen": 141867685,
      "step": 6584,
      "time_per_iteration": 3.340477705001831
    },
    {
      "auxiliary_loss_clip": 0.01149258,
      "auxiliary_loss_mlp": 0.01026979,
      "balance_loss_clip": 1.0487802,
      "balance_loss_mlp": 1.01952195,
      "epoch": 0.791799434858414,
      "flos": 18844340474880.0,
      "grad_norm": 1.9644286835395501,
      "language_loss": 0.78514743,
      "learning_rate": 4.375867127749655e-07,
      "loss": 0.8069098,
      "num_input_tokens_seen": 141885960,
      "step": 6585,
      "time_per_iteration": 2.4824368953704834
    },
    {
      "auxiliary_loss_clip": 0.01124678,
      "auxiliary_loss_mlp": 0.01025529,
      "balance_loss_clip": 1.04537022,
      "balance_loss_mlp": 1.01841843,
      "epoch": 0.7919196777490531,
      "flos": 25812015672960.0,
      "grad_norm": 2.371266189956271,
      "language_loss": 0.67202908,
      "learning_rate": 4.3710054070194744e-07,
      "loss": 0.69353116,
      "num_input_tokens_seen": 141905655,
      "step": 6586,
      "time_per_iteration": 3.39778733253479
    },
    {
      "auxiliary_loss_clip": 0.01165373,
      "auxiliary_loss_mlp": 0.00761136,
      "balance_loss_clip": 1.04705024,
      "balance_loss_mlp": 1.00025678,
      "epoch": 0.7920399206396922,
      "flos": 11947624594560.0,
      "grad_norm": 2.7176262955585675,
      "language_loss": 0.66416442,
      "learning_rate": 4.3661460571953455e-07,
      "loss": 0.68342948,
      "num_input_tokens_seen": 141922390,
      "step": 6587,
      "time_per_iteration": 3.166304588317871
    },
    {
      "auxiliary_loss_clip": 0.01150951,
      "auxiliary_loss_mlp": 0.01021042,
      "balance_loss_clip": 1.0450902,
      "balance_loss_mlp": 1.01450372,
      "epoch": 0.7921601635303313,
      "flos": 21579907438080.0,
      "grad_norm": 1.5334389942060427,
      "language_loss": 0.68643111,
      "learning_rate": 4.36128907901443e-07,
      "loss": 0.70815104,
      "num_input_tokens_seen": 141941985,
      "step": 6588,
      "time_per_iteration": 2.5147759914398193
    },
    {
      "auxiliary_loss_clip": 0.01122664,
      "auxiliary_loss_mlp": 0.0102457,
      "balance_loss_clip": 1.0421598,
      "balance_loss_mlp": 1.01728654,
      "epoch": 0.7922804064209703,
      "flos": 18113989236480.0,
      "grad_norm": 1.9616576293152,
      "language_loss": 0.72468239,
      "learning_rate": 4.356434473213519e-07,
      "loss": 0.74615467,
      "num_input_tokens_seen": 141959435,
      "step": 6589,
      "time_per_iteration": 2.5261049270629883
    },
    {
      "auxiliary_loss_clip": 0.01136681,
      "auxiliary_loss_mlp": 0.01027195,
      "balance_loss_clip": 1.04841745,
      "balance_loss_mlp": 1.02073669,
      "epoch": 0.7924006493116095,
      "flos": 21652806090240.0,
      "grad_norm": 1.7140157467386665,
      "language_loss": 0.79905188,
      "learning_rate": 4.351582240529068e-07,
      "loss": 0.82069057,
      "num_input_tokens_seen": 141980265,
      "step": 6590,
      "time_per_iteration": 2.5357437133789062
    },
    {
      "auxiliary_loss_clip": 0.01038962,
      "auxiliary_loss_mlp": 0.01001461,
      "balance_loss_clip": 1.00795221,
      "balance_loss_mlp": 1.00031018,
      "epoch": 0.7925208922022485,
      "flos": 64242755694720.0,
      "grad_norm": 0.7038560101057136,
      "language_loss": 0.58225006,
      "learning_rate": 4.346732381697149e-07,
      "loss": 0.60265428,
      "num_input_tokens_seen": 142044395,
      "step": 6591,
      "time_per_iteration": 3.1610167026519775
    },
    {
      "auxiliary_loss_clip": 0.01134014,
      "auxiliary_loss_mlp": 0.01029904,
      "balance_loss_clip": 1.04685128,
      "balance_loss_mlp": 1.02311814,
      "epoch": 0.7926411350928876,
      "flos": 16941541403520.0,
      "grad_norm": 2.0025836263401926,
      "language_loss": 0.81038892,
      "learning_rate": 4.3418848974534825e-07,
      "loss": 0.83202803,
      "num_input_tokens_seen": 142061335,
      "step": 6592,
      "time_per_iteration": 2.5060744285583496
    },
    {
      "auxiliary_loss_clip": 0.01127394,
      "auxiliary_loss_mlp": 0.01028063,
      "balance_loss_clip": 1.04485846,
      "balance_loss_mlp": 1.02104473,
      "epoch": 0.7927613779835267,
      "flos": 34459987144320.0,
      "grad_norm": 1.583174532761633,
      "language_loss": 0.68685764,
      "learning_rate": 4.3370397885334276e-07,
      "loss": 0.70841217,
      "num_input_tokens_seen": 142081965,
      "step": 6593,
      "time_per_iteration": 2.666896343231201
    },
    {
      "auxiliary_loss_clip": 0.01146083,
      "auxiliary_loss_mlp": 0.01030254,
      "balance_loss_clip": 1.04801726,
      "balance_loss_mlp": 1.02318764,
      "epoch": 0.7928816208741658,
      "flos": 18951174501120.0,
      "grad_norm": 1.7026733696224499,
      "language_loss": 0.75381243,
      "learning_rate": 4.3321970556719777e-07,
      "loss": 0.77557588,
      "num_input_tokens_seen": 142100260,
      "step": 6594,
      "time_per_iteration": 2.486280918121338
    },
    {
      "auxiliary_loss_clip": 0.01162719,
      "auxiliary_loss_mlp": 0.01029566,
      "balance_loss_clip": 1.04746091,
      "balance_loss_mlp": 1.02179384,
      "epoch": 0.7930018637648049,
      "flos": 18623022825600.0,
      "grad_norm": 2.148946758554579,
      "language_loss": 0.71746063,
      "learning_rate": 4.3273566996037856e-07,
      "loss": 0.73938346,
      "num_input_tokens_seen": 142116955,
      "step": 6595,
      "time_per_iteration": 2.4435129165649414
    },
    {
      "auxiliary_loss_clip": 0.01138315,
      "auxiliary_loss_mlp": 0.01025137,
      "balance_loss_clip": 1.04625237,
      "balance_loss_mlp": 1.01851773,
      "epoch": 0.793122106655444,
      "flos": 24530650824960.0,
      "grad_norm": 1.9765610188175706,
      "language_loss": 0.80409986,
      "learning_rate": 4.322518721063113e-07,
      "loss": 0.82573438,
      "num_input_tokens_seen": 142135505,
      "step": 6596,
      "time_per_iteration": 2.5627260208129883
    },
    {
      "auxiliary_loss_clip": 0.01152308,
      "auxiliary_loss_mlp": 0.01027149,
      "balance_loss_clip": 1.04783463,
      "balance_loss_mlp": 1.01985383,
      "epoch": 0.7932423495460831,
      "flos": 34421203434240.0,
      "grad_norm": 2.0239949022226553,
      "language_loss": 0.70101738,
      "learning_rate": 4.3176831207838906e-07,
      "loss": 0.722812,
      "num_input_tokens_seen": 142158915,
      "step": 6597,
      "time_per_iteration": 2.6100752353668213
    },
    {
      "auxiliary_loss_clip": 0.01149724,
      "auxiliary_loss_mlp": 0.01024577,
      "balance_loss_clip": 1.04844832,
      "balance_loss_mlp": 1.01825011,
      "epoch": 0.7933625924367221,
      "flos": 26980333441920.0,
      "grad_norm": 1.7893116262424522,
      "language_loss": 0.74648607,
      "learning_rate": 4.3128498994996685e-07,
      "loss": 0.76822907,
      "num_input_tokens_seen": 142178390,
      "step": 6598,
      "time_per_iteration": 2.5418381690979004
    },
    {
      "auxiliary_loss_clip": 0.01155891,
      "auxiliary_loss_mlp": 0.01029783,
      "balance_loss_clip": 1.04768372,
      "balance_loss_mlp": 1.02251148,
      "epoch": 0.7934828353273613,
      "flos": 29568630643200.0,
      "grad_norm": 3.398297190593282,
      "language_loss": 0.71302295,
      "learning_rate": 4.308019057943646e-07,
      "loss": 0.73487973,
      "num_input_tokens_seen": 142200115,
      "step": 6599,
      "time_per_iteration": 2.5587828159332275
    },
    {
      "auxiliary_loss_clip": 0.01114495,
      "auxiliary_loss_mlp": 0.01022518,
      "balance_loss_clip": 1.04413605,
      "balance_loss_mlp": 1.01605415,
      "epoch": 0.7936030782180004,
      "flos": 28615381557120.0,
      "grad_norm": 1.5925909807756664,
      "language_loss": 0.74619514,
      "learning_rate": 4.3031905968486535e-07,
      "loss": 0.76756531,
      "num_input_tokens_seen": 142220945,
      "step": 6600,
      "time_per_iteration": 3.368894338607788
    },
    {
      "auxiliary_loss_clip": 0.01109897,
      "auxiliary_loss_mlp": 0.01025383,
      "balance_loss_clip": 1.04560375,
      "balance_loss_mlp": 1.01850152,
      "epoch": 0.7937233211086394,
      "flos": 16392574869120.0,
      "grad_norm": 1.9948816406834091,
      "language_loss": 0.68417454,
      "learning_rate": 4.298364516947162e-07,
      "loss": 0.70552725,
      "num_input_tokens_seen": 142238175,
      "step": 6601,
      "time_per_iteration": 2.5723087787628174
    },
    {
      "auxiliary_loss_clip": 0.01105033,
      "auxiliary_loss_mlp": 0.01022177,
      "balance_loss_clip": 1.0414629,
      "balance_loss_mlp": 1.01566505,
      "epoch": 0.7938435639992786,
      "flos": 22013420682240.0,
      "grad_norm": 1.7909913735749496,
      "language_loss": 0.6523906,
      "learning_rate": 4.293540818971295e-07,
      "loss": 0.67366266,
      "num_input_tokens_seen": 142255980,
      "step": 6602,
      "time_per_iteration": 2.587277889251709
    },
    {
      "auxiliary_loss_clip": 0.01153989,
      "auxiliary_loss_mlp": 0.0102398,
      "balance_loss_clip": 1.04600072,
      "balance_loss_mlp": 1.0171349,
      "epoch": 0.7939638068899176,
      "flos": 22197032029440.0,
      "grad_norm": 1.9606597166038036,
      "language_loss": 0.76525217,
      "learning_rate": 4.2887195036527934e-07,
      "loss": 0.78703183,
      "num_input_tokens_seen": 142274785,
      "step": 6603,
      "time_per_iteration": 2.5142898559570312
    },
    {
      "auxiliary_loss_clip": 0.01142244,
      "auxiliary_loss_mlp": 0.01024075,
      "balance_loss_clip": 1.04367864,
      "balance_loss_mlp": 1.01688051,
      "epoch": 0.7940840497805567,
      "flos": 17745186343680.0,
      "grad_norm": 2.6100554389737978,
      "language_loss": 0.73181641,
      "learning_rate": 4.28390057172306e-07,
      "loss": 0.75347966,
      "num_input_tokens_seen": 142291290,
      "step": 6604,
      "time_per_iteration": 2.462362766265869
    },
    {
      "auxiliary_loss_clip": 0.01115957,
      "auxiliary_loss_mlp": 0.0102642,
      "balance_loss_clip": 1.0402174,
      "balance_loss_mlp": 1.01905608,
      "epoch": 0.7942042926711959,
      "flos": 23805435231360.0,
      "grad_norm": 2.2607773672117957,
      "language_loss": 0.72081506,
      "learning_rate": 4.279084023913111e-07,
      "loss": 0.74223888,
      "num_input_tokens_seen": 142309165,
      "step": 6605,
      "time_per_iteration": 2.5864293575286865
    },
    {
      "auxiliary_loss_clip": 0.01149802,
      "auxiliary_loss_mlp": 0.0102848,
      "balance_loss_clip": 1.04692185,
      "balance_loss_mlp": 1.0218637,
      "epoch": 0.7943245355618349,
      "flos": 19244959839360.0,
      "grad_norm": 1.6500159770580556,
      "language_loss": 0.69047308,
      "learning_rate": 4.2742698609536096e-07,
      "loss": 0.7122559,
      "num_input_tokens_seen": 142327475,
      "step": 6606,
      "time_per_iteration": 2.485440969467163
    },
    {
      "auxiliary_loss_clip": 0.0114003,
      "auxiliary_loss_mlp": 0.01025886,
      "balance_loss_clip": 1.04618764,
      "balance_loss_mlp": 1.01915419,
      "epoch": 0.794444778452474,
      "flos": 25007616547200.0,
      "grad_norm": 1.6323515738981953,
      "language_loss": 0.78206521,
      "learning_rate": 4.2694580835748706e-07,
      "loss": 0.80372441,
      "num_input_tokens_seen": 142347335,
      "step": 6607,
      "time_per_iteration": 2.569101572036743
    },
    {
      "auxiliary_loss_clip": 0.01134812,
      "auxiliary_loss_mlp": 0.01026422,
      "balance_loss_clip": 1.04673719,
      "balance_loss_mlp": 1.01935911,
      "epoch": 0.7945650213431131,
      "flos": 23221491828480.0,
      "grad_norm": 1.9419003508898611,
      "language_loss": 0.74163848,
      "learning_rate": 4.264648692506836e-07,
      "loss": 0.76325083,
      "num_input_tokens_seen": 142366125,
      "step": 6608,
      "time_per_iteration": 2.527162551879883
    },
    {
      "auxiliary_loss_clip": 0.01130562,
      "auxiliary_loss_mlp": 0.01040733,
      "balance_loss_clip": 1.04375148,
      "balance_loss_mlp": 1.03288007,
      "epoch": 0.7946852642337522,
      "flos": 26062887237120.0,
      "grad_norm": 2.104910375317638,
      "language_loss": 0.72087961,
      "learning_rate": 4.2598416884790824e-07,
      "loss": 0.74259257,
      "num_input_tokens_seen": 142385175,
      "step": 6609,
      "time_per_iteration": 2.5647928714752197
    },
    {
      "auxiliary_loss_clip": 0.01144377,
      "auxiliary_loss_mlp": 0.01030507,
      "balance_loss_clip": 1.0443393,
      "balance_loss_mlp": 1.02356887,
      "epoch": 0.7948055071243912,
      "flos": 23769704177280.0,
      "grad_norm": 1.910184028955897,
      "language_loss": 0.80661285,
      "learning_rate": 4.255037072220828e-07,
      "loss": 0.82836163,
      "num_input_tokens_seen": 142406545,
      "step": 6610,
      "time_per_iteration": 3.3658664226531982
    },
    {
      "auxiliary_loss_clip": 0.01161056,
      "auxiliary_loss_mlp": 0.01022036,
      "balance_loss_clip": 1.04748893,
      "balance_loss_mlp": 1.01522303,
      "epoch": 0.7949257500150304,
      "flos": 21980814111360.0,
      "grad_norm": 1.5703541711064393,
      "language_loss": 0.71702212,
      "learning_rate": 4.2502348444609293e-07,
      "loss": 0.73885298,
      "num_input_tokens_seen": 142426165,
      "step": 6611,
      "time_per_iteration": 3.303710699081421
    },
    {
      "auxiliary_loss_clip": 0.01104208,
      "auxiliary_loss_mlp": 0.01026188,
      "balance_loss_clip": 1.03860843,
      "balance_loss_mlp": 1.01914275,
      "epoch": 0.7950459929056695,
      "flos": 25774129802880.0,
      "grad_norm": 1.770151523844581,
      "language_loss": 0.69898438,
      "learning_rate": 4.2454350059278844e-07,
      "loss": 0.72028828,
      "num_input_tokens_seen": 142447225,
      "step": 6612,
      "time_per_iteration": 2.643118381500244
    },
    {
      "auxiliary_loss_clip": 0.01129147,
      "auxiliary_loss_mlp": 0.01024179,
      "balance_loss_clip": 1.03966749,
      "balance_loss_mlp": 1.01757777,
      "epoch": 0.7951662357963085,
      "flos": 22158068751360.0,
      "grad_norm": 1.6778179939915567,
      "language_loss": 0.8396762,
      "learning_rate": 4.240637557349824e-07,
      "loss": 0.86120945,
      "num_input_tokens_seen": 142464440,
      "step": 6613,
      "time_per_iteration": 3.274561643600464
    },
    {
      "auxiliary_loss_clip": 0.01122231,
      "auxiliary_loss_mlp": 0.01026468,
      "balance_loss_clip": 1.04258704,
      "balance_loss_mlp": 1.01925027,
      "epoch": 0.7952864786869477,
      "flos": 24641938137600.0,
      "grad_norm": 1.8290358827188562,
      "language_loss": 0.66872811,
      "learning_rate": 4.235842499454516e-07,
      "loss": 0.69021499,
      "num_input_tokens_seen": 142484355,
      "step": 6614,
      "time_per_iteration": 2.582728147506714
    },
    {
      "auxiliary_loss_clip": 0.01136936,
      "auxiliary_loss_mlp": 0.01033628,
      "balance_loss_clip": 1.04491997,
      "balance_loss_mlp": 1.02680922,
      "epoch": 0.7954067215775867,
      "flos": 21830922656640.0,
      "grad_norm": 1.6836686437911408,
      "language_loss": 0.82876384,
      "learning_rate": 4.2310498329693687e-07,
      "loss": 0.85046947,
      "num_input_tokens_seen": 142505255,
      "step": 6615,
      "time_per_iteration": 2.5627050399780273
    },
    {
      "auxiliary_loss_clip": 0.01153452,
      "auxiliary_loss_mlp": 0.01027729,
      "balance_loss_clip": 1.04745746,
      "balance_loss_mlp": 1.01995039,
      "epoch": 0.7955269644682258,
      "flos": 24060652341120.0,
      "grad_norm": 1.5446957459098858,
      "language_loss": 0.80763626,
      "learning_rate": 4.2262595586214164e-07,
      "loss": 0.8294481,
      "num_input_tokens_seen": 142526350,
      "step": 6616,
      "time_per_iteration": 2.5603179931640625
    },
    {
      "auxiliary_loss_clip": 0.01153408,
      "auxiliary_loss_mlp": 0.0102833,
      "balance_loss_clip": 1.04655671,
      "balance_loss_mlp": 1.02113533,
      "epoch": 0.795647207358865,
      "flos": 25010741030400.0,
      "grad_norm": 1.7578453221163097,
      "language_loss": 0.76657647,
      "learning_rate": 4.221471677137358e-07,
      "loss": 0.78839391,
      "num_input_tokens_seen": 142547165,
      "step": 6617,
      "time_per_iteration": 2.5413687229156494
    },
    {
      "auxiliary_loss_clip": 0.01126073,
      "auxiliary_loss_mlp": 0.01021984,
      "balance_loss_clip": 1.0436511,
      "balance_loss_mlp": 1.01548672,
      "epoch": 0.795767450249504,
      "flos": 14648358343680.0,
      "grad_norm": 1.5626755915499213,
      "language_loss": 0.70025235,
      "learning_rate": 4.216686189243492e-07,
      "loss": 0.72173297,
      "num_input_tokens_seen": 142565955,
      "step": 6618,
      "time_per_iteration": 2.5288848876953125
    },
    {
      "auxiliary_loss_clip": 0.01118781,
      "auxiliary_loss_mlp": 0.01031614,
      "balance_loss_clip": 1.04407048,
      "balance_loss_mlp": 1.02368641,
      "epoch": 0.7958876931401431,
      "flos": 18547897530240.0,
      "grad_norm": 1.5816847304286525,
      "language_loss": 0.72719777,
      "learning_rate": 4.211903095665785e-07,
      "loss": 0.74870169,
      "num_input_tokens_seen": 142585340,
      "step": 6619,
      "time_per_iteration": 2.5549676418304443
    },
    {
      "auxiliary_loss_clip": 0.01144839,
      "auxiliary_loss_mlp": 0.01025014,
      "balance_loss_clip": 1.04430747,
      "balance_loss_mlp": 1.01821017,
      "epoch": 0.7960079360307821,
      "flos": 21543960902400.0,
      "grad_norm": 1.908723232002634,
      "language_loss": 0.75564939,
      "learning_rate": 4.2071223971298277e-07,
      "loss": 0.77734786,
      "num_input_tokens_seen": 142602525,
      "step": 6620,
      "time_per_iteration": 2.5222716331481934
    },
    {
      "auxiliary_loss_clip": 0.01151193,
      "auxiliary_loss_mlp": 0.01030363,
      "balance_loss_clip": 1.04591358,
      "balance_loss_mlp": 1.02294517,
      "epoch": 0.7961281789214213,
      "flos": 25481745095040.0,
      "grad_norm": 2.4195932859947993,
      "language_loss": 0.60969168,
      "learning_rate": 4.2023440943608433e-07,
      "loss": 0.63150728,
      "num_input_tokens_seen": 142622490,
      "step": 6621,
      "time_per_iteration": 2.52785325050354
    },
    {
      "auxiliary_loss_clip": 0.01147638,
      "auxiliary_loss_mlp": 0.01027321,
      "balance_loss_clip": 1.04353511,
      "balance_loss_mlp": 1.02084231,
      "epoch": 0.7962484218120603,
      "flos": 21944436612480.0,
      "grad_norm": 1.501799860551746,
      "language_loss": 0.78099811,
      "learning_rate": 4.1975681880837023e-07,
      "loss": 0.80274773,
      "num_input_tokens_seen": 142642495,
      "step": 6622,
      "time_per_iteration": 2.488443374633789
    },
    {
      "auxiliary_loss_clip": 0.01115712,
      "auxiliary_loss_mlp": 0.01025577,
      "balance_loss_clip": 1.03789806,
      "balance_loss_mlp": 1.01858246,
      "epoch": 0.7963686647026994,
      "flos": 18876264687360.0,
      "grad_norm": 1.8143565417476109,
      "language_loss": 0.82225442,
      "learning_rate": 4.192794679022895e-07,
      "loss": 0.84366727,
      "num_input_tokens_seen": 142660820,
      "step": 6623,
      "time_per_iteration": 2.5728352069854736
    },
    {
      "auxiliary_loss_clip": 0.01149274,
      "auxiliary_loss_mlp": 0.01028157,
      "balance_loss_clip": 1.04451632,
      "balance_loss_mlp": 1.02171707,
      "epoch": 0.7964889075933386,
      "flos": 29716582763520.0,
      "grad_norm": 1.91859315275791,
      "language_loss": 0.72212398,
      "learning_rate": 4.1880235679025743e-07,
      "loss": 0.74389827,
      "num_input_tokens_seen": 142680915,
      "step": 6624,
      "time_per_iteration": 2.575850248336792
    },
    {
      "auxiliary_loss_clip": 0.01095833,
      "auxiliary_loss_mlp": 0.01023294,
      "balance_loss_clip": 1.04130685,
      "balance_loss_mlp": 1.01589096,
      "epoch": 0.7966091504839776,
      "flos": 29491458272640.0,
      "grad_norm": 1.75079322329521,
      "language_loss": 0.63889122,
      "learning_rate": 4.1832548554464986e-07,
      "loss": 0.66008246,
      "num_input_tokens_seen": 142699210,
      "step": 6625,
      "time_per_iteration": 2.658163547515869
    },
    {
      "auxiliary_loss_clip": 0.01042895,
      "auxiliary_loss_mlp": 0.01001104,
      "balance_loss_clip": 1.00769544,
      "balance_loss_mlp": 1.000103,
      "epoch": 0.7967293933746167,
      "flos": 67288697101440.0,
      "grad_norm": 0.7427544454454184,
      "language_loss": 0.58796155,
      "learning_rate": 4.178488542378098e-07,
      "loss": 0.60840154,
      "num_input_tokens_seen": 142756790,
      "step": 6626,
      "time_per_iteration": 3.8376355171203613
    },
    {
      "auxiliary_loss_clip": 0.01167082,
      "auxiliary_loss_mlp": 0.01025702,
      "balance_loss_clip": 1.04791689,
      "balance_loss_mlp": 1.01863563,
      "epoch": 0.7968496362652558,
      "flos": 25554679660800.0,
      "grad_norm": 1.7135897055860592,
      "language_loss": 0.88920403,
      "learning_rate": 4.173724629420401e-07,
      "loss": 0.91113192,
      "num_input_tokens_seen": 142778150,
      "step": 6627,
      "time_per_iteration": 2.549018383026123
    },
    {
      "auxiliary_loss_clip": 0.01140039,
      "auxiliary_loss_mlp": 0.01025869,
      "balance_loss_clip": 1.04665935,
      "balance_loss_mlp": 1.01903534,
      "epoch": 0.7969698791558949,
      "flos": 14501088581760.0,
      "grad_norm": 3.2001831721414007,
      "language_loss": 0.68231958,
      "learning_rate": 4.168963117296087e-07,
      "loss": 0.70397866,
      "num_input_tokens_seen": 142795485,
      "step": 6628,
      "time_per_iteration": 2.5316152572631836
    },
    {
      "auxiliary_loss_clip": 0.0116657,
      "auxiliary_loss_mlp": 0.01026233,
      "balance_loss_clip": 1.04945374,
      "balance_loss_mlp": 1.01925623,
      "epoch": 0.797090122046534,
      "flos": 22127545169280.0,
      "grad_norm": 2.913187759763036,
      "language_loss": 0.7602942,
      "learning_rate": 4.1642040067274876e-07,
      "loss": 0.78222221,
      "num_input_tokens_seen": 142815155,
      "step": 6629,
      "time_per_iteration": 2.4640471935272217
    },
    {
      "auxiliary_loss_clip": 0.01139551,
      "auxiliary_loss_mlp": 0.01022963,
      "balance_loss_clip": 1.04398465,
      "balance_loss_mlp": 1.01629925,
      "epoch": 0.7972103649371731,
      "flos": 19897671830400.0,
      "grad_norm": 1.6379379016483138,
      "language_loss": 0.72587901,
      "learning_rate": 4.1594472984365493e-07,
      "loss": 0.74750417,
      "num_input_tokens_seen": 142833840,
      "step": 6630,
      "time_per_iteration": 2.534360647201538
    },
    {
      "auxiliary_loss_clip": 0.01145152,
      "auxiliary_loss_mlp": 0.01025236,
      "balance_loss_clip": 1.04581189,
      "balance_loss_mlp": 1.01842606,
      "epoch": 0.7973306078278122,
      "flos": 36058621847040.0,
      "grad_norm": 1.8830406408743918,
      "language_loss": 0.77529478,
      "learning_rate": 4.154692993144862e-07,
      "loss": 0.79699862,
      "num_input_tokens_seen": 142853610,
      "step": 6631,
      "time_per_iteration": 2.6311869621276855
    },
    {
      "auxiliary_loss_clip": 0.01164311,
      "auxiliary_loss_mlp": 0.00760702,
      "balance_loss_clip": 1.04761207,
      "balance_loss_mlp": 1.00027835,
      "epoch": 0.7974508507184512,
      "flos": 21360600950400.0,
      "grad_norm": 2.017246618167033,
      "language_loss": 0.71483135,
      "learning_rate": 4.1499410915736476e-07,
      "loss": 0.73408151,
      "num_input_tokens_seen": 142872540,
      "step": 6632,
      "time_per_iteration": 2.466123104095459
    },
    {
      "auxiliary_loss_clip": 0.01048968,
      "auxiliary_loss_mlp": 0.01004043,
      "balance_loss_clip": 1.01034272,
      "balance_loss_mlp": 1.0029757,
      "epoch": 0.7975710936090904,
      "flos": 68253115317120.0,
      "grad_norm": 0.7618349192891694,
      "language_loss": 0.64265859,
      "learning_rate": 4.145191594443762e-07,
      "loss": 0.6631887,
      "num_input_tokens_seen": 142936895,
      "step": 6633,
      "time_per_iteration": 3.2255425453186035
    },
    {
      "auxiliary_loss_clip": 0.01116541,
      "auxiliary_loss_mlp": 0.01033868,
      "balance_loss_clip": 1.04356503,
      "balance_loss_mlp": 1.0264442,
      "epoch": 0.7976913364997295,
      "flos": 22492433479680.0,
      "grad_norm": 1.5259003757178222,
      "language_loss": 0.70639694,
      "learning_rate": 4.140444502475713e-07,
      "loss": 0.72790104,
      "num_input_tokens_seen": 142956445,
      "step": 6634,
      "time_per_iteration": 2.545480966567993
    },
    {
      "auxiliary_loss_clip": 0.01148883,
      "auxiliary_loss_mlp": 0.0103161,
      "balance_loss_clip": 1.04664421,
      "balance_loss_mlp": 1.02458858,
      "epoch": 0.7978115793903685,
      "flos": 15263220378240.0,
      "grad_norm": 1.7931122894568114,
      "language_loss": 0.70022273,
      "learning_rate": 4.1356998163896216e-07,
      "loss": 0.72202766,
      "num_input_tokens_seen": 142973495,
      "step": 6635,
      "time_per_iteration": 2.4722578525543213
    },
    {
      "auxiliary_loss_clip": 0.01128401,
      "auxiliary_loss_mlp": 0.01024946,
      "balance_loss_clip": 1.04485011,
      "balance_loss_mlp": 1.01804411,
      "epoch": 0.7979318222810077,
      "flos": 19719232041600.0,
      "grad_norm": 1.9760259237073492,
      "language_loss": 0.74668521,
      "learning_rate": 4.130957536905255e-07,
      "loss": 0.76821864,
      "num_input_tokens_seen": 142991510,
      "step": 6636,
      "time_per_iteration": 2.556365489959717
    },
    {
      "auxiliary_loss_clip": 0.0114273,
      "auxiliary_loss_mlp": 0.01031861,
      "balance_loss_clip": 1.04714036,
      "balance_loss_mlp": 1.02442253,
      "epoch": 0.7980520651716467,
      "flos": 15560273854080.0,
      "grad_norm": 2.4764999787077873,
      "language_loss": 0.70832217,
      "learning_rate": 4.1262176647420134e-07,
      "loss": 0.73006809,
      "num_input_tokens_seen": 143009675,
      "step": 6637,
      "time_per_iteration": 3.325406551361084
    },
    {
      "auxiliary_loss_clip": 0.01141177,
      "auxiliary_loss_mlp": 0.01025411,
      "balance_loss_clip": 1.04513645,
      "balance_loss_mlp": 1.01854479,
      "epoch": 0.7981723080622858,
      "flos": 22309432663680.0,
      "grad_norm": 1.7280980014924456,
      "language_loss": 0.79478908,
      "learning_rate": 4.121480200618923e-07,
      "loss": 0.81645489,
      "num_input_tokens_seen": 143029330,
      "step": 6638,
      "time_per_iteration": 4.008262872695923
    },
    {
      "auxiliary_loss_clip": 0.01129792,
      "auxiliary_loss_mlp": 0.01025868,
      "balance_loss_clip": 1.04415441,
      "balance_loss_mlp": 1.01836395,
      "epoch": 0.798292550952925,
      "flos": 22929573997440.0,
      "grad_norm": 1.734862808406453,
      "language_loss": 0.80258644,
      "learning_rate": 4.116745145254674e-07,
      "loss": 0.82414305,
      "num_input_tokens_seen": 143048865,
      "step": 6639,
      "time_per_iteration": 2.5202112197875977
    },
    {
      "auxiliary_loss_clip": 0.0103253,
      "auxiliary_loss_mlp": 0.01001289,
      "balance_loss_clip": 1.00897694,
      "balance_loss_mlp": 1.00018072,
      "epoch": 0.798412793843564,
      "flos": 64497936890880.0,
      "grad_norm": 0.7685837897642603,
      "language_loss": 0.58041567,
      "learning_rate": 4.1120124993675476e-07,
      "loss": 0.60075384,
      "num_input_tokens_seen": 143113295,
      "step": 6640,
      "time_per_iteration": 3.1237452030181885
    },
    {
      "auxiliary_loss_clip": 0.01144942,
      "auxiliary_loss_mlp": 0.01028602,
      "balance_loss_clip": 1.04604197,
      "balance_loss_mlp": 1.02133667,
      "epoch": 0.7985330367342031,
      "flos": 13586910514560.0,
      "grad_norm": 1.867603598885377,
      "language_loss": 0.61795938,
      "learning_rate": 4.107282263675498e-07,
      "loss": 0.63969481,
      "num_input_tokens_seen": 143130965,
      "step": 6641,
      "time_per_iteration": 2.499401330947876
    },
    {
      "auxiliary_loss_clip": 0.01033041,
      "auxiliary_loss_mlp": 0.00751179,
      "balance_loss_clip": 1.01074207,
      "balance_loss_mlp": 1.00009394,
      "epoch": 0.7986532796248422,
      "flos": 67698797656320.0,
      "grad_norm": 0.7697077992936361,
      "language_loss": 0.52542174,
      "learning_rate": 4.1025544388960907e-07,
      "loss": 0.54326397,
      "num_input_tokens_seen": 143192005,
      "step": 6642,
      "time_per_iteration": 3.150620460510254
    },
    {
      "auxiliary_loss_clip": 0.01148833,
      "auxiliary_loss_mlp": 0.01024567,
      "balance_loss_clip": 1.0469141,
      "balance_loss_mlp": 1.01733959,
      "epoch": 0.7987735225154813,
      "flos": 22455373622400.0,
      "grad_norm": 1.7254431601377087,
      "language_loss": 0.71554267,
      "learning_rate": 4.097829025746538e-07,
      "loss": 0.73727673,
      "num_input_tokens_seen": 143213550,
      "step": 6643,
      "time_per_iteration": 2.5658533573150635
    },
    {
      "auxiliary_loss_clip": 0.01046766,
      "auxiliary_loss_mlp": 0.01000616,
      "balance_loss_clip": 1.00984478,
      "balance_loss_mlp": 0.99953759,
      "epoch": 0.7988937654061203,
      "flos": 68864098682880.0,
      "grad_norm": 0.6576310862352848,
      "language_loss": 0.61018163,
      "learning_rate": 4.0931060249436757e-07,
      "loss": 0.63065547,
      "num_input_tokens_seen": 143277390,
      "step": 6644,
      "time_per_iteration": 3.118819236755371
    },
    {
      "auxiliary_loss_clip": 0.01148992,
      "auxiliary_loss_mlp": 0.01030481,
      "balance_loss_clip": 1.04840231,
      "balance_loss_mlp": 1.02314687,
      "epoch": 0.7990140082967595,
      "flos": 20806893820800.0,
      "grad_norm": 1.9479141169163285,
      "language_loss": 0.69514942,
      "learning_rate": 4.088385437203978e-07,
      "loss": 0.71694422,
      "num_input_tokens_seen": 143294400,
      "step": 6645,
      "time_per_iteration": 2.4912755489349365
    },
    {
      "auxiliary_loss_clip": 0.01165528,
      "auxiliary_loss_mlp": 0.01029238,
      "balance_loss_clip": 1.04766035,
      "balance_loss_mlp": 1.02172232,
      "epoch": 0.7991342511873986,
      "flos": 18985289443200.0,
      "grad_norm": 1.9436977801899171,
      "language_loss": 0.77824122,
      "learning_rate": 4.083667263243564e-07,
      "loss": 0.8001889,
      "num_input_tokens_seen": 143312745,
      "step": 6646,
      "time_per_iteration": 2.444019317626953
    },
    {
      "auxiliary_loss_clip": 0.01146411,
      "auxiliary_loss_mlp": 0.01029233,
      "balance_loss_clip": 1.0472039,
      "balance_loss_mlp": 1.021824,
      "epoch": 0.7992544940780376,
      "flos": 20816805974400.0,
      "grad_norm": 1.83554357887033,
      "language_loss": 0.71876597,
      "learning_rate": 4.0789515037781653e-07,
      "loss": 0.74052233,
      "num_input_tokens_seen": 143333470,
      "step": 6647,
      "time_per_iteration": 2.5194504261016846
    },
    {
      "auxiliary_loss_clip": 0.01153593,
      "auxiliary_loss_mlp": 0.0103105,
      "balance_loss_clip": 1.04677653,
      "balance_loss_mlp": 1.023772,
      "epoch": 0.7993747369686768,
      "flos": 12640772321280.0,
      "grad_norm": 1.714062898542251,
      "language_loss": 0.82390612,
      "learning_rate": 4.0742381595231755e-07,
      "loss": 0.84575254,
      "num_input_tokens_seen": 143350195,
      "step": 6648,
      "time_per_iteration": 2.4593935012817383
    },
    {
      "auxiliary_loss_clip": 0.01124924,
      "auxiliary_loss_mlp": 0.01025515,
      "balance_loss_clip": 1.04492974,
      "balance_loss_mlp": 1.01866353,
      "epoch": 0.7994949798593158,
      "flos": 20078769225600.0,
      "grad_norm": 2.535993953280498,
      "language_loss": 0.7816413,
      "learning_rate": 4.06952723119359e-07,
      "loss": 0.80314571,
      "num_input_tokens_seen": 143370070,
      "step": 6649,
      "time_per_iteration": 2.5530855655670166
    },
    {
      "auxiliary_loss_clip": 0.01130984,
      "auxiliary_loss_mlp": 0.01033476,
      "balance_loss_clip": 1.04555178,
      "balance_loss_mlp": 1.026425,
      "epoch": 0.7996152227499549,
      "flos": 38654209509120.0,
      "grad_norm": 1.8148378036008224,
      "language_loss": 0.67171383,
      "learning_rate": 4.0648187195040504e-07,
      "loss": 0.69335842,
      "num_input_tokens_seen": 143392275,
      "step": 6650,
      "time_per_iteration": 2.6629462242126465
    },
    {
      "auxiliary_loss_clip": 0.01042793,
      "auxiliary_loss_mlp": 0.01001356,
      "balance_loss_clip": 1.00783908,
      "balance_loss_mlp": 1.00024164,
      "epoch": 0.799735465640594,
      "flos": 70243821947520.0,
      "grad_norm": 1.4580852878270008,
      "language_loss": 0.67622507,
      "learning_rate": 4.060112625168848e-07,
      "loss": 0.6966666,
      "num_input_tokens_seen": 143457385,
      "step": 6651,
      "time_per_iteration": 3.171459197998047
    },
    {
      "auxiliary_loss_clip": 0.0116375,
      "auxiliary_loss_mlp": 0.01027719,
      "balance_loss_clip": 1.04872084,
      "balance_loss_mlp": 1.02021146,
      "epoch": 0.7998557085312331,
      "flos": 24240995550720.0,
      "grad_norm": 1.8402820938763091,
      "language_loss": 0.74077052,
      "learning_rate": 4.055408948901886e-07,
      "loss": 0.76268518,
      "num_input_tokens_seen": 143478785,
      "step": 6652,
      "time_per_iteration": 3.2113137245178223
    },
    {
      "auxiliary_loss_clip": 0.01153984,
      "auxiliary_loss_mlp": 0.01031917,
      "balance_loss_clip": 1.04714298,
      "balance_loss_mlp": 1.02452052,
      "epoch": 0.7999759514218722,
      "flos": 27564025449600.0,
      "grad_norm": 1.6890959991114904,
      "language_loss": 0.7120319,
      "learning_rate": 4.050707691416708e-07,
      "loss": 0.73389089,
      "num_input_tokens_seen": 143500095,
      "step": 6653,
      "time_per_iteration": 2.537909984588623
    },
    {
      "auxiliary_loss_clip": 0.01043095,
      "auxiliary_loss_mlp": 0.01001118,
      "balance_loss_clip": 1.0077951,
      "balance_loss_mlp": 1.0000447,
      "epoch": 0.8000961943125112,
      "flos": 67337428878720.0,
      "grad_norm": 0.6815958035072053,
      "language_loss": 0.5981791,
      "learning_rate": 4.046008853426495e-07,
      "loss": 0.61862123,
      "num_input_tokens_seen": 143563410,
      "step": 6654,
      "time_per_iteration": 3.1463377475738525
    },
    {
      "auxiliary_loss_clip": 0.01118954,
      "auxiliary_loss_mlp": 0.010251,
      "balance_loss_clip": 1.04474628,
      "balance_loss_mlp": 1.01783729,
      "epoch": 0.8002164372031504,
      "flos": 28733815676160.0,
      "grad_norm": 1.679718745540682,
      "language_loss": 0.62453675,
      "learning_rate": 4.0413124356440464e-07,
      "loss": 0.64597726,
      "num_input_tokens_seen": 143587455,
      "step": 6655,
      "time_per_iteration": 2.6492037773132324
    },
    {
      "auxiliary_loss_clip": 0.01110931,
      "auxiliary_loss_mlp": 0.0102511,
      "balance_loss_clip": 1.04154587,
      "balance_loss_mlp": 1.01786852,
      "epoch": 0.8003366800937894,
      "flos": 17639429725440.0,
      "grad_norm": 2.849524047350204,
      "language_loss": 0.82267046,
      "learning_rate": 4.0366184387818223e-07,
      "loss": 0.84403086,
      "num_input_tokens_seen": 143605915,
      "step": 6656,
      "time_per_iteration": 2.602039098739624
    },
    {
      "auxiliary_loss_clip": 0.01168857,
      "auxiliary_loss_mlp": 0.01024928,
      "balance_loss_clip": 1.04915285,
      "balance_loss_mlp": 1.01763225,
      "epoch": 0.8004569229844285,
      "flos": 25995303797760.0,
      "grad_norm": 2.2324361412677165,
      "language_loss": 0.84952039,
      "learning_rate": 4.0319268635518797e-07,
      "loss": 0.87145823,
      "num_input_tokens_seen": 143626490,
      "step": 6657,
      "time_per_iteration": 2.516737461090088
    },
    {
      "auxiliary_loss_clip": 0.01152529,
      "auxiliary_loss_mlp": 0.01026825,
      "balance_loss_clip": 1.04691815,
      "balance_loss_mlp": 1.0199523,
      "epoch": 0.8005771658750677,
      "flos": 20812352688000.0,
      "grad_norm": 1.495801578830262,
      "language_loss": 0.74953699,
      "learning_rate": 4.027237710665943e-07,
      "loss": 0.7713306,
      "num_input_tokens_seen": 143644955,
      "step": 6658,
      "time_per_iteration": 2.5172903537750244
    },
    {
      "auxiliary_loss_clip": 0.01127109,
      "auxiliary_loss_mlp": 0.0102462,
      "balance_loss_clip": 1.04311562,
      "balance_loss_mlp": 1.01690102,
      "epoch": 0.8006974087657067,
      "flos": 25812626204160.0,
      "grad_norm": 1.685217632641309,
      "language_loss": 0.69631088,
      "learning_rate": 4.022550980835344e-07,
      "loss": 0.71782815,
      "num_input_tokens_seen": 143667200,
      "step": 6659,
      "time_per_iteration": 2.610893487930298
    },
    {
      "auxiliary_loss_clip": 0.01120327,
      "auxiliary_loss_mlp": 0.01030568,
      "balance_loss_clip": 1.04117632,
      "balance_loss_mlp": 1.02388024,
      "epoch": 0.8008176516563458,
      "flos": 17164690646400.0,
      "grad_norm": 2.283537488800438,
      "language_loss": 0.79491341,
      "learning_rate": 4.017866674771051e-07,
      "loss": 0.81642228,
      "num_input_tokens_seen": 143684685,
      "step": 6660,
      "time_per_iteration": 2.5495524406433105
    },
    {
      "auxiliary_loss_clip": 0.0109941,
      "auxiliary_loss_mlp": 0.01027934,
      "balance_loss_clip": 1.03772807,
      "balance_loss_mlp": 1.0207696,
      "epoch": 0.8009378945469849,
      "flos": 24207311571840.0,
      "grad_norm": 1.6538804311885569,
      "language_loss": 0.74134356,
      "learning_rate": 4.013184793183688e-07,
      "loss": 0.76261705,
      "num_input_tokens_seen": 143706780,
      "step": 6661,
      "time_per_iteration": 2.6161577701568604
    },
    {
      "auxiliary_loss_clip": 0.01151512,
      "auxiliary_loss_mlp": 0.01027766,
      "balance_loss_clip": 1.04620838,
      "balance_loss_mlp": 1.02122414,
      "epoch": 0.801058137437624,
      "flos": 19787318271360.0,
      "grad_norm": 1.7714687482939586,
      "language_loss": 0.72577608,
      "learning_rate": 4.008505336783472e-07,
      "loss": 0.74756885,
      "num_input_tokens_seen": 143724505,
      "step": 6662,
      "time_per_iteration": 3.2638626098632812
    },
    {
      "auxiliary_loss_clip": 0.01141274,
      "auxiliary_loss_mlp": 0.01027749,
      "balance_loss_clip": 1.04392076,
      "balance_loss_mlp": 1.02080822,
      "epoch": 0.801178380328263,
      "flos": 18659400324480.0,
      "grad_norm": 1.7530701023660586,
      "language_loss": 0.80653691,
      "learning_rate": 4.003828306280284e-07,
      "loss": 0.82822716,
      "num_input_tokens_seen": 143742180,
      "step": 6663,
      "time_per_iteration": 3.304621696472168
    },
    {
      "auxiliary_loss_clip": 0.01148187,
      "auxiliary_loss_mlp": 0.01023801,
      "balance_loss_clip": 1.04645801,
      "balance_loss_mlp": 1.01735187,
      "epoch": 0.8012986232189022,
      "flos": 15706573948800.0,
      "grad_norm": 1.7454552554022325,
      "language_loss": 0.78191954,
      "learning_rate": 3.999153702383626e-07,
      "loss": 0.80363941,
      "num_input_tokens_seen": 143760070,
      "step": 6664,
      "time_per_iteration": 3.2237563133239746
    },
    {
      "auxiliary_loss_clip": 0.01154592,
      "auxiliary_loss_mlp": 0.01027577,
      "balance_loss_clip": 1.04677439,
      "balance_loss_mlp": 1.02038002,
      "epoch": 0.8014188661095413,
      "flos": 28584139703040.0,
      "grad_norm": 1.6331992235062247,
      "language_loss": 0.73574305,
      "learning_rate": 3.9944815258026263e-07,
      "loss": 0.75756478,
      "num_input_tokens_seen": 143781890,
      "step": 6665,
      "time_per_iteration": 2.562884569168091
    },
    {
      "auxiliary_loss_clip": 0.0115534,
      "auxiliary_loss_mlp": 0.01034389,
      "balance_loss_clip": 1.04824936,
      "balance_loss_mlp": 1.02694988,
      "epoch": 0.8015391090001803,
      "flos": 29310360877440.0,
      "grad_norm": 1.6012660306524586,
      "language_loss": 0.82895231,
      "learning_rate": 3.989811777246057e-07,
      "loss": 0.85084963,
      "num_input_tokens_seen": 143802060,
      "step": 6666,
      "time_per_iteration": 2.545543670654297
    },
    {
      "auxiliary_loss_clip": 0.01056077,
      "auxiliary_loss_mlp": 0.01002363,
      "balance_loss_clip": 1.00847709,
      "balance_loss_mlp": 1.00133157,
      "epoch": 0.8016593518908195,
      "flos": 70397340675840.0,
      "grad_norm": 0.8487935945178949,
      "language_loss": 0.66268891,
      "learning_rate": 3.985144457422305e-07,
      "loss": 0.68327332,
      "num_input_tokens_seen": 143856345,
      "step": 6667,
      "time_per_iteration": 3.001004695892334
    },
    {
      "auxiliary_loss_clip": 0.01164849,
      "auxiliary_loss_mlp": 0.01029286,
      "balance_loss_clip": 1.0479877,
      "balance_loss_mlp": 1.0227859,
      "epoch": 0.8017795947814585,
      "flos": 26026114688640.0,
      "grad_norm": 1.7844329873290972,
      "language_loss": 0.76428503,
      "learning_rate": 3.9804795670394096e-07,
      "loss": 0.78622633,
      "num_input_tokens_seen": 143876470,
      "step": 6668,
      "time_per_iteration": 2.48689603805542
    },
    {
      "auxiliary_loss_clip": 0.01127435,
      "auxiliary_loss_mlp": 0.01031529,
      "balance_loss_clip": 1.04303479,
      "balance_loss_mlp": 1.02462065,
      "epoch": 0.8018998376720976,
      "flos": 22087181260800.0,
      "grad_norm": 1.641842118719,
      "language_loss": 0.70469052,
      "learning_rate": 3.975817106805022e-07,
      "loss": 0.72628009,
      "num_input_tokens_seen": 143895170,
      "step": 6669,
      "time_per_iteration": 2.5324020385742188
    },
    {
      "auxiliary_loss_clip": 0.01125939,
      "auxiliary_loss_mlp": 0.01027595,
      "balance_loss_clip": 1.04499674,
      "balance_loss_mlp": 1.02050543,
      "epoch": 0.8020200805627368,
      "flos": 34568545023360.0,
      "grad_norm": 1.9323712904365193,
      "language_loss": 0.65140826,
      "learning_rate": 3.97115707742645e-07,
      "loss": 0.67294359,
      "num_input_tokens_seen": 143915845,
      "step": 6670,
      "time_per_iteration": 2.674323558807373
    },
    {
      "auxiliary_loss_clip": 0.01138966,
      "auxiliary_loss_mlp": 0.01030608,
      "balance_loss_clip": 1.04710543,
      "balance_loss_mlp": 1.02381265,
      "epoch": 0.8021403234533758,
      "flos": 20120354196480.0,
      "grad_norm": 1.7624285274749412,
      "language_loss": 0.64961398,
      "learning_rate": 3.966499479610599e-07,
      "loss": 0.67130971,
      "num_input_tokens_seen": 143933940,
      "step": 6671,
      "time_per_iteration": 2.5324535369873047
    },
    {
      "auxiliary_loss_clip": 0.01119725,
      "auxiliary_loss_mlp": 0.0102988,
      "balance_loss_clip": 1.04435897,
      "balance_loss_mlp": 1.02337706,
      "epoch": 0.8022605663440149,
      "flos": 27746200252800.0,
      "grad_norm": 2.03527951757548,
      "language_loss": 0.65211892,
      "learning_rate": 3.9618443140640225e-07,
      "loss": 0.67361498,
      "num_input_tokens_seen": 143952850,
      "step": 6672,
      "time_per_iteration": 2.608290433883667
    },
    {
      "auxiliary_loss_clip": 0.01012436,
      "auxiliary_loss_mlp": 0.01000528,
      "balance_loss_clip": 1.00780821,
      "balance_loss_mlp": 0.99947882,
      "epoch": 0.802380809234654,
      "flos": 60244998768000.0,
      "grad_norm": 0.6880317878935087,
      "language_loss": 0.51449841,
      "learning_rate": 3.957191581492918e-07,
      "loss": 0.53462803,
      "num_input_tokens_seen": 144013610,
      "step": 6673,
      "time_per_iteration": 3.183957099914551
    },
    {
      "auxiliary_loss_clip": 0.01135307,
      "auxiliary_loss_mlp": 0.01029182,
      "balance_loss_clip": 1.0466615,
      "balance_loss_mlp": 1.02146304,
      "epoch": 0.8025010521252931,
      "flos": 15080722352640.0,
      "grad_norm": 2.9885558338290727,
      "language_loss": 0.71160913,
      "learning_rate": 3.952541282603097e-07,
      "loss": 0.73325408,
      "num_input_tokens_seen": 144028715,
      "step": 6674,
      "time_per_iteration": 2.4792304039001465
    },
    {
      "auxiliary_loss_clip": 0.01150101,
      "auxiliary_loss_mlp": 0.01021727,
      "balance_loss_clip": 1.04842544,
      "balance_loss_mlp": 1.01458883,
      "epoch": 0.8026212950159322,
      "flos": 22163527618560.0,
      "grad_norm": 1.9290309047659324,
      "language_loss": 0.83466089,
      "learning_rate": 3.9478934181000013e-07,
      "loss": 0.85637915,
      "num_input_tokens_seen": 144048740,
      "step": 6675,
      "time_per_iteration": 2.503544569015503
    },
    {
      "auxiliary_loss_clip": 0.01170032,
      "auxiliary_loss_mlp": 0.01030372,
      "balance_loss_clip": 1.050017,
      "balance_loss_mlp": 1.02296877,
      "epoch": 0.8027415379065713,
      "flos": 17675986792320.0,
      "grad_norm": 2.365832707706217,
      "language_loss": 0.84222734,
      "learning_rate": 3.943247988688714e-07,
      "loss": 0.86423135,
      "num_input_tokens_seen": 144067435,
      "step": 6676,
      "time_per_iteration": 2.4614522457122803
    },
    {
      "auxiliary_loss_clip": 0.01150056,
      "auxiliary_loss_mlp": 0.01024906,
      "balance_loss_clip": 1.04495823,
      "balance_loss_mlp": 1.01781607,
      "epoch": 0.8028617807972104,
      "flos": 21979593048960.0,
      "grad_norm": 1.6478960613640738,
      "language_loss": 0.7192058,
      "learning_rate": 3.938604995073933e-07,
      "loss": 0.74095541,
      "num_input_tokens_seen": 144085905,
      "step": 6677,
      "time_per_iteration": 2.518571376800537
    },
    {
      "auxiliary_loss_clip": 0.0114096,
      "auxiliary_loss_mlp": 0.0103284,
      "balance_loss_clip": 1.04631925,
      "balance_loss_mlp": 1.02557397,
      "epoch": 0.8029820236878494,
      "flos": 26428457905920.0,
      "grad_norm": 1.605553215703312,
      "language_loss": 0.65403855,
      "learning_rate": 3.9339644379600157e-07,
      "loss": 0.67577654,
      "num_input_tokens_seen": 144105735,
      "step": 6678,
      "time_per_iteration": 3.426527738571167
    },
    {
      "auxiliary_loss_clip": 0.01149863,
      "auxiliary_loss_mlp": 0.01030208,
      "balance_loss_clip": 1.04743719,
      "balance_loss_mlp": 1.02325845,
      "epoch": 0.8031022665784886,
      "flos": 17676489582720.0,
      "grad_norm": 2.643536410323045,
      "language_loss": 0.71594584,
      "learning_rate": 3.929326318050907e-07,
      "loss": 0.7377466,
      "num_input_tokens_seen": 144123405,
      "step": 6679,
      "time_per_iteration": 2.4547054767608643
    },
    {
      "auxiliary_loss_clip": 0.01161236,
      "auxiliary_loss_mlp": 0.01021845,
      "balance_loss_clip": 1.04605997,
      "balance_loss_mlp": 1.01514888,
      "epoch": 0.8032225094691277,
      "flos": 15450279431040.0,
      "grad_norm": 1.7960272059538764,
      "language_loss": 0.79191053,
      "learning_rate": 3.924690636050225e-07,
      "loss": 0.81374133,
      "num_input_tokens_seen": 144140815,
      "step": 6680,
      "time_per_iteration": 2.445206642150879
    },
    {
      "auxiliary_loss_clip": 0.011528,
      "auxiliary_loss_mlp": 0.01031975,
      "balance_loss_clip": 1.04813957,
      "balance_loss_mlp": 1.02496529,
      "epoch": 0.8033427523597667,
      "flos": 26179202453760.0,
      "grad_norm": 1.8894868394396995,
      "language_loss": 0.72918546,
      "learning_rate": 3.9200573926611915e-07,
      "loss": 0.75103325,
      "num_input_tokens_seen": 144162230,
      "step": 6681,
      "time_per_iteration": 2.5514700412750244
    },
    {
      "auxiliary_loss_clip": 0.01149317,
      "auxiliary_loss_mlp": 0.01025578,
      "balance_loss_clip": 1.04880595,
      "balance_loss_mlp": 1.01893485,
      "epoch": 0.8034629952504058,
      "flos": 21324905809920.0,
      "grad_norm": 2.0276199517586733,
      "language_loss": 0.72728634,
      "learning_rate": 3.9154265885866613e-07,
      "loss": 0.74903524,
      "num_input_tokens_seen": 144181540,
      "step": 6682,
      "time_per_iteration": 2.5415868759155273
    },
    {
      "auxiliary_loss_clip": 0.01151347,
      "auxiliary_loss_mlp": 0.01028849,
      "balance_loss_clip": 1.04850113,
      "balance_loss_mlp": 1.02099061,
      "epoch": 0.8035832381410449,
      "flos": 21651585027840.0,
      "grad_norm": 5.854196220570995,
      "language_loss": 0.74186611,
      "learning_rate": 3.9107982245291394e-07,
      "loss": 0.76366818,
      "num_input_tokens_seen": 144199665,
      "step": 6683,
      "time_per_iteration": 2.476905107498169
    },
    {
      "auxiliary_loss_clip": 0.01124317,
      "auxiliary_loss_mlp": 0.01023802,
      "balance_loss_clip": 1.04688287,
      "balance_loss_mlp": 1.01668859,
      "epoch": 0.803703481031684,
      "flos": 20518818744960.0,
      "grad_norm": 1.9915898959440057,
      "language_loss": 0.77391601,
      "learning_rate": 3.9061723011907245e-07,
      "loss": 0.79539716,
      "num_input_tokens_seen": 144219020,
      "step": 6684,
      "time_per_iteration": 2.559946060180664
    },
    {
      "auxiliary_loss_clip": 0.01136992,
      "auxiliary_loss_mlp": 0.0102626,
      "balance_loss_clip": 1.04439902,
      "balance_loss_mlp": 1.01872897,
      "epoch": 0.803823723922323,
      "flos": 22854807838080.0,
      "grad_norm": 2.0954607479584393,
      "language_loss": 0.79343569,
      "learning_rate": 3.901548819273179e-07,
      "loss": 0.81506819,
      "num_input_tokens_seen": 144239035,
      "step": 6685,
      "time_per_iteration": 2.5323619842529297
    },
    {
      "auxiliary_loss_clip": 0.01151585,
      "auxiliary_loss_mlp": 0.0102857,
      "balance_loss_clip": 1.04761231,
      "balance_loss_mlp": 1.02142668,
      "epoch": 0.8039439668129622,
      "flos": 21362145235200.0,
      "grad_norm": 1.9447530107700597,
      "language_loss": 0.69353664,
      "learning_rate": 3.896927779477881e-07,
      "loss": 0.71533823,
      "num_input_tokens_seen": 144258295,
      "step": 6686,
      "time_per_iteration": 2.4954659938812256
    },
    {
      "auxiliary_loss_clip": 0.01121965,
      "auxiliary_loss_mlp": 0.01023998,
      "balance_loss_clip": 1.04195094,
      "balance_loss_mlp": 1.0173198,
      "epoch": 0.8040642097036013,
      "flos": 23802382575360.0,
      "grad_norm": 2.1215832357854043,
      "language_loss": 0.67175412,
      "learning_rate": 3.892309182505833e-07,
      "loss": 0.69321376,
      "num_input_tokens_seen": 144276110,
      "step": 6687,
      "time_per_iteration": 2.5862650871276855
    },
    {
      "auxiliary_loss_clip": 0.01161662,
      "auxiliary_loss_mlp": 0.0102387,
      "balance_loss_clip": 1.04596066,
      "balance_loss_mlp": 1.01703632,
      "epoch": 0.8041844525942403,
      "flos": 25922046009600.0,
      "grad_norm": 1.9449071905057924,
      "language_loss": 0.86011076,
      "learning_rate": 3.887693029057675e-07,
      "loss": 0.88196611,
      "num_input_tokens_seen": 144295620,
      "step": 6688,
      "time_per_iteration": 2.496030569076538
    },
    {
      "auxiliary_loss_clip": 0.01136816,
      "auxiliary_loss_mlp": 0.01019386,
      "balance_loss_clip": 1.04496086,
      "balance_loss_mlp": 1.01280534,
      "epoch": 0.8043046954848795,
      "flos": 25191120153600.0,
      "grad_norm": 1.8961911110740348,
      "language_loss": 0.81279945,
      "learning_rate": 3.8830793198336684e-07,
      "loss": 0.83436143,
      "num_input_tokens_seen": 144315210,
      "step": 6689,
      "time_per_iteration": 4.181029319763184
    },
    {
      "auxiliary_loss_clip": 0.01152948,
      "auxiliary_loss_mlp": 0.0102833,
      "balance_loss_clip": 1.04657817,
      "balance_loss_mlp": 1.02080822,
      "epoch": 0.8044249383755185,
      "flos": 41719185123840.0,
      "grad_norm": 1.5776949225936465,
      "language_loss": 0.70259988,
      "learning_rate": 3.878468055533721e-07,
      "loss": 0.72441268,
      "num_input_tokens_seen": 144337750,
      "step": 6690,
      "time_per_iteration": 3.3972089290618896
    },
    {
      "auxiliary_loss_clip": 0.01132415,
      "auxiliary_loss_mlp": 0.01035214,
      "balance_loss_clip": 1.04741144,
      "balance_loss_mlp": 1.02752233,
      "epoch": 0.8045451812661576,
      "flos": 20631434860800.0,
      "grad_norm": 2.4804830211681663,
      "language_loss": 0.84846658,
      "learning_rate": 3.8738592368573464e-07,
      "loss": 0.87014282,
      "num_input_tokens_seen": 144355305,
      "step": 6691,
      "time_per_iteration": 2.5666115283966064
    },
    {
      "auxiliary_loss_clip": 0.01114125,
      "auxiliary_loss_mlp": 0.01030963,
      "balance_loss_clip": 1.04348254,
      "balance_loss_mlp": 1.0236522,
      "epoch": 0.8046654241567968,
      "flos": 29711806254720.0,
      "grad_norm": 1.841292785953817,
      "language_loss": 0.88128179,
      "learning_rate": 3.8692528645037137e-07,
      "loss": 0.90273273,
      "num_input_tokens_seen": 144374485,
      "step": 6692,
      "time_per_iteration": 2.6131651401519775
    },
    {
      "auxiliary_loss_clip": 0.01164159,
      "auxiliary_loss_mlp": 0.01031742,
      "balance_loss_clip": 1.04866779,
      "balance_loss_mlp": 1.02431548,
      "epoch": 0.8047856670474358,
      "flos": 17671389851520.0,
      "grad_norm": 2.001952394632532,
      "language_loss": 0.7796886,
      "learning_rate": 3.8646489391715907e-07,
      "loss": 0.8016476,
      "num_input_tokens_seen": 144388780,
      "step": 6693,
      "time_per_iteration": 2.4217946529388428
    },
    {
      "auxiliary_loss_clip": 0.01135809,
      "auxiliary_loss_mlp": 0.01027924,
      "balance_loss_clip": 1.04510224,
      "balance_loss_mlp": 1.02065182,
      "epoch": 0.8049059099380749,
      "flos": 17120699464320.0,
      "grad_norm": 4.5463865477578835,
      "language_loss": 0.87971389,
      "learning_rate": 3.8600474615593903e-07,
      "loss": 0.90135121,
      "num_input_tokens_seen": 144403395,
      "step": 6694,
      "time_per_iteration": 2.4824376106262207
    },
    {
      "auxiliary_loss_clip": 0.01029227,
      "auxiliary_loss_mlp": 0.0100294,
      "balance_loss_clip": 1.01031947,
      "balance_loss_mlp": 1.00174773,
      "epoch": 0.805026152828714,
      "flos": 62212903240320.0,
      "grad_norm": 0.779508188922491,
      "language_loss": 0.59740317,
      "learning_rate": 3.8554484323651605e-07,
      "loss": 0.61772478,
      "num_input_tokens_seen": 144465265,
      "step": 6695,
      "time_per_iteration": 3.220430374145508
    },
    {
      "auxiliary_loss_clip": 0.01148319,
      "auxiliary_loss_mlp": 0.00761062,
      "balance_loss_clip": 1.04554617,
      "balance_loss_mlp": 1.00029039,
      "epoch": 0.8051463957193531,
      "flos": 21688608971520.0,
      "grad_norm": 1.4743117208008705,
      "language_loss": 0.79375672,
      "learning_rate": 3.85085185228657e-07,
      "loss": 0.81285048,
      "num_input_tokens_seen": 144484235,
      "step": 6696,
      "time_per_iteration": 2.518263816833496
    },
    {
      "auxiliary_loss_clip": 0.01132258,
      "auxiliary_loss_mlp": 0.01030495,
      "balance_loss_clip": 1.04550076,
      "balance_loss_mlp": 1.02361643,
      "epoch": 0.8052666386099921,
      "flos": 32051458535040.0,
      "grad_norm": 1.8142725539054405,
      "language_loss": 0.73034286,
      "learning_rate": 3.8462577220209114e-07,
      "loss": 0.75197041,
      "num_input_tokens_seen": 144504610,
      "step": 6697,
      "time_per_iteration": 2.6155641078948975
    },
    {
      "auxiliary_loss_clip": 0.0105543,
      "auxiliary_loss_mlp": 0.0100095,
      "balance_loss_clip": 1.00794053,
      "balance_loss_mlp": 0.9998886,
      "epoch": 0.8053868815006313,
      "flos": 67157875768320.0,
      "grad_norm": 0.702858997951354,
      "language_loss": 0.5902074,
      "learning_rate": 3.8416660422651127e-07,
      "loss": 0.61077118,
      "num_input_tokens_seen": 144574260,
      "step": 6698,
      "time_per_iteration": 3.152697801589966
    },
    {
      "auxiliary_loss_clip": 0.01125314,
      "auxiliary_loss_mlp": 0.01022617,
      "balance_loss_clip": 1.04254293,
      "balance_loss_mlp": 1.01526213,
      "epoch": 0.8055071243912704,
      "flos": 23837000307840.0,
      "grad_norm": 1.7597572798502605,
      "language_loss": 0.67632967,
      "learning_rate": 3.837076813715723e-07,
      "loss": 0.69780898,
      "num_input_tokens_seen": 144594145,
      "step": 6699,
      "time_per_iteration": 2.5903286933898926
    },
    {
      "auxiliary_loss_clip": 0.01119698,
      "auxiliary_loss_mlp": 0.01027848,
      "balance_loss_clip": 1.04146528,
      "balance_loss_mlp": 1.01999247,
      "epoch": 0.8056273672819094,
      "flos": 21324510760320.0,
      "grad_norm": 1.742423916132846,
      "language_loss": 0.75181413,
      "learning_rate": 3.832490037068941e-07,
      "loss": 0.77328962,
      "num_input_tokens_seen": 144612935,
      "step": 6700,
      "time_per_iteration": 2.563020706176758
    },
    {
      "auxiliary_loss_clip": 0.01091064,
      "auxiliary_loss_mlp": 0.01026101,
      "balance_loss_clip": 1.03992414,
      "balance_loss_mlp": 1.01940489,
      "epoch": 0.8057476101725486,
      "flos": 25768383626880.0,
      "grad_norm": 1.8379550891336747,
      "language_loss": 0.75529301,
      "learning_rate": 3.827905713020554e-07,
      "loss": 0.7764647,
      "num_input_tokens_seen": 144630580,
      "step": 6701,
      "time_per_iteration": 2.645224094390869
    },
    {
      "auxiliary_loss_clip": 0.0112464,
      "auxiliary_loss_mlp": 0.01030745,
      "balance_loss_clip": 1.03975511,
      "balance_loss_mlp": 1.02241862,
      "epoch": 0.8058678530631876,
      "flos": 24535283679360.0,
      "grad_norm": 1.9727058312082406,
      "language_loss": 0.68778884,
      "learning_rate": 3.823323842266017e-07,
      "loss": 0.70934272,
      "num_input_tokens_seen": 144649975,
      "step": 6702,
      "time_per_iteration": 2.6198668479919434
    },
    {
      "auxiliary_loss_clip": 0.01151788,
      "auxiliary_loss_mlp": 0.01033394,
      "balance_loss_clip": 1.04427111,
      "balance_loss_mlp": 1.02631545,
      "epoch": 0.8059880959538267,
      "flos": 24753728240640.0,
      "grad_norm": 2.6559278005139912,
      "language_loss": 0.72923744,
      "learning_rate": 3.818744425500393e-07,
      "loss": 0.75108922,
      "num_input_tokens_seen": 144667990,
      "step": 6703,
      "time_per_iteration": 2.526411771774292
    },
    {
      "auxiliary_loss_clip": 0.01117412,
      "auxiliary_loss_mlp": 0.01029744,
      "balance_loss_clip": 1.04204512,
      "balance_loss_mlp": 1.0224216,
      "epoch": 0.8061083388444659,
      "flos": 22196349671040.0,
      "grad_norm": 1.6525712077960464,
      "language_loss": 0.80407345,
      "learning_rate": 3.8141674634183675e-07,
      "loss": 0.82554501,
      "num_input_tokens_seen": 144687020,
      "step": 6704,
      "time_per_iteration": 3.369621515274048
    },
    {
      "auxiliary_loss_clip": 0.01106729,
      "auxiliary_loss_mlp": 0.01028758,
      "balance_loss_clip": 1.04418302,
      "balance_loss_mlp": 1.02177227,
      "epoch": 0.8062285817351049,
      "flos": 30044195735040.0,
      "grad_norm": 2.0343958009564957,
      "language_loss": 0.66459018,
      "learning_rate": 3.809592956714278e-07,
      "loss": 0.68594503,
      "num_input_tokens_seen": 144710255,
      "step": 6705,
      "time_per_iteration": 2.645833969116211
    },
    {
      "auxiliary_loss_clip": 0.01157221,
      "auxiliary_loss_mlp": 0.01034193,
      "balance_loss_clip": 1.04933286,
      "balance_loss_mlp": 1.0267067,
      "epoch": 0.806348824625744,
      "flos": 22782591544320.0,
      "grad_norm": 1.8477915980738315,
      "language_loss": 0.74499363,
      "learning_rate": 3.805020906082057e-07,
      "loss": 0.76690781,
      "num_input_tokens_seen": 144728830,
      "step": 6706,
      "time_per_iteration": 2.496737003326416
    },
    {
      "auxiliary_loss_clip": 0.01138799,
      "auxiliary_loss_mlp": 0.0102896,
      "balance_loss_clip": 1.04427445,
      "balance_loss_mlp": 1.02125561,
      "epoch": 0.8064690675163831,
      "flos": 23404600385280.0,
      "grad_norm": 2.060984940040578,
      "language_loss": 0.80828321,
      "learning_rate": 3.8004513122152917e-07,
      "loss": 0.82996076,
      "num_input_tokens_seen": 144747140,
      "step": 6707,
      "time_per_iteration": 2.5241291522979736
    },
    {
      "auxiliary_loss_clip": 0.01124897,
      "auxiliary_loss_mlp": 0.01026539,
      "balance_loss_clip": 1.04355729,
      "balance_loss_mlp": 1.01997626,
      "epoch": 0.8065893104070222,
      "flos": 24060903736320.0,
      "grad_norm": 1.7450470920517303,
      "language_loss": 0.67172277,
      "learning_rate": 3.79588417580718e-07,
      "loss": 0.69323719,
      "num_input_tokens_seen": 144765250,
      "step": 6708,
      "time_per_iteration": 2.547727108001709
    },
    {
      "auxiliary_loss_clip": 0.01152235,
      "auxiliary_loss_mlp": 0.01029335,
      "balance_loss_clip": 1.0465771,
      "balance_loss_mlp": 1.02228403,
      "epoch": 0.8067095532976613,
      "flos": 22305410340480.0,
      "grad_norm": 1.83534674388313,
      "language_loss": 0.76346213,
      "learning_rate": 3.791319497550558e-07,
      "loss": 0.78527784,
      "num_input_tokens_seen": 144783080,
      "step": 6709,
      "time_per_iteration": 2.47678804397583
    },
    {
      "auxiliary_loss_clip": 0.01126549,
      "auxiliary_loss_mlp": 0.00760902,
      "balance_loss_clip": 1.04563761,
      "balance_loss_mlp": 1.00025964,
      "epoch": 0.8068297961883004,
      "flos": 17129498296320.0,
      "grad_norm": 1.903325463263411,
      "language_loss": 0.70772928,
      "learning_rate": 3.78675727813788e-07,
      "loss": 0.72660381,
      "num_input_tokens_seen": 144800645,
      "step": 6710,
      "time_per_iteration": 2.5286831855773926
    },
    {
      "auxiliary_loss_clip": 0.01137828,
      "auxiliary_loss_mlp": 0.01028915,
      "balance_loss_clip": 1.04566443,
      "balance_loss_mlp": 1.02161372,
      "epoch": 0.8069500390789395,
      "flos": 22018843635840.0,
      "grad_norm": 1.6049888872539804,
      "language_loss": 0.73669183,
      "learning_rate": 3.782197518261225e-07,
      "loss": 0.75835925,
      "num_input_tokens_seen": 144820085,
      "step": 6711,
      "time_per_iteration": 2.542816638946533
    },
    {
      "auxiliary_loss_clip": 0.01144603,
      "auxiliary_loss_mlp": 0.01024824,
      "balance_loss_clip": 1.04745007,
      "balance_loss_mlp": 1.01786804,
      "epoch": 0.8070702819695785,
      "flos": 19244241567360.0,
      "grad_norm": 1.937689601658259,
      "language_loss": 0.95622593,
      "learning_rate": 3.777640218612319e-07,
      "loss": 0.97792017,
      "num_input_tokens_seen": 144838070,
      "step": 6712,
      "time_per_iteration": 2.5099611282348633
    },
    {
      "auxiliary_loss_clip": 0.01145868,
      "auxiliary_loss_mlp": 0.01024339,
      "balance_loss_clip": 1.04698896,
      "balance_loss_mlp": 1.01722217,
      "epoch": 0.8071905248602176,
      "flos": 21544320038400.0,
      "grad_norm": 2.5282700047701567,
      "language_loss": 0.72120076,
      "learning_rate": 3.773085379882488e-07,
      "loss": 0.74290276,
      "num_input_tokens_seen": 144857125,
      "step": 6713,
      "time_per_iteration": 2.4920568466186523
    },
    {
      "auxiliary_loss_clip": 0.01152537,
      "auxiliary_loss_mlp": 0.0076128,
      "balance_loss_clip": 1.04582644,
      "balance_loss_mlp": 1.00032365,
      "epoch": 0.8073107677508568,
      "flos": 37268309105280.0,
      "grad_norm": 1.926765461329152,
      "language_loss": 0.75744176,
      "learning_rate": 3.768533002762715e-07,
      "loss": 0.77657992,
      "num_input_tokens_seen": 144880660,
      "step": 6714,
      "time_per_iteration": 2.627150535583496
    },
    {
      "auxiliary_loss_clip": 0.01135223,
      "auxiliary_loss_mlp": 0.01025903,
      "balance_loss_clip": 1.04274833,
      "balance_loss_mlp": 1.01931715,
      "epoch": 0.8074310106414958,
      "flos": 28366269759360.0,
      "grad_norm": 1.645101991601082,
      "language_loss": 0.76655716,
      "learning_rate": 3.763983087943572e-07,
      "loss": 0.78816843,
      "num_input_tokens_seen": 144900050,
      "step": 6715,
      "time_per_iteration": 4.181788444519043
    },
    {
      "auxiliary_loss_clip": 0.01142821,
      "auxiliary_loss_mlp": 0.00760995,
      "balance_loss_clip": 1.04484391,
      "balance_loss_mlp": 1.00030255,
      "epoch": 0.8075512535321349,
      "flos": 24281646768000.0,
      "grad_norm": 1.6026081494774203,
      "language_loss": 0.80991662,
      "learning_rate": 3.759435636115282e-07,
      "loss": 0.82895476,
      "num_input_tokens_seen": 144920835,
      "step": 6716,
      "time_per_iteration": 3.1999378204345703
    },
    {
      "auxiliary_loss_clip": 0.01088327,
      "auxiliary_loss_mlp": 0.00760457,
      "balance_loss_clip": 1.04193711,
      "balance_loss_mlp": 1.00029731,
      "epoch": 0.807671496422774,
      "flos": 26030855283840.0,
      "grad_norm": 1.704914704259896,
      "language_loss": 0.72834814,
      "learning_rate": 3.7548906479676967e-07,
      "loss": 0.74683595,
      "num_input_tokens_seen": 144940430,
      "step": 6717,
      "time_per_iteration": 2.6426594257354736
    },
    {
      "auxiliary_loss_clip": 0.01153823,
      "auxiliary_loss_mlp": 0.0102417,
      "balance_loss_clip": 1.04604197,
      "balance_loss_mlp": 1.01717198,
      "epoch": 0.8077917393134131,
      "flos": 23730740899200.0,
      "grad_norm": 1.5988757626103358,
      "language_loss": 0.71594125,
      "learning_rate": 3.7503481241902855e-07,
      "loss": 0.7377212,
      "num_input_tokens_seen": 144960405,
      "step": 6718,
      "time_per_iteration": 2.51042103767395
    },
    {
      "auxiliary_loss_clip": 0.01137502,
      "auxiliary_loss_mlp": 0.00760262,
      "balance_loss_clip": 1.045223,
      "balance_loss_mlp": 1.00028014,
      "epoch": 0.8079119822040521,
      "flos": 18402028398720.0,
      "grad_norm": 1.5924042388100152,
      "language_loss": 0.80211306,
      "learning_rate": 3.745808065472145e-07,
      "loss": 0.8210907,
      "num_input_tokens_seen": 144977700,
      "step": 6719,
      "time_per_iteration": 2.4997830390930176
    },
    {
      "auxiliary_loss_clip": 0.01147754,
      "auxiliary_loss_mlp": 0.01027129,
      "balance_loss_clip": 1.04965103,
      "balance_loss_mlp": 1.02073944,
      "epoch": 0.8080322250946913,
      "flos": 23621787970560.0,
      "grad_norm": 1.4719904616025614,
      "language_loss": 0.76315981,
      "learning_rate": 3.741270472501994e-07,
      "loss": 0.78490865,
      "num_input_tokens_seen": 144998340,
      "step": 6720,
      "time_per_iteration": 2.5011088848114014
    },
    {
      "auxiliary_loss_clip": 0.01135769,
      "auxiliary_loss_mlp": 0.01024271,
      "balance_loss_clip": 1.04733884,
      "balance_loss_mlp": 1.0172075,
      "epoch": 0.8081524679853304,
      "flos": 22820692896000.0,
      "grad_norm": 1.618210669297089,
      "language_loss": 0.72948617,
      "learning_rate": 3.736735345968183e-07,
      "loss": 0.75108659,
      "num_input_tokens_seen": 145017950,
      "step": 6721,
      "time_per_iteration": 2.5314385890960693
    },
    {
      "auxiliary_loss_clip": 0.01152405,
      "auxiliary_loss_mlp": 0.01028242,
      "balance_loss_clip": 1.04780173,
      "balance_loss_mlp": 1.02115798,
      "epoch": 0.8082727108759694,
      "flos": 17640004343040.0,
      "grad_norm": 1.5907601751374836,
      "language_loss": 0.78560448,
      "learning_rate": 3.7322026865586986e-07,
      "loss": 0.80741096,
      "num_input_tokens_seen": 145036985,
      "step": 6722,
      "time_per_iteration": 2.458688974380493
    },
    {
      "auxiliary_loss_clip": 0.011605,
      "auxiliary_loss_mlp": 0.01026154,
      "balance_loss_clip": 1.05203772,
      "balance_loss_mlp": 1.01898909,
      "epoch": 0.8083929537666086,
      "flos": 25958172113280.0,
      "grad_norm": 1.8972568071915916,
      "language_loss": 0.73357159,
      "learning_rate": 3.7276724949611206e-07,
      "loss": 0.75543815,
      "num_input_tokens_seen": 145057095,
      "step": 6723,
      "time_per_iteration": 2.520411729812622
    },
    {
      "auxiliary_loss_clip": 0.01140305,
      "auxiliary_loss_mlp": 0.01031713,
      "balance_loss_clip": 1.04695344,
      "balance_loss_mlp": 1.02400637,
      "epoch": 0.8085131966572476,
      "flos": 27089178629760.0,
      "grad_norm": 1.8306855338065091,
      "language_loss": 0.7516911,
      "learning_rate": 3.723144771862694e-07,
      "loss": 0.77341133,
      "num_input_tokens_seen": 145077735,
      "step": 6724,
      "time_per_iteration": 2.5508809089660645
    },
    {
      "auxiliary_loss_clip": 0.01126359,
      "auxiliary_loss_mlp": 0.01029275,
      "balance_loss_clip": 1.0433315,
      "balance_loss_mlp": 1.02262902,
      "epoch": 0.8086334395478867,
      "flos": 23988543788160.0,
      "grad_norm": 1.5236435885284427,
      "language_loss": 0.76609969,
      "learning_rate": 3.718619517950263e-07,
      "loss": 0.78765601,
      "num_input_tokens_seen": 145098330,
      "step": 6725,
      "time_per_iteration": 2.5840864181518555
    },
    {
      "auxiliary_loss_clip": 0.01166679,
      "auxiliary_loss_mlp": 0.0102606,
      "balance_loss_clip": 1.05063558,
      "balance_loss_mlp": 1.01946223,
      "epoch": 0.8087536824385259,
      "flos": 20405879406720.0,
      "grad_norm": 1.7449578403070263,
      "language_loss": 0.76797557,
      "learning_rate": 3.714096733910301e-07,
      "loss": 0.78990299,
      "num_input_tokens_seen": 145115855,
      "step": 6726,
      "time_per_iteration": 2.4257378578186035
    },
    {
      "auxiliary_loss_clip": 0.01156273,
      "auxiliary_loss_mlp": 0.01027665,
      "balance_loss_clip": 1.04830873,
      "balance_loss_mlp": 1.02041137,
      "epoch": 0.8088739253291649,
      "flos": 25919639798400.0,
      "grad_norm": 1.971728595241666,
      "language_loss": 0.70091099,
      "learning_rate": 3.709576420428926e-07,
      "loss": 0.72275031,
      "num_input_tokens_seen": 145136655,
      "step": 6727,
      "time_per_iteration": 2.5234479904174805
    },
    {
      "auxiliary_loss_clip": 0.01136417,
      "auxiliary_loss_mlp": 0.01023025,
      "balance_loss_clip": 1.04202986,
      "balance_loss_mlp": 1.01636732,
      "epoch": 0.808994168219804,
      "flos": 28402072640640.0,
      "grad_norm": 1.970576635328107,
      "language_loss": 0.73199034,
      "learning_rate": 3.7050585781918463e-07,
      "loss": 0.75358474,
      "num_input_tokens_seen": 145156955,
      "step": 6728,
      "time_per_iteration": 2.558981418609619
    },
    {
      "auxiliary_loss_clip": 0.01154172,
      "auxiliary_loss_mlp": 0.01032518,
      "balance_loss_clip": 1.04801977,
      "balance_loss_mlp": 1.02524877,
      "epoch": 0.8091144111104431,
      "flos": 17421056991360.0,
      "grad_norm": 1.9345798547360664,
      "language_loss": 0.69021749,
      "learning_rate": 3.700543207884428e-07,
      "loss": 0.71208435,
      "num_input_tokens_seen": 145173865,
      "step": 6729,
      "time_per_iteration": 2.46881103515625
    },
    {
      "auxiliary_loss_clip": 0.01151473,
      "auxiliary_loss_mlp": 0.01028184,
      "balance_loss_clip": 1.04837394,
      "balance_loss_mlp": 1.02124608,
      "epoch": 0.8092346540010822,
      "flos": 32153803361280.0,
      "grad_norm": 1.6996815326324561,
      "language_loss": 0.70940953,
      "learning_rate": 3.6960303101916466e-07,
      "loss": 0.73120612,
      "num_input_tokens_seen": 145193780,
      "step": 6730,
      "time_per_iteration": 3.2852981090545654
    },
    {
      "auxiliary_loss_clip": 0.01055522,
      "auxiliary_loss_mlp": 0.00751323,
      "balance_loss_clip": 1.00796902,
      "balance_loss_mlp": 1.00009811,
      "epoch": 0.8093548968917212,
      "flos": 58035093390720.0,
      "grad_norm": 0.7618086512161657,
      "language_loss": 0.55576146,
      "learning_rate": 3.6915198857981047e-07,
      "loss": 0.57382989,
      "num_input_tokens_seen": 145258980,
      "step": 6731,
      "time_per_iteration": 3.10363507270813
    },
    {
      "auxiliary_loss_clip": 0.01118688,
      "auxiliary_loss_mlp": 0.01031525,
      "balance_loss_clip": 1.04278994,
      "balance_loss_mlp": 1.02448559,
      "epoch": 0.8094751397823604,
      "flos": 27381599251200.0,
      "grad_norm": 1.8236806996200032,
      "language_loss": 0.68096799,
      "learning_rate": 3.687011935388027e-07,
      "loss": 0.70247012,
      "num_input_tokens_seen": 145281875,
      "step": 6732,
      "time_per_iteration": 2.608764410018921
    },
    {
      "auxiliary_loss_clip": 0.01149085,
      "auxiliary_loss_mlp": 0.01029726,
      "balance_loss_clip": 1.04638386,
      "balance_loss_mlp": 1.02237332,
      "epoch": 0.8095953826729995,
      "flos": 24061083304320.0,
      "grad_norm": 2.167976816476336,
      "language_loss": 0.72803485,
      "learning_rate": 3.6825064596452646e-07,
      "loss": 0.74982303,
      "num_input_tokens_seen": 145302220,
      "step": 6733,
      "time_per_iteration": 2.530694007873535
    },
    {
      "auxiliary_loss_clip": 0.01151344,
      "auxiliary_loss_mlp": 0.0102875,
      "balance_loss_clip": 1.04611301,
      "balance_loss_mlp": 1.02226162,
      "epoch": 0.8097156255636385,
      "flos": 23951412103680.0,
      "grad_norm": 1.539155486737868,
      "language_loss": 0.70544779,
      "learning_rate": 3.678003459253305e-07,
      "loss": 0.72724867,
      "num_input_tokens_seen": 145323070,
      "step": 6734,
      "time_per_iteration": 2.5403409004211426
    },
    {
      "auxiliary_loss_clip": 0.01124446,
      "auxiliary_loss_mlp": 0.0102607,
      "balance_loss_clip": 1.0455519,
      "balance_loss_mlp": 1.01927519,
      "epoch": 0.8098358684542777,
      "flos": 21799142098560.0,
      "grad_norm": 2.491277754960323,
      "language_loss": 0.74140179,
      "learning_rate": 3.673502934895236e-07,
      "loss": 0.76290691,
      "num_input_tokens_seen": 145342575,
      "step": 6735,
      "time_per_iteration": 2.5551421642303467
    },
    {
      "auxiliary_loss_clip": 0.01055881,
      "auxiliary_loss_mlp": 0.01003393,
      "balance_loss_clip": 1.00841248,
      "balance_loss_mlp": 1.00235558,
      "epoch": 0.8099561113449167,
      "flos": 68809515966720.0,
      "grad_norm": 0.6887222967758427,
      "language_loss": 0.57960469,
      "learning_rate": 3.669004887253802e-07,
      "loss": 0.60019743,
      "num_input_tokens_seen": 145408865,
      "step": 6736,
      "time_per_iteration": 3.195160150527954
    },
    {
      "auxiliary_loss_clip": 0.01141629,
      "auxiliary_loss_mlp": 0.01030366,
      "balance_loss_clip": 1.0475893,
      "balance_loss_mlp": 1.0238421,
      "epoch": 0.8100763542355558,
      "flos": 23586056916480.0,
      "grad_norm": 1.7017866657285234,
      "language_loss": 0.78836453,
      "learning_rate": 3.664509317011335e-07,
      "loss": 0.81008446,
      "num_input_tokens_seen": 145429200,
      "step": 6737,
      "time_per_iteration": 2.5371413230895996
    },
    {
      "auxiliary_loss_clip": 0.01152692,
      "auxiliary_loss_mlp": 0.01028601,
      "balance_loss_clip": 1.04989457,
      "balance_loss_mlp": 1.02091825,
      "epoch": 0.810196597126195,
      "flos": 31650408207360.0,
      "grad_norm": 1.7709280712288527,
      "language_loss": 0.74110758,
      "learning_rate": 3.6600162248498134e-07,
      "loss": 0.7629205,
      "num_input_tokens_seen": 145452830,
      "step": 6738,
      "time_per_iteration": 2.567059278488159
    },
    {
      "auxiliary_loss_clip": 0.01078657,
      "auxiliary_loss_mlp": 0.01024554,
      "balance_loss_clip": 1.03702044,
      "balance_loss_mlp": 1.01806319,
      "epoch": 0.810316840016834,
      "flos": 24900459298560.0,
      "grad_norm": 1.8157210869276579,
      "language_loss": 0.75805569,
      "learning_rate": 3.6555256114508426e-07,
      "loss": 0.77908778,
      "num_input_tokens_seen": 145472625,
      "step": 6739,
      "time_per_iteration": 2.635758638381958
    },
    {
      "auxiliary_loss_clip": 0.01136741,
      "auxiliary_loss_mlp": 0.01029549,
      "balance_loss_clip": 1.04206276,
      "balance_loss_mlp": 1.02228606,
      "epoch": 0.8104370829074731,
      "flos": 27965003950080.0,
      "grad_norm": 2.0783908195107736,
      "language_loss": 0.72613406,
      "learning_rate": 3.651037477495642e-07,
      "loss": 0.74779689,
      "num_input_tokens_seen": 145494075,
      "step": 6740,
      "time_per_iteration": 2.575836181640625
    },
    {
      "auxiliary_loss_clip": 0.01163842,
      "auxiliary_loss_mlp": 0.01029328,
      "balance_loss_clip": 1.04745913,
      "balance_loss_mlp": 1.02207422,
      "epoch": 0.8105573257981122,
      "flos": 24640752988800.0,
      "grad_norm": 1.8101517450497882,
      "language_loss": 0.68040943,
      "learning_rate": 3.6465518236650584e-07,
      "loss": 0.70234114,
      "num_input_tokens_seen": 145514220,
      "step": 6741,
      "time_per_iteration": 4.061638832092285
    },
    {
      "auxiliary_loss_clip": 0.01123312,
      "auxiliary_loss_mlp": 0.01023713,
      "balance_loss_clip": 1.04241037,
      "balance_loss_mlp": 1.01691473,
      "epoch": 0.8106775686887513,
      "flos": 26358935132160.0,
      "grad_norm": 1.5725752626689806,
      "language_loss": 0.7821238,
      "learning_rate": 3.642068650639558e-07,
      "loss": 0.80359405,
      "num_input_tokens_seen": 145533965,
      "step": 6742,
      "time_per_iteration": 2.5929312705993652
    },
    {
      "auxiliary_loss_clip": 0.01128984,
      "auxiliary_loss_mlp": 0.01024416,
      "balance_loss_clip": 1.04025817,
      "balance_loss_mlp": 1.0175916,
      "epoch": 0.8107978115793903,
      "flos": 27271892136960.0,
      "grad_norm": 1.6979431590537866,
      "language_loss": 0.64317226,
      "learning_rate": 3.6375879590992334e-07,
      "loss": 0.66470623,
      "num_input_tokens_seen": 145554310,
      "step": 6743,
      "time_per_iteration": 2.56786847114563
    },
    {
      "auxiliary_loss_clip": 0.0113379,
      "auxiliary_loss_mlp": 0.01024721,
      "balance_loss_clip": 1.04592228,
      "balance_loss_mlp": 1.01730299,
      "epoch": 0.8109180544700295,
      "flos": 24934322845440.0,
      "grad_norm": 1.799478112163573,
      "language_loss": 0.81111926,
      "learning_rate": 3.6331097497238173e-07,
      "loss": 0.83270431,
      "num_input_tokens_seen": 145573755,
      "step": 6744,
      "time_per_iteration": 2.538419008255005
    },
    {
      "auxiliary_loss_clip": 0.01123682,
      "auxiliary_loss_mlp": 0.01026351,
      "balance_loss_clip": 1.04448152,
      "balance_loss_mlp": 1.02002132,
      "epoch": 0.8110382973606686,
      "flos": 21105383840640.0,
      "grad_norm": 1.9542343161377356,
      "language_loss": 0.79800558,
      "learning_rate": 3.628634023192627e-07,
      "loss": 0.81950593,
      "num_input_tokens_seen": 145594000,
      "step": 6745,
      "time_per_iteration": 2.580892324447632
    },
    {
      "auxiliary_loss_clip": 0.01152705,
      "auxiliary_loss_mlp": 0.01022049,
      "balance_loss_clip": 1.0462873,
      "balance_loss_mlp": 1.01482487,
      "epoch": 0.8111585402513076,
      "flos": 15414081500160.0,
      "grad_norm": 8.988291761890526,
      "language_loss": 0.75489938,
      "learning_rate": 3.624160780184644e-07,
      "loss": 0.77664691,
      "num_input_tokens_seen": 145611215,
      "step": 6746,
      "time_per_iteration": 2.4571259021759033
    },
    {
      "auxiliary_loss_clip": 0.01133144,
      "auxiliary_loss_mlp": 0.01025759,
      "balance_loss_clip": 1.04535651,
      "balance_loss_mlp": 1.01826715,
      "epoch": 0.8112787831419467,
      "flos": 24095736950400.0,
      "grad_norm": 1.5834022826619456,
      "language_loss": 0.74227405,
      "learning_rate": 3.6196900213784496e-07,
      "loss": 0.76386315,
      "num_input_tokens_seen": 145630530,
      "step": 6747,
      "time_per_iteration": 2.542808771133423
    },
    {
      "auxiliary_loss_clip": 0.01151554,
      "auxiliary_loss_mlp": 0.01027883,
      "balance_loss_clip": 1.04640019,
      "balance_loss_mlp": 1.02112949,
      "epoch": 0.8113990260325858,
      "flos": 20483374999680.0,
      "grad_norm": 1.883831502598574,
      "language_loss": 0.86661965,
      "learning_rate": 3.6152217474522527e-07,
      "loss": 0.88841397,
      "num_input_tokens_seen": 145647345,
      "step": 6748,
      "time_per_iteration": 2.4763739109039307
    },
    {
      "auxiliary_loss_clip": 0.01152566,
      "auxiliary_loss_mlp": 0.01021474,
      "balance_loss_clip": 1.04987192,
      "balance_loss_mlp": 1.01476598,
      "epoch": 0.8115192689232249,
      "flos": 24901141656960.0,
      "grad_norm": 1.4925523394449782,
      "language_loss": 0.72677743,
      "learning_rate": 3.6107559590838975e-07,
      "loss": 0.74851781,
      "num_input_tokens_seen": 145666330,
      "step": 6749,
      "time_per_iteration": 2.533613681793213
    },
    {
      "auxiliary_loss_clip": 0.01087644,
      "auxiliary_loss_mlp": 0.01031666,
      "balance_loss_clip": 1.03838658,
      "balance_loss_mlp": 1.02435815,
      "epoch": 0.811639511813864,
      "flos": 24057204635520.0,
      "grad_norm": 2.467950668594518,
      "language_loss": 0.66181672,
      "learning_rate": 3.606292656950822e-07,
      "loss": 0.68300986,
      "num_input_tokens_seen": 145684740,
      "step": 6750,
      "time_per_iteration": 2.625168561935425
    },
    {
      "auxiliary_loss_clip": 0.0112998,
      "auxiliary_loss_mlp": 0.01029176,
      "balance_loss_clip": 1.04189444,
      "balance_loss_mlp": 1.02159762,
      "epoch": 0.8117597547045031,
      "flos": 23185150243200.0,
      "grad_norm": 3.4462262494780127,
      "language_loss": 0.86629862,
      "learning_rate": 3.601831841730121e-07,
      "loss": 0.88789022,
      "num_input_tokens_seen": 145702660,
      "step": 6751,
      "time_per_iteration": 2.515833854675293
    },
    {
      "auxiliary_loss_clip": 0.01150318,
      "auxiliary_loss_mlp": 0.01025291,
      "balance_loss_clip": 1.04660249,
      "balance_loss_mlp": 1.01807272,
      "epoch": 0.8118799975951422,
      "flos": 23040250778880.0,
      "grad_norm": 1.581954657765605,
      "language_loss": 0.72588289,
      "learning_rate": 3.5973735140984916e-07,
      "loss": 0.74763894,
      "num_input_tokens_seen": 145722830,
      "step": 6752,
      "time_per_iteration": 2.5380849838256836
    },
    {
      "auxiliary_loss_clip": 0.01104501,
      "auxiliary_loss_mlp": 0.00760812,
      "balance_loss_clip": 1.04093456,
      "balance_loss_mlp": 1.00024033,
      "epoch": 0.8120002404857812,
      "flos": 24639962889600.0,
      "grad_norm": 1.8245109382896405,
      "language_loss": 0.79391062,
      "learning_rate": 3.5929176747322607e-07,
      "loss": 0.81256378,
      "num_input_tokens_seen": 145741935,
      "step": 6753,
      "time_per_iteration": 2.6144087314605713
    },
    {
      "auxiliary_loss_clip": 0.01037767,
      "auxiliary_loss_mlp": 0.01000695,
      "balance_loss_clip": 1.00821829,
      "balance_loss_mlp": 0.99963969,
      "epoch": 0.8121204833764204,
      "flos": 57415742156160.0,
      "grad_norm": 0.818770455586398,
      "language_loss": 0.56304932,
      "learning_rate": 3.588464324307372e-07,
      "loss": 0.58343399,
      "num_input_tokens_seen": 145805560,
      "step": 6754,
      "time_per_iteration": 3.16375732421875
    },
    {
      "auxiliary_loss_clip": 0.01151771,
      "auxiliary_loss_mlp": 0.01028759,
      "balance_loss_clip": 1.0447278,
      "balance_loss_mlp": 1.02141309,
      "epoch": 0.8122407262670595,
      "flos": 19464589549440.0,
      "grad_norm": 1.7051218188005042,
      "language_loss": 0.75398338,
      "learning_rate": 3.584013463499391e-07,
      "loss": 0.77578866,
      "num_input_tokens_seen": 145824180,
      "step": 6755,
      "time_per_iteration": 2.48321795463562
    },
    {
      "auxiliary_loss_clip": 0.01039315,
      "auxiliary_loss_mlp": 0.00999642,
      "balance_loss_clip": 1.01176035,
      "balance_loss_mlp": 0.99853289,
      "epoch": 0.8123609691576985,
      "flos": 56425325472000.0,
      "grad_norm": 0.7378537864920456,
      "language_loss": 0.6448555,
      "learning_rate": 3.579565092983521e-07,
      "loss": 0.66524506,
      "num_input_tokens_seen": 145885300,
      "step": 6756,
      "time_per_iteration": 3.781142473220825
    },
    {
      "auxiliary_loss_clip": 0.0116515,
      "auxiliary_loss_mlp": 0.01033846,
      "balance_loss_clip": 1.04942262,
      "balance_loss_mlp": 1.02675903,
      "epoch": 0.8124812120483377,
      "flos": 20631973564800.0,
      "grad_norm": 1.9962546057621966,
      "language_loss": 0.83676046,
      "learning_rate": 3.575119213434565e-07,
      "loss": 0.85875034,
      "num_input_tokens_seen": 145903815,
      "step": 6757,
      "time_per_iteration": 2.4570813179016113
    },
    {
      "auxiliary_loss_clip": 0.01151727,
      "auxiliary_loss_mlp": 0.0102495,
      "balance_loss_clip": 1.04844308,
      "balance_loss_mlp": 1.01777399,
      "epoch": 0.8126014549389767,
      "flos": 22492397566080.0,
      "grad_norm": 1.7695805055040827,
      "language_loss": 0.81749421,
      "learning_rate": 3.5706758255269765e-07,
      "loss": 0.83926105,
      "num_input_tokens_seen": 145922270,
      "step": 6758,
      "time_per_iteration": 2.4993133544921875
    },
    {
      "auxiliary_loss_clip": 0.01142559,
      "auxiliary_loss_mlp": 0.01031767,
      "balance_loss_clip": 1.04732573,
      "balance_loss_mlp": 1.02483797,
      "epoch": 0.8127216978296158,
      "flos": 23287961946240.0,
      "grad_norm": 2.0578144999559935,
      "language_loss": 0.69767725,
      "learning_rate": 3.566234929934795e-07,
      "loss": 0.71942049,
      "num_input_tokens_seen": 145941470,
      "step": 6759,
      "time_per_iteration": 2.5252883434295654
    },
    {
      "auxiliary_loss_clip": 0.01147126,
      "auxiliary_loss_mlp": 0.01030459,
      "balance_loss_clip": 1.04710674,
      "balance_loss_mlp": 1.02359581,
      "epoch": 0.812841940720255,
      "flos": 25154994049920.0,
      "grad_norm": 1.596941612907656,
      "language_loss": 0.71791756,
      "learning_rate": 3.561796527331706e-07,
      "loss": 0.7396934,
      "num_input_tokens_seen": 145963145,
      "step": 6760,
      "time_per_iteration": 2.534383773803711
    },
    {
      "auxiliary_loss_clip": 0.01123581,
      "auxiliary_loss_mlp": 0.0102466,
      "balance_loss_clip": 1.04295051,
      "balance_loss_mlp": 1.01790082,
      "epoch": 0.812962183610894,
      "flos": 26648446752000.0,
      "grad_norm": 1.891489208631019,
      "language_loss": 0.77819705,
      "learning_rate": 3.5573606183910163e-07,
      "loss": 0.79967952,
      "num_input_tokens_seen": 145983150,
      "step": 6761,
      "time_per_iteration": 2.5945050716400146
    },
    {
      "auxiliary_loss_clip": 0.0115632,
      "auxiliary_loss_mlp": 0.0102808,
      "balance_loss_clip": 1.04561031,
      "balance_loss_mlp": 1.0207572,
      "epoch": 0.8130824265015331,
      "flos": 24966965329920.0,
      "grad_norm": 1.7507876507399072,
      "language_loss": 0.78691852,
      "learning_rate": 3.5529272037856493e-07,
      "loss": 0.80876249,
      "num_input_tokens_seen": 146001365,
      "step": 6762,
      "time_per_iteration": 2.516690492630005
    },
    {
      "auxiliary_loss_clip": 0.0101233,
      "auxiliary_loss_mlp": 0.01000116,
      "balance_loss_clip": 1.00964868,
      "balance_loss_mlp": 0.9989835,
      "epoch": 0.8132026693921722,
      "flos": 67622918175360.0,
      "grad_norm": 0.709413191553818,
      "language_loss": 0.53910118,
      "learning_rate": 3.548496284188149e-07,
      "loss": 0.55922568,
      "num_input_tokens_seen": 146061570,
      "step": 6763,
      "time_per_iteration": 3.2818145751953125
    },
    {
      "auxiliary_loss_clip": 0.01104555,
      "auxiliary_loss_mlp": 0.01026683,
      "balance_loss_clip": 1.04551721,
      "balance_loss_mlp": 1.02014422,
      "epoch": 0.8133229122828113,
      "flos": 19495149045120.0,
      "grad_norm": 5.560505104844581,
      "language_loss": 0.79016185,
      "learning_rate": 3.544067860270681e-07,
      "loss": 0.81147426,
      "num_input_tokens_seen": 146079145,
      "step": 6764,
      "time_per_iteration": 2.719434976577759
    },
    {
      "auxiliary_loss_clip": 0.01124448,
      "auxiliary_loss_mlp": 0.01027776,
      "balance_loss_clip": 1.04329753,
      "balance_loss_mlp": 1.02047718,
      "epoch": 0.8134431551734503,
      "flos": 20668135582080.0,
      "grad_norm": 1.5919296591999543,
      "language_loss": 0.70750415,
      "learning_rate": 3.539641932705029e-07,
      "loss": 0.72902638,
      "num_input_tokens_seen": 146097625,
      "step": 6765,
      "time_per_iteration": 2.5440261363983154
    },
    {
      "auxiliary_loss_clip": 0.01167195,
      "auxiliary_loss_mlp": 0.01026836,
      "balance_loss_clip": 1.04766536,
      "balance_loss_mlp": 1.0191884,
      "epoch": 0.8135633980640895,
      "flos": 21507332008320.0,
      "grad_norm": 2.1650447136805737,
      "language_loss": 0.77138084,
      "learning_rate": 3.53521850216262e-07,
      "loss": 0.79332113,
      "num_input_tokens_seen": 146117195,
      "step": 6766,
      "time_per_iteration": 2.4680590629577637
    },
    {
      "auxiliary_loss_clip": 0.01165743,
      "auxiliary_loss_mlp": 0.01033154,
      "balance_loss_clip": 1.04956222,
      "balance_loss_mlp": 1.02561438,
      "epoch": 0.8136836409547286,
      "flos": 20554442058240.0,
      "grad_norm": 2.344517666628521,
      "language_loss": 0.76946139,
      "learning_rate": 3.530797569314461e-07,
      "loss": 0.79145038,
      "num_input_tokens_seen": 146136220,
      "step": 6767,
      "time_per_iteration": 3.6066341400146484
    },
    {
      "auxiliary_loss_clip": 0.01164379,
      "auxiliary_loss_mlp": 0.01031328,
      "balance_loss_clip": 1.04797709,
      "balance_loss_mlp": 1.0240978,
      "epoch": 0.8138038838453676,
      "flos": 20299045380480.0,
      "grad_norm": 1.7332947765378452,
      "language_loss": 0.77709574,
      "learning_rate": 3.5263791348312235e-07,
      "loss": 0.79905283,
      "num_input_tokens_seen": 146155415,
      "step": 6768,
      "time_per_iteration": 3.20334792137146
    },
    {
      "auxiliary_loss_clip": 0.01134449,
      "auxiliary_loss_mlp": 0.01027207,
      "balance_loss_clip": 1.04436159,
      "balance_loss_mlp": 1.01991153,
      "epoch": 0.8139241267360068,
      "flos": 29789840551680.0,
      "grad_norm": 1.7127113297632148,
      "language_loss": 0.70757997,
      "learning_rate": 3.521963199383171e-07,
      "loss": 0.72919655,
      "num_input_tokens_seen": 146178370,
      "step": 6769,
      "time_per_iteration": 2.6040656566619873
    },
    {
      "auxiliary_loss_clip": 0.01109011,
      "auxiliary_loss_mlp": 0.01026551,
      "balance_loss_clip": 1.04203653,
      "balance_loss_mlp": 1.01852834,
      "epoch": 0.8140443696266458,
      "flos": 19713270384000.0,
      "grad_norm": 2.0914741343791996,
      "language_loss": 0.76594824,
      "learning_rate": 3.517549763640197e-07,
      "loss": 0.78730381,
      "num_input_tokens_seen": 146196010,
      "step": 6770,
      "time_per_iteration": 2.616095542907715
    },
    {
      "auxiliary_loss_clip": 0.01149709,
      "auxiliary_loss_mlp": 0.00760684,
      "balance_loss_clip": 1.04857063,
      "balance_loss_mlp": 1.00026846,
      "epoch": 0.8141646125172849,
      "flos": 27160568910720.0,
      "grad_norm": 1.716762393948832,
      "language_loss": 0.7085669,
      "learning_rate": 3.513138828271829e-07,
      "loss": 0.72767091,
      "num_input_tokens_seen": 146215880,
      "step": 6771,
      "time_per_iteration": 2.5563526153564453
    },
    {
      "auxiliary_loss_clip": 0.01117001,
      "auxiliary_loss_mlp": 0.01024273,
      "balance_loss_clip": 1.04289639,
      "balance_loss_mlp": 1.01793098,
      "epoch": 0.8142848554079241,
      "flos": 39673102700160.0,
      "grad_norm": 2.8984066415245184,
      "language_loss": 0.69872916,
      "learning_rate": 3.508730393947179e-07,
      "loss": 0.72014195,
      "num_input_tokens_seen": 146239135,
      "step": 6772,
      "time_per_iteration": 2.6939961910247803
    },
    {
      "auxiliary_loss_clip": 0.01125326,
      "auxiliary_loss_mlp": 0.01023774,
      "balance_loss_clip": 1.04822803,
      "balance_loss_mlp": 1.01735437,
      "epoch": 0.8144050982985631,
      "flos": 22237288197120.0,
      "grad_norm": 1.7386539981088658,
      "language_loss": 0.71851611,
      "learning_rate": 3.504324461335024e-07,
      "loss": 0.74000704,
      "num_input_tokens_seen": 146259245,
      "step": 6773,
      "time_per_iteration": 2.564802885055542
    },
    {
      "auxiliary_loss_clip": 0.01099527,
      "auxiliary_loss_mlp": 0.01026524,
      "balance_loss_clip": 1.0393703,
      "balance_loss_mlp": 1.0192579,
      "epoch": 0.8145253411892022,
      "flos": 23038239617280.0,
      "grad_norm": 2.191220537628639,
      "language_loss": 0.88013035,
      "learning_rate": 3.499921031103732e-07,
      "loss": 0.90139091,
      "num_input_tokens_seen": 146280015,
      "step": 6774,
      "time_per_iteration": 2.6176445484161377
    },
    {
      "auxiliary_loss_clip": 0.01127747,
      "auxiliary_loss_mlp": 0.01023461,
      "balance_loss_clip": 1.04220748,
      "balance_loss_mlp": 1.01640129,
      "epoch": 0.8146455840798413,
      "flos": 24827668387200.0,
      "grad_norm": 1.630796501862283,
      "language_loss": 0.7833156,
      "learning_rate": 3.4955201039212987e-07,
      "loss": 0.80482769,
      "num_input_tokens_seen": 146300935,
      "step": 6775,
      "time_per_iteration": 2.5870580673217773
    },
    {
      "auxiliary_loss_clip": 0.01151766,
      "auxiliary_loss_mlp": 0.01032086,
      "balance_loss_clip": 1.04703748,
      "balance_loss_mlp": 1.02486753,
      "epoch": 0.8147658269704804,
      "flos": 19974520978560.0,
      "grad_norm": 2.083598147556066,
      "language_loss": 0.65456808,
      "learning_rate": 3.4911216804553465e-07,
      "loss": 0.67640662,
      "num_input_tokens_seen": 146319835,
      "step": 6776,
      "time_per_iteration": 2.4784457683563232
    },
    {
      "auxiliary_loss_clip": 0.01136192,
      "auxiliary_loss_mlp": 0.01025903,
      "balance_loss_clip": 1.04501462,
      "balance_loss_mlp": 1.01852703,
      "epoch": 0.8148860698611194,
      "flos": 21178031097600.0,
      "grad_norm": 1.840022980978433,
      "language_loss": 0.70380008,
      "learning_rate": 3.4867257613731017e-07,
      "loss": 0.72542107,
      "num_input_tokens_seen": 146339030,
      "step": 6777,
      "time_per_iteration": 2.520625591278076
    },
    {
      "auxiliary_loss_clip": 0.01139684,
      "auxiliary_loss_mlp": 0.01027055,
      "balance_loss_clip": 1.0448817,
      "balance_loss_mlp": 1.02023363,
      "epoch": 0.8150063127517585,
      "flos": 19606903234560.0,
      "grad_norm": 1.9377258407557598,
      "language_loss": 0.8550638,
      "learning_rate": 3.4823323473414343e-07,
      "loss": 0.87673122,
      "num_input_tokens_seen": 146358550,
      "step": 6778,
      "time_per_iteration": 2.50909686088562
    },
    {
      "auxiliary_loss_clip": 0.01125726,
      "auxiliary_loss_mlp": 0.01028351,
      "balance_loss_clip": 1.04460657,
      "balance_loss_mlp": 1.02048957,
      "epoch": 0.8151265556423977,
      "flos": 22638374438400.0,
      "grad_norm": 2.068153244170498,
      "language_loss": 0.76184547,
      "learning_rate": 3.477941439026812e-07,
      "loss": 0.78338623,
      "num_input_tokens_seen": 146376770,
      "step": 6779,
      "time_per_iteration": 2.5577714443206787
    },
    {
      "auxiliary_loss_clip": 0.01135389,
      "auxiliary_loss_mlp": 0.01030166,
      "balance_loss_clip": 1.04782701,
      "balance_loss_mlp": 1.02332294,
      "epoch": 0.8152467985330367,
      "flos": 17968048277760.0,
      "grad_norm": 1.775856162213348,
      "language_loss": 0.73086643,
      "learning_rate": 3.473553037095349e-07,
      "loss": 0.75252193,
      "num_input_tokens_seen": 146395795,
      "step": 6780,
      "time_per_iteration": 2.489179849624634
    },
    {
      "auxiliary_loss_clip": 0.01131427,
      "auxiliary_loss_mlp": 0.01032215,
      "balance_loss_clip": 1.0441339,
      "balance_loss_mlp": 1.02540481,
      "epoch": 0.8153670414236758,
      "flos": 24969012405120.0,
      "grad_norm": 1.689369790478916,
      "language_loss": 0.83334225,
      "learning_rate": 3.469167142212743e-07,
      "loss": 0.85497868,
      "num_input_tokens_seen": 146417640,
      "step": 6781,
      "time_per_iteration": 2.5664689540863037
    },
    {
      "auxiliary_loss_clip": 0.01152065,
      "auxiliary_loss_mlp": 0.01030564,
      "balance_loss_clip": 1.04850936,
      "balance_loss_mlp": 1.02345026,
      "epoch": 0.8154872843143149,
      "flos": 31066069754880.0,
      "grad_norm": 2.1672605136354846,
      "language_loss": 0.63251305,
      "learning_rate": 3.4647837550443337e-07,
      "loss": 0.65433931,
      "num_input_tokens_seen": 146436205,
      "step": 6782,
      "time_per_iteration": 3.3715970516204834
    },
    {
      "auxiliary_loss_clip": 0.01124221,
      "auxiliary_loss_mlp": 0.01025467,
      "balance_loss_clip": 1.04431021,
      "balance_loss_mlp": 1.01814795,
      "epoch": 0.815607527204954,
      "flos": 19391654983680.0,
      "grad_norm": 2.4830154440441685,
      "language_loss": 0.74504256,
      "learning_rate": 3.460402876255086e-07,
      "loss": 0.76653945,
      "num_input_tokens_seen": 146453595,
      "step": 6783,
      "time_per_iteration": 2.536693811416626
    },
    {
      "auxiliary_loss_clip": 0.0115476,
      "auxiliary_loss_mlp": 0.01028891,
      "balance_loss_clip": 1.04780483,
      "balance_loss_mlp": 1.02202487,
      "epoch": 0.815727770095593,
      "flos": 26140418743680.0,
      "grad_norm": 2.5239171649958467,
      "language_loss": 0.7185452,
      "learning_rate": 3.456024506509574e-07,
      "loss": 0.74038172,
      "num_input_tokens_seen": 146474515,
      "step": 6784,
      "time_per_iteration": 2.5285286903381348
    },
    {
      "auxiliary_loss_clip": 0.01150049,
      "auxiliary_loss_mlp": 0.00760824,
      "balance_loss_clip": 1.04705393,
      "balance_loss_mlp": 1.00023758,
      "epoch": 0.8158480129862322,
      "flos": 25337527989120.0,
      "grad_norm": 1.505612895754183,
      "language_loss": 0.73600054,
      "learning_rate": 3.4516486464719873e-07,
      "loss": 0.75510925,
      "num_input_tokens_seen": 146493905,
      "step": 6785,
      "time_per_iteration": 2.534344434738159
    },
    {
      "auxiliary_loss_clip": 0.01104304,
      "auxiliary_loss_mlp": 0.01029443,
      "balance_loss_clip": 1.04094911,
      "balance_loss_mlp": 1.02235007,
      "epoch": 0.8159682558768713,
      "flos": 34423645559040.0,
      "grad_norm": 1.792305967465305,
      "language_loss": 0.62114483,
      "learning_rate": 3.4472752968061445e-07,
      "loss": 0.64248228,
      "num_input_tokens_seen": 146518335,
      "step": 6786,
      "time_per_iteration": 2.70717453956604
    },
    {
      "auxiliary_loss_clip": 0.01153084,
      "auxiliary_loss_mlp": 0.01021324,
      "balance_loss_clip": 1.04647362,
      "balance_loss_mlp": 1.01423097,
      "epoch": 0.8160884987675103,
      "flos": 18653223185280.0,
      "grad_norm": 1.799602755052315,
      "language_loss": 0.73584104,
      "learning_rate": 3.442904458175475e-07,
      "loss": 0.75758517,
      "num_input_tokens_seen": 146535655,
      "step": 6787,
      "time_per_iteration": 2.4612183570861816
    },
    {
      "auxiliary_loss_clip": 0.01147999,
      "auxiliary_loss_mlp": 0.0102956,
      "balance_loss_clip": 1.04408431,
      "balance_loss_mlp": 1.02189755,
      "epoch": 0.8162087416581495,
      "flos": 31430527102080.0,
      "grad_norm": 1.4950397508550584,
      "language_loss": 0.76229119,
      "learning_rate": 3.438536131243044e-07,
      "loss": 0.7840668,
      "num_input_tokens_seen": 146556815,
      "step": 6788,
      "time_per_iteration": 2.576498508453369
    },
    {
      "auxiliary_loss_clip": 0.01141827,
      "auxiliary_loss_mlp": 0.010286,
      "balance_loss_clip": 1.04599166,
      "balance_loss_mlp": 1.02104235,
      "epoch": 0.8163289845487885,
      "flos": 37593910915200.0,
      "grad_norm": 2.2249442790910683,
      "language_loss": 0.62193418,
      "learning_rate": 3.434170316671503e-07,
      "loss": 0.64363849,
      "num_input_tokens_seen": 146581845,
      "step": 6789,
      "time_per_iteration": 2.6674773693084717
    },
    {
      "auxiliary_loss_clip": 0.01121033,
      "auxiliary_loss_mlp": 0.01024138,
      "balance_loss_clip": 1.0479759,
      "balance_loss_mlp": 1.01733065,
      "epoch": 0.8164492274394276,
      "flos": 13953989554560.0,
      "grad_norm": 2.243991278622339,
      "language_loss": 0.89954734,
      "learning_rate": 3.4298070151231583e-07,
      "loss": 0.92099905,
      "num_input_tokens_seen": 146597245,
      "step": 6790,
      "time_per_iteration": 2.5472185611724854
    },
    {
      "auxiliary_loss_clip": 0.01142311,
      "auxiliary_loss_mlp": 0.01029438,
      "balance_loss_clip": 1.04588342,
      "balance_loss_mlp": 1.02272677,
      "epoch": 0.8165694703300668,
      "flos": 28986554747520.0,
      "grad_norm": 1.802792638527892,
      "language_loss": 0.59682214,
      "learning_rate": 3.425446227259916e-07,
      "loss": 0.61853969,
      "num_input_tokens_seen": 146618210,
      "step": 6791,
      "time_per_iteration": 2.571389675140381
    },
    {
      "auxiliary_loss_clip": 0.01137973,
      "auxiliary_loss_mlp": 0.01026695,
      "balance_loss_clip": 1.04461789,
      "balance_loss_mlp": 1.02038026,
      "epoch": 0.8166897132207058,
      "flos": 25118365155840.0,
      "grad_norm": 1.8719098817674136,
      "language_loss": 0.82111728,
      "learning_rate": 3.421087953743296e-07,
      "loss": 0.84276396,
      "num_input_tokens_seen": 146637975,
      "step": 6792,
      "time_per_iteration": 2.563957691192627
    },
    {
      "auxiliary_loss_clip": 0.01150136,
      "auxiliary_loss_mlp": 0.01025388,
      "balance_loss_clip": 1.04419184,
      "balance_loss_mlp": 1.01784849,
      "epoch": 0.8168099561113449,
      "flos": 23148593176320.0,
      "grad_norm": 2.1546609669312846,
      "language_loss": 0.80113047,
      "learning_rate": 3.416732195234464e-07,
      "loss": 0.82288575,
      "num_input_tokens_seen": 146658030,
      "step": 6793,
      "time_per_iteration": 4.050412893295288
    },
    {
      "auxiliary_loss_clip": 0.01152731,
      "auxiliary_loss_mlp": 0.01031242,
      "balance_loss_clip": 1.04563594,
      "balance_loss_mlp": 1.0241518,
      "epoch": 0.816930199001984,
      "flos": 18407666833920.0,
      "grad_norm": 1.496879881042609,
      "language_loss": 0.79234207,
      "learning_rate": 3.4123789523941613e-07,
      "loss": 0.8141818,
      "num_input_tokens_seen": 146677855,
      "step": 6794,
      "time_per_iteration": 2.4817593097686768
    },
    {
      "auxiliary_loss_clip": 0.01141613,
      "auxiliary_loss_mlp": 0.01030434,
      "balance_loss_clip": 1.04194415,
      "balance_loss_mlp": 1.02294755,
      "epoch": 0.8170504418926231,
      "flos": 21251324799360.0,
      "grad_norm": 1.5346526064888462,
      "language_loss": 0.63283229,
      "learning_rate": 3.4080282258827884e-07,
      "loss": 0.65455282,
      "num_input_tokens_seen": 146696230,
      "step": 6795,
      "time_per_iteration": 2.4972455501556396
    },
    {
      "auxiliary_loss_clip": 0.0115562,
      "auxiliary_loss_mlp": 0.01025782,
      "balance_loss_clip": 1.0471822,
      "balance_loss_mlp": 1.01857579,
      "epoch": 0.8171706847832622,
      "flos": 19099234362240.0,
      "grad_norm": 2.06796628126924,
      "language_loss": 0.72356987,
      "learning_rate": 3.403680016360342e-07,
      "loss": 0.74538392,
      "num_input_tokens_seen": 146714835,
      "step": 6796,
      "time_per_iteration": 2.484015703201294
    },
    {
      "auxiliary_loss_clip": 0.0114594,
      "auxiliary_loss_mlp": 0.01024617,
      "balance_loss_clip": 1.04778814,
      "balance_loss_mlp": 1.01761591,
      "epoch": 0.8172909276739013,
      "flos": 21470128496640.0,
      "grad_norm": 1.4579927137009172,
      "language_loss": 0.67463434,
      "learning_rate": 3.3993343244864403e-07,
      "loss": 0.69633985,
      "num_input_tokens_seen": 146734425,
      "step": 6797,
      "time_per_iteration": 2.48726487159729
    },
    {
      "auxiliary_loss_clip": 0.01149933,
      "auxiliary_loss_mlp": 0.01027877,
      "balance_loss_clip": 1.04656386,
      "balance_loss_mlp": 1.02130008,
      "epoch": 0.8174111705645404,
      "flos": 27599792417280.0,
      "grad_norm": 1.4730682539015059,
      "language_loss": 0.72492278,
      "learning_rate": 3.394991150920323e-07,
      "loss": 0.74670088,
      "num_input_tokens_seen": 146757545,
      "step": 6798,
      "time_per_iteration": 2.547252893447876
    },
    {
      "auxiliary_loss_clip": 0.01112818,
      "auxiliary_loss_mlp": 0.00761696,
      "balance_loss_clip": 1.04452348,
      "balance_loss_mlp": 1.00028348,
      "epoch": 0.8175314134551794,
      "flos": 14064594508800.0,
      "grad_norm": 2.055649009633026,
      "language_loss": 0.74143159,
      "learning_rate": 3.3906504963208396e-07,
      "loss": 0.76017678,
      "num_input_tokens_seen": 146774240,
      "step": 6799,
      "time_per_iteration": 2.5501158237457275
    },
    {
      "auxiliary_loss_clip": 0.01105179,
      "auxiliary_loss_mlp": 0.01025162,
      "balance_loss_clip": 1.04329848,
      "balance_loss_mlp": 1.01783991,
      "epoch": 0.8176516563458186,
      "flos": 22708076780160.0,
      "grad_norm": 1.8999517073164542,
      "language_loss": 0.6653204,
      "learning_rate": 3.3863123613464774e-07,
      "loss": 0.68662381,
      "num_input_tokens_seen": 146793140,
      "step": 6800,
      "time_per_iteration": 2.5917744636535645
    },
    {
      "auxiliary_loss_clip": 0.01134959,
      "auxiliary_loss_mlp": 0.01021384,
      "balance_loss_clip": 1.03994441,
      "balance_loss_mlp": 1.01457071,
      "epoch": 0.8177718992364577,
      "flos": 21945406279680.0,
      "grad_norm": 1.5819923857465883,
      "language_loss": 0.75060475,
      "learning_rate": 3.381976746655317e-07,
      "loss": 0.77216816,
      "num_input_tokens_seen": 146812895,
      "step": 6801,
      "time_per_iteration": 2.539149522781372
    },
    {
      "auxiliary_loss_clip": 0.01104498,
      "auxiliary_loss_mlp": 0.01030434,
      "balance_loss_clip": 1.04713869,
      "balance_loss_mlp": 1.02336168,
      "epoch": 0.8178921421270967,
      "flos": 22017443005440.0,
      "grad_norm": 1.8890408002682166,
      "language_loss": 0.66964477,
      "learning_rate": 3.3776436529050756e-07,
      "loss": 0.69099414,
      "num_input_tokens_seen": 146832445,
      "step": 6802,
      "time_per_iteration": 2.5756068229675293
    },
    {
      "auxiliary_loss_clip": 0.01162288,
      "auxiliary_loss_mlp": 0.01025139,
      "balance_loss_clip": 1.04683077,
      "balance_loss_mlp": 1.01823068,
      "epoch": 0.8180123850177359,
      "flos": 33183111496320.0,
      "grad_norm": 1.6717131366120992,
      "language_loss": 0.72862494,
      "learning_rate": 3.373313080753073e-07,
      "loss": 0.75049925,
      "num_input_tokens_seen": 146856505,
      "step": 6803,
      "time_per_iteration": 2.5665318965911865
    },
    {
      "auxiliary_loss_clip": 0.01144085,
      "auxiliary_loss_mlp": 0.01023697,
      "balance_loss_clip": 1.04217649,
      "balance_loss_mlp": 1.01660669,
      "epoch": 0.8181326279083749,
      "flos": 22091167670400.0,
      "grad_norm": 1.734407486016436,
      "language_loss": 0.77299374,
      "learning_rate": 3.3689850308562527e-07,
      "loss": 0.79467154,
      "num_input_tokens_seen": 146876950,
      "step": 6804,
      "time_per_iteration": 2.5004937648773193
    },
    {
      "auxiliary_loss_clip": 0.01102951,
      "auxiliary_loss_mlp": 0.01024072,
      "balance_loss_clip": 1.04397893,
      "balance_loss_mlp": 1.01739669,
      "epoch": 0.818252870799014,
      "flos": 15705747936000.0,
      "grad_norm": 1.6824076174678317,
      "language_loss": 0.77604461,
      "learning_rate": 3.364659503871183e-07,
      "loss": 0.79731488,
      "num_input_tokens_seen": 146894885,
      "step": 6805,
      "time_per_iteration": 2.566399574279785
    },
    {
      "auxiliary_loss_clip": 0.0112044,
      "auxiliary_loss_mlp": 0.01020262,
      "balance_loss_clip": 1.04145527,
      "balance_loss_mlp": 1.01399767,
      "epoch": 0.8183731136896532,
      "flos": 18770687637120.0,
      "grad_norm": 1.8677862605192477,
      "language_loss": 0.83699358,
      "learning_rate": 3.3603365004540417e-07,
      "loss": 0.85840058,
      "num_input_tokens_seen": 146913180,
      "step": 6806,
      "time_per_iteration": 2.543564796447754
    },
    {
      "auxiliary_loss_clip": 0.01165511,
      "auxiliary_loss_mlp": 0.01030528,
      "balance_loss_clip": 1.04925919,
      "balance_loss_mlp": 1.02335215,
      "epoch": 0.8184933565802922,
      "flos": 26541792293760.0,
      "grad_norm": 2.008387377506147,
      "language_loss": 0.76614594,
      "learning_rate": 3.356016021260624e-07,
      "loss": 0.78810632,
      "num_input_tokens_seen": 146933510,
      "step": 6807,
      "time_per_iteration": 2.489863872528076
    },
    {
      "auxiliary_loss_clip": 0.01152968,
      "auxiliary_loss_mlp": 0.01027096,
      "balance_loss_clip": 1.04792619,
      "balance_loss_mlp": 1.0200274,
      "epoch": 0.8186135994709313,
      "flos": 17530117660800.0,
      "grad_norm": 2.3955410922460496,
      "language_loss": 0.65548855,
      "learning_rate": 3.35169806694634e-07,
      "loss": 0.67728925,
      "num_input_tokens_seen": 146951760,
      "step": 6808,
      "time_per_iteration": 3.266005277633667
    },
    {
      "auxiliary_loss_clip": 0.01022873,
      "auxiliary_loss_mlp": 0.0100213,
      "balance_loss_clip": 1.0093224,
      "balance_loss_mlp": 1.00111663,
      "epoch": 0.8187338423615703,
      "flos": 63480300675840.0,
      "grad_norm": 0.7170233531181492,
      "language_loss": 0.60645962,
      "learning_rate": 3.3473826381662186e-07,
      "loss": 0.62670958,
      "num_input_tokens_seen": 147022900,
      "step": 6809,
      "time_per_iteration": 3.279881000518799
    },
    {
      "auxiliary_loss_clip": 0.01146391,
      "auxiliary_loss_mlp": 0.01023215,
      "balance_loss_clip": 1.04767883,
      "balance_loss_mlp": 1.01613069,
      "epoch": 0.8188540852522095,
      "flos": 17529974006400.0,
      "grad_norm": 1.857789863573793,
      "language_loss": 0.81561327,
      "learning_rate": 3.3430697355749216e-07,
      "loss": 0.83730936,
      "num_input_tokens_seen": 147040590,
      "step": 6810,
      "time_per_iteration": 2.4627137184143066
    },
    {
      "auxiliary_loss_clip": 0.01105402,
      "auxiliary_loss_mlp": 0.01033205,
      "balance_loss_clip": 1.0410676,
      "balance_loss_mlp": 1.02570343,
      "epoch": 0.8189743281428485,
      "flos": 14392530702720.0,
      "grad_norm": 1.91207718907085,
      "language_loss": 0.7542488,
      "learning_rate": 3.3387593598266907e-07,
      "loss": 0.77563488,
      "num_input_tokens_seen": 147057200,
      "step": 6811,
      "time_per_iteration": 2.5692479610443115
    },
    {
      "auxiliary_loss_clip": 0.01114699,
      "auxiliary_loss_mlp": 0.01021324,
      "balance_loss_clip": 1.04039669,
      "balance_loss_mlp": 1.01421022,
      "epoch": 0.8190945710334876,
      "flos": 25080479285760.0,
      "grad_norm": 1.5540986087225956,
      "language_loss": 0.78187108,
      "learning_rate": 3.3344515115754225e-07,
      "loss": 0.8032313,
      "num_input_tokens_seen": 147076180,
      "step": 6812,
      "time_per_iteration": 2.5800628662109375
    },
    {
      "auxiliary_loss_clip": 0.01123109,
      "auxiliary_loss_mlp": 0.01025121,
      "balance_loss_clip": 1.04266214,
      "balance_loss_mlp": 1.01782572,
      "epoch": 0.8192148139241268,
      "flos": 21507152440320.0,
      "grad_norm": 3.1682263257073973,
      "language_loss": 0.80145115,
      "learning_rate": 3.33014619147461e-07,
      "loss": 0.82293344,
      "num_input_tokens_seen": 147094205,
      "step": 6813,
      "time_per_iteration": 2.547830581665039
    },
    {
      "auxiliary_loss_clip": 0.01140156,
      "auxiliary_loss_mlp": 0.0103017,
      "balance_loss_clip": 1.04951453,
      "balance_loss_mlp": 1.02248693,
      "epoch": 0.8193350568147658,
      "flos": 23952166289280.0,
      "grad_norm": 1.8780589548372724,
      "language_loss": 0.71717393,
      "learning_rate": 3.325843400177362e-07,
      "loss": 0.73887718,
      "num_input_tokens_seen": 147115545,
      "step": 6814,
      "time_per_iteration": 2.5305545330047607
    },
    {
      "auxiliary_loss_clip": 0.01153809,
      "auxiliary_loss_mlp": 0.00760936,
      "balance_loss_clip": 1.04655004,
      "balance_loss_mlp": 1.00028479,
      "epoch": 0.8194552997054049,
      "flos": 20559469962240.0,
      "grad_norm": 1.718044440466529,
      "language_loss": 0.73602092,
      "learning_rate": 3.32154313833642e-07,
      "loss": 0.75516832,
      "num_input_tokens_seen": 147135700,
      "step": 6815,
      "time_per_iteration": 2.5088226795196533
    },
    {
      "auxiliary_loss_clip": 0.01165435,
      "auxiliary_loss_mlp": 0.01028785,
      "balance_loss_clip": 1.0475446,
      "balance_loss_mlp": 1.02103364,
      "epoch": 0.819575542596044,
      "flos": 26031753123840.0,
      "grad_norm": 2.3459036276879646,
      "language_loss": 0.59545577,
      "learning_rate": 3.3172454066041164e-07,
      "loss": 0.6173979,
      "num_input_tokens_seen": 147155205,
      "step": 6816,
      "time_per_iteration": 2.5126585960388184
    },
    {
      "auxiliary_loss_clip": 0.01093438,
      "auxiliary_loss_mlp": 0.00760079,
      "balance_loss_clip": 1.0417769,
      "balance_loss_mlp": 1.00025511,
      "epoch": 0.8196957854866831,
      "flos": 29096944220160.0,
      "grad_norm": 1.9010772360846389,
      "language_loss": 0.76310009,
      "learning_rate": 3.3129502056324234e-07,
      "loss": 0.78163528,
      "num_input_tokens_seen": 147176570,
      "step": 6817,
      "time_per_iteration": 2.6811389923095703
    },
    {
      "auxiliary_loss_clip": 0.00998317,
      "auxiliary_loss_mlp": 0.01002152,
      "balance_loss_clip": 1.01183498,
      "balance_loss_mlp": 1.00093043,
      "epoch": 0.8198160283773221,
      "flos": 69033631898880.0,
      "grad_norm": 0.9788672172636785,
      "language_loss": 0.5976724,
      "learning_rate": 3.3086575360729165e-07,
      "loss": 0.61767709,
      "num_input_tokens_seen": 147234105,
      "step": 6818,
      "time_per_iteration": 3.158604383468628
    },
    {
      "auxiliary_loss_clip": 0.01135642,
      "auxiliary_loss_mlp": 0.01026591,
      "balance_loss_clip": 1.04582453,
      "balance_loss_mlp": 1.01892579,
      "epoch": 0.8199362712679613,
      "flos": 16618058496000.0,
      "grad_norm": 1.619071396711758,
      "language_loss": 0.71539617,
      "learning_rate": 3.3043673985767906e-07,
      "loss": 0.73701853,
      "num_input_tokens_seen": 147253170,
      "step": 6819,
      "time_per_iteration": 3.8724524974823
    },
    {
      "auxiliary_loss_clip": 0.01112507,
      "auxiliary_loss_mlp": 0.01028755,
      "balance_loss_clip": 1.04082072,
      "balance_loss_mlp": 1.02165031,
      "epoch": 0.8200565141586004,
      "flos": 21757664868480.0,
      "grad_norm": 1.733740093195914,
      "language_loss": 0.77560717,
      "learning_rate": 3.3000797937948564e-07,
      "loss": 0.79701972,
      "num_input_tokens_seen": 147271465,
      "step": 6820,
      "time_per_iteration": 3.3193514347076416
    },
    {
      "auxiliary_loss_clip": 0.01022646,
      "auxiliary_loss_mlp": 0.0100292,
      "balance_loss_clip": 1.00926697,
      "balance_loss_mlp": 1.00186503,
      "epoch": 0.8201767570492394,
      "flos": 69807112392960.0,
      "grad_norm": 0.9274597134168876,
      "language_loss": 0.64963251,
      "learning_rate": 3.295794722377534e-07,
      "loss": 0.66988814,
      "num_input_tokens_seen": 147335070,
      "step": 6821,
      "time_per_iteration": 3.179305076599121
    },
    {
      "auxiliary_loss_clip": 0.01159561,
      "auxiliary_loss_mlp": 0.01025235,
      "balance_loss_clip": 1.04500449,
      "balance_loss_mlp": 1.018345,
      "epoch": 0.8202969999398786,
      "flos": 23111892455040.0,
      "grad_norm": 1.4973075312463564,
      "language_loss": 0.79638612,
      "learning_rate": 3.291512184974876e-07,
      "loss": 0.81823409,
      "num_input_tokens_seen": 147355460,
      "step": 6822,
      "time_per_iteration": 2.475217819213867
    },
    {
      "auxiliary_loss_clip": 0.01133746,
      "auxiliary_loss_mlp": 0.01029674,
      "balance_loss_clip": 1.04197192,
      "balance_loss_mlp": 1.02231884,
      "epoch": 0.8204172428305176,
      "flos": 28220616109440.0,
      "grad_norm": 1.6497699352216888,
      "language_loss": 0.66492373,
      "learning_rate": 3.2872321822365346e-07,
      "loss": 0.68655795,
      "num_input_tokens_seen": 147375675,
      "step": 6823,
      "time_per_iteration": 2.581940174102783
    },
    {
      "auxiliary_loss_clip": 0.0114775,
      "auxiliary_loss_mlp": 0.01026508,
      "balance_loss_clip": 1.04593551,
      "balance_loss_mlp": 1.01954901,
      "epoch": 0.8205374857211567,
      "flos": 20887011106560.0,
      "grad_norm": 1.7508468158134722,
      "language_loss": 0.73225856,
      "learning_rate": 3.282954714811783e-07,
      "loss": 0.75400114,
      "num_input_tokens_seen": 147394580,
      "step": 6824,
      "time_per_iteration": 2.4890992641448975
    },
    {
      "auxiliary_loss_clip": 0.01123604,
      "auxiliary_loss_mlp": 0.01023972,
      "balance_loss_clip": 1.04060197,
      "balance_loss_mlp": 1.01701951,
      "epoch": 0.8206577286117959,
      "flos": 13152140294400.0,
      "grad_norm": 2.551101378042838,
      "language_loss": 0.7101748,
      "learning_rate": 3.2786797833495093e-07,
      "loss": 0.73165053,
      "num_input_tokens_seen": 147409935,
      "step": 6825,
      "time_per_iteration": 2.4846715927124023
    },
    {
      "auxiliary_loss_clip": 0.01162042,
      "auxiliary_loss_mlp": 0.01025972,
      "balance_loss_clip": 1.04708838,
      "balance_loss_mlp": 1.01902533,
      "epoch": 0.8207779715024349,
      "flos": 25265634917760.0,
      "grad_norm": 1.8982591425304003,
      "language_loss": 0.72883856,
      "learning_rate": 3.274407388498213e-07,
      "loss": 0.75071871,
      "num_input_tokens_seen": 147428065,
      "step": 6826,
      "time_per_iteration": 2.4966301918029785
    },
    {
      "auxiliary_loss_clip": 0.01116006,
      "auxiliary_loss_mlp": 0.01017874,
      "balance_loss_clip": 1.04152691,
      "balance_loss_mlp": 1.0108223,
      "epoch": 0.820898214393074,
      "flos": 19610243199360.0,
      "grad_norm": 1.7985357007481377,
      "language_loss": 0.7404412,
      "learning_rate": 3.270137530906021e-07,
      "loss": 0.76178002,
      "num_input_tokens_seen": 147447300,
      "step": 6827,
      "time_per_iteration": 2.577653408050537
    },
    {
      "auxiliary_loss_clip": 0.01101013,
      "auxiliary_loss_mlp": 0.01023981,
      "balance_loss_clip": 1.04293096,
      "balance_loss_mlp": 1.01749647,
      "epoch": 0.8210184572837131,
      "flos": 15596615439360.0,
      "grad_norm": 1.938158448934175,
      "language_loss": 0.83396029,
      "learning_rate": 3.265870211220665e-07,
      "loss": 0.85521019,
      "num_input_tokens_seen": 147465135,
      "step": 6828,
      "time_per_iteration": 2.5631890296936035
    },
    {
      "auxiliary_loss_clip": 0.01118219,
      "auxiliary_loss_mlp": 0.01031921,
      "balance_loss_clip": 1.04302478,
      "balance_loss_mlp": 1.02410626,
      "epoch": 0.8211387001743522,
      "flos": 20813932886400.0,
      "grad_norm": 2.579462168645033,
      "language_loss": 0.82113957,
      "learning_rate": 3.2616054300894934e-07,
      "loss": 0.842641,
      "num_input_tokens_seen": 147484585,
      "step": 6829,
      "time_per_iteration": 2.5804059505462646
    },
    {
      "auxiliary_loss_clip": 0.01124786,
      "auxiliary_loss_mlp": 0.01030319,
      "balance_loss_clip": 1.0433774,
      "balance_loss_mlp": 1.02311265,
      "epoch": 0.8212589430649913,
      "flos": 27704579368320.0,
      "grad_norm": 1.9695810141866232,
      "language_loss": 0.84545052,
      "learning_rate": 3.2573431881594693e-07,
      "loss": 0.86700153,
      "num_input_tokens_seen": 147504130,
      "step": 6830,
      "time_per_iteration": 2.5663044452667236
    },
    {
      "auxiliary_loss_clip": 0.01092955,
      "auxiliary_loss_mlp": 0.01021861,
      "balance_loss_clip": 1.0382359,
      "balance_loss_mlp": 1.01482153,
      "epoch": 0.8213791859556304,
      "flos": 22455625017600.0,
      "grad_norm": 1.9470781660651826,
      "language_loss": 0.65750802,
      "learning_rate": 3.2530834860771663e-07,
      "loss": 0.67865622,
      "num_input_tokens_seen": 147523510,
      "step": 6831,
      "time_per_iteration": 2.6149399280548096
    },
    {
      "auxiliary_loss_clip": 0.01150593,
      "auxiliary_loss_mlp": 0.01024935,
      "balance_loss_clip": 1.04559052,
      "balance_loss_mlp": 1.01780033,
      "epoch": 0.8214994288462695,
      "flos": 16654471908480.0,
      "grad_norm": 1.9227332179090004,
      "language_loss": 0.74230158,
      "learning_rate": 3.248826324488794e-07,
      "loss": 0.7640568,
      "num_input_tokens_seen": 147540805,
      "step": 6832,
      "time_per_iteration": 2.475402593612671
    },
    {
      "auxiliary_loss_clip": 0.01165325,
      "auxiliary_loss_mlp": 0.01023703,
      "balance_loss_clip": 1.0503155,
      "balance_loss_mlp": 1.0169853,
      "epoch": 0.8216196717369085,
      "flos": 25221787390080.0,
      "grad_norm": 1.7356946631938204,
      "language_loss": 0.87901926,
      "learning_rate": 3.244571704040138e-07,
      "loss": 0.90090954,
      "num_input_tokens_seen": 147560965,
      "step": 6833,
      "time_per_iteration": 2.4789998531341553
    },
    {
      "auxiliary_loss_clip": 0.0114732,
      "auxiliary_loss_mlp": 0.01026011,
      "balance_loss_clip": 1.04402959,
      "balance_loss_mlp": 1.01866829,
      "epoch": 0.8217399146275477,
      "flos": 25371930240000.0,
      "grad_norm": 2.5092900202712363,
      "language_loss": 0.73286933,
      "learning_rate": 3.2403196253766374e-07,
      "loss": 0.75460267,
      "num_input_tokens_seen": 147580045,
      "step": 6834,
      "time_per_iteration": 3.6174943447113037
    },
    {
      "auxiliary_loss_clip": 0.01148503,
      "auxiliary_loss_mlp": 0.01029643,
      "balance_loss_clip": 1.04620886,
      "balance_loss_mlp": 1.02179575,
      "epoch": 0.8218601575181868,
      "flos": 25629625388160.0,
      "grad_norm": 2.4784791358861042,
      "language_loss": 0.789873,
      "learning_rate": 3.2360700891433254e-07,
      "loss": 0.81165451,
      "num_input_tokens_seen": 147599070,
      "step": 6835,
      "time_per_iteration": 2.558473825454712
    },
    {
      "auxiliary_loss_clip": 0.01017003,
      "auxiliary_loss_mlp": 0.01002488,
      "balance_loss_clip": 1.0106312,
      "balance_loss_mlp": 1.00146317,
      "epoch": 0.8219804004088258,
      "flos": 67660229427840.0,
      "grad_norm": 0.8016290706564742,
      "language_loss": 0.57334048,
      "learning_rate": 3.231823095984847e-07,
      "loss": 0.59353542,
      "num_input_tokens_seen": 147653710,
      "step": 6836,
      "time_per_iteration": 3.085634469985962
    },
    {
      "auxiliary_loss_clip": 0.01136089,
      "auxiliary_loss_mlp": 0.01026388,
      "balance_loss_clip": 1.04609787,
      "balance_loss_mlp": 1.01967645,
      "epoch": 0.822100643299465,
      "flos": 19464266327040.0,
      "grad_norm": 2.190555133817007,
      "language_loss": 0.75894642,
      "learning_rate": 3.2275786465454814e-07,
      "loss": 0.78057116,
      "num_input_tokens_seen": 147670360,
      "step": 6837,
      "time_per_iteration": 2.494976282119751
    },
    {
      "auxiliary_loss_clip": 0.01120757,
      "auxiliary_loss_mlp": 0.0102207,
      "balance_loss_clip": 1.0434984,
      "balance_loss_mlp": 1.0149529,
      "epoch": 0.822220886190104,
      "flos": 24681368292480.0,
      "grad_norm": 1.634330416389713,
      "language_loss": 0.75182664,
      "learning_rate": 3.2233367414690917e-07,
      "loss": 0.77325487,
      "num_input_tokens_seen": 147692550,
      "step": 6838,
      "time_per_iteration": 2.5809335708618164
    },
    {
      "auxiliary_loss_clip": 0.01119415,
      "auxiliary_loss_mlp": 0.01023596,
      "balance_loss_clip": 1.04212165,
      "balance_loss_mlp": 1.01709032,
      "epoch": 0.8223411290807431,
      "flos": 27819062991360.0,
      "grad_norm": 2.104766468206614,
      "language_loss": 0.8487277,
      "learning_rate": 3.219097381399183e-07,
      "loss": 0.87015784,
      "num_input_tokens_seen": 147709725,
      "step": 6839,
      "time_per_iteration": 2.6004672050476074
    },
    {
      "auxiliary_loss_clip": 0.01143641,
      "auxiliary_loss_mlp": 0.01026429,
      "balance_loss_clip": 1.04651105,
      "balance_loss_mlp": 1.01946759,
      "epoch": 0.8224613719713821,
      "flos": 23218546913280.0,
      "grad_norm": 1.6922950298044408,
      "language_loss": 0.81252682,
      "learning_rate": 3.2148605669788584e-07,
      "loss": 0.83422756,
      "num_input_tokens_seen": 147729615,
      "step": 6840,
      "time_per_iteration": 2.5499320030212402
    },
    {
      "auxiliary_loss_clip": 0.01140134,
      "auxiliary_loss_mlp": 0.01025919,
      "balance_loss_clip": 1.04629135,
      "balance_loss_mlp": 1.01896369,
      "epoch": 0.8225816148620213,
      "flos": 15706250726400.0,
      "grad_norm": 2.172386854654803,
      "language_loss": 0.77545398,
      "learning_rate": 3.2106262988508405e-07,
      "loss": 0.79711449,
      "num_input_tokens_seen": 147747665,
      "step": 6841,
      "time_per_iteration": 2.5042037963867188
    },
    {
      "auxiliary_loss_clip": 0.01138113,
      "auxiliary_loss_mlp": 0.01029096,
      "balance_loss_clip": 1.04552114,
      "balance_loss_mlp": 1.02191329,
      "epoch": 0.8227018577526604,
      "flos": 18515111391360.0,
      "grad_norm": 1.6786229747416912,
      "language_loss": 0.74013418,
      "learning_rate": 3.206394577657465e-07,
      "loss": 0.76180625,
      "num_input_tokens_seen": 147765445,
      "step": 6842,
      "time_per_iteration": 2.506777763366699
    },
    {
      "auxiliary_loss_clip": 0.01155044,
      "auxiliary_loss_mlp": 0.0102489,
      "balance_loss_clip": 1.04723024,
      "balance_loss_mlp": 1.01718903,
      "epoch": 0.8228221006432994,
      "flos": 22236785406720.0,
      "grad_norm": 2.2329974980578755,
      "language_loss": 0.72520655,
      "learning_rate": 3.202165404040675e-07,
      "loss": 0.74700588,
      "num_input_tokens_seen": 147783365,
      "step": 6843,
      "time_per_iteration": 2.494231939315796
    },
    {
      "auxiliary_loss_clip": 0.01093347,
      "auxiliary_loss_mlp": 0.01024843,
      "balance_loss_clip": 1.0417459,
      "balance_loss_mlp": 1.01771474,
      "epoch": 0.8229423435339386,
      "flos": 24097532630400.0,
      "grad_norm": 2.2906272524392084,
      "language_loss": 0.74515563,
      "learning_rate": 3.1979387786420396e-07,
      "loss": 0.76633751,
      "num_input_tokens_seen": 147803605,
      "step": 6844,
      "time_per_iteration": 3.541156053543091
    },
    {
      "auxiliary_loss_clip": 0.01135149,
      "auxiliary_loss_mlp": 0.01025479,
      "balance_loss_clip": 1.04108143,
      "balance_loss_mlp": 1.01845479,
      "epoch": 0.8230625864245776,
      "flos": 23878549365120.0,
      "grad_norm": 1.7268959727786048,
      "language_loss": 0.81789529,
      "learning_rate": 3.1937147021027346e-07,
      "loss": 0.83950162,
      "num_input_tokens_seen": 147822060,
      "step": 6845,
      "time_per_iteration": 3.302060127258301
    },
    {
      "auxiliary_loss_clip": 0.01150495,
      "auxiliary_loss_mlp": 0.01025866,
      "balance_loss_clip": 1.04745793,
      "balance_loss_mlp": 1.01888633,
      "epoch": 0.8231828293152167,
      "flos": 16581106379520.0,
      "grad_norm": 2.2872735110871476,
      "language_loss": 0.76380217,
      "learning_rate": 3.189493175063547e-07,
      "loss": 0.78556579,
      "num_input_tokens_seen": 147839295,
      "step": 6846,
      "time_per_iteration": 2.4687583446502686
    },
    {
      "auxiliary_loss_clip": 0.01139044,
      "auxiliary_loss_mlp": 0.01025159,
      "balance_loss_clip": 1.04629302,
      "balance_loss_mlp": 1.01783323,
      "epoch": 0.8233030722058559,
      "flos": 18880071528960.0,
      "grad_norm": 2.0049695139086436,
      "language_loss": 0.66900969,
      "learning_rate": 3.1852741981648776e-07,
      "loss": 0.69065177,
      "num_input_tokens_seen": 147857945,
      "step": 6847,
      "time_per_iteration": 2.5121095180511475
    },
    {
      "auxiliary_loss_clip": 0.01109686,
      "auxiliary_loss_mlp": 0.01025549,
      "balance_loss_clip": 1.04235137,
      "balance_loss_mlp": 1.01848602,
      "epoch": 0.8234233150964949,
      "flos": 28439024757120.0,
      "grad_norm": 1.860320517922056,
      "language_loss": 0.69983637,
      "learning_rate": 3.1810577720467404e-07,
      "loss": 0.72118878,
      "num_input_tokens_seen": 147879675,
      "step": 6848,
      "time_per_iteration": 2.612328290939331
    },
    {
      "auxiliary_loss_clip": 0.01138714,
      "auxiliary_loss_mlp": 0.0102345,
      "balance_loss_clip": 1.04459381,
      "balance_loss_mlp": 1.01633656,
      "epoch": 0.823543557987134,
      "flos": 33765941577600.0,
      "grad_norm": 1.6355701032923053,
      "language_loss": 0.56227827,
      "learning_rate": 3.176843897348769e-07,
      "loss": 0.58389986,
      "num_input_tokens_seen": 147902870,
      "step": 6849,
      "time_per_iteration": 2.6879642009735107
    },
    {
      "auxiliary_loss_clip": 0.01131796,
      "auxiliary_loss_mlp": 0.01028514,
      "balance_loss_clip": 1.04397655,
      "balance_loss_mlp": 1.02116442,
      "epoch": 0.8236638008777731,
      "flos": 17092366611840.0,
      "grad_norm": 2.7398267244715586,
      "language_loss": 0.75384313,
      "learning_rate": 3.1726325747102034e-07,
      "loss": 0.77544618,
      "num_input_tokens_seen": 147921245,
      "step": 6850,
      "time_per_iteration": 2.485578775405884
    },
    {
      "auxiliary_loss_clip": 0.01101845,
      "auxiliary_loss_mlp": 0.01030538,
      "balance_loss_clip": 1.03677607,
      "balance_loss_mlp": 1.0230993,
      "epoch": 0.8237840437684122,
      "flos": 61639982334720.0,
      "grad_norm": 1.474646525181093,
      "language_loss": 0.63907039,
      "learning_rate": 3.1684238047698974e-07,
      "loss": 0.66039419,
      "num_input_tokens_seen": 147949515,
      "step": 6851,
      "time_per_iteration": 2.958782434463501
    },
    {
      "auxiliary_loss_clip": 0.011385,
      "auxiliary_loss_mlp": 0.01029227,
      "balance_loss_clip": 1.044976,
      "balance_loss_mlp": 1.02192807,
      "epoch": 0.8239042866590512,
      "flos": 27309023821440.0,
      "grad_norm": 2.460208046094842,
      "language_loss": 0.53089166,
      "learning_rate": 3.1642175881663155e-07,
      "loss": 0.55256885,
      "num_input_tokens_seen": 147969245,
      "step": 6852,
      "time_per_iteration": 2.584423065185547
    },
    {
      "auxiliary_loss_clip": 0.01162184,
      "auxiliary_loss_mlp": 0.01021517,
      "balance_loss_clip": 1.04606497,
      "balance_loss_mlp": 1.01441216,
      "epoch": 0.8240245295496904,
      "flos": 21726351187200.0,
      "grad_norm": 2.2317803227243336,
      "language_loss": 0.84044206,
      "learning_rate": 3.160013925537537e-07,
      "loss": 0.86227906,
      "num_input_tokens_seen": 147990080,
      "step": 6853,
      "time_per_iteration": 2.468045473098755
    },
    {
      "auxiliary_loss_clip": 0.01122142,
      "auxiliary_loss_mlp": 0.01026732,
      "balance_loss_clip": 1.04285288,
      "balance_loss_mlp": 1.01989198,
      "epoch": 0.8241447724403295,
      "flos": 20009318279040.0,
      "grad_norm": 2.2926231382498825,
      "language_loss": 0.75748992,
      "learning_rate": 3.155812817521266e-07,
      "loss": 0.77897865,
      "num_input_tokens_seen": 148010455,
      "step": 6854,
      "time_per_iteration": 2.5640451908111572
    },
    {
      "auxiliary_loss_clip": 0.01143246,
      "auxiliary_loss_mlp": 0.01027387,
      "balance_loss_clip": 1.04832053,
      "balance_loss_mlp": 1.02053273,
      "epoch": 0.8242650153309685,
      "flos": 22272983337600.0,
      "grad_norm": 1.826629152210343,
      "language_loss": 0.78189373,
      "learning_rate": 3.151614264754787e-07,
      "loss": 0.80360007,
      "num_input_tokens_seen": 148028400,
      "step": 6855,
      "time_per_iteration": 2.5237014293670654
    },
    {
      "auxiliary_loss_clip": 0.01163963,
      "auxiliary_loss_mlp": 0.01025369,
      "balance_loss_clip": 1.04616857,
      "balance_loss_mlp": 1.01807094,
      "epoch": 0.8243852582216077,
      "flos": 22309971367680.0,
      "grad_norm": 1.9510005595881217,
      "language_loss": 0.79374212,
      "learning_rate": 3.147418267875035e-07,
      "loss": 0.81563544,
      "num_input_tokens_seen": 148046530,
      "step": 6856,
      "time_per_iteration": 2.4558022022247314
    },
    {
      "auxiliary_loss_clip": 0.01087832,
      "auxiliary_loss_mlp": 0.00760799,
      "balance_loss_clip": 1.03655076,
      "balance_loss_mlp": 1.00028992,
      "epoch": 0.8245055011122467,
      "flos": 24645421756800.0,
      "grad_norm": 2.1169697292430163,
      "language_loss": 0.65284157,
      "learning_rate": 3.1432248275185315e-07,
      "loss": 0.67132783,
      "num_input_tokens_seen": 148067040,
      "step": 6857,
      "time_per_iteration": 2.6704189777374268
    },
    {
      "auxiliary_loss_clip": 0.01149333,
      "auxiliary_loss_mlp": 0.01024836,
      "balance_loss_clip": 1.04811621,
      "balance_loss_mlp": 1.01767182,
      "epoch": 0.8246257440028858,
      "flos": 17487275713920.0,
      "grad_norm": 1.9437145071144222,
      "language_loss": 0.76817179,
      "learning_rate": 3.139033944321412e-07,
      "loss": 0.78991348,
      "num_input_tokens_seen": 148084400,
      "step": 6858,
      "time_per_iteration": 2.459165096282959
    },
    {
      "auxiliary_loss_clip": 0.01153492,
      "auxiliary_loss_mlp": 0.01026324,
      "balance_loss_clip": 1.04622531,
      "balance_loss_mlp": 1.01913011,
      "epoch": 0.824745986893525,
      "flos": 25010130499200.0,
      "grad_norm": 1.6532719057592513,
      "language_loss": 0.79160112,
      "learning_rate": 3.1348456189194507e-07,
      "loss": 0.81339931,
      "num_input_tokens_seen": 148104860,
      "step": 6859,
      "time_per_iteration": 2.530111074447632
    },
    {
      "auxiliary_loss_clip": 0.01112303,
      "auxiliary_loss_mlp": 0.01024522,
      "balance_loss_clip": 1.04065561,
      "balance_loss_mlp": 1.01766133,
      "epoch": 0.824866229784164,
      "flos": 18772698798720.0,
      "grad_norm": 1.750713222753815,
      "language_loss": 0.82620406,
      "learning_rate": 3.1306598519479876e-07,
      "loss": 0.84757227,
      "num_input_tokens_seen": 148124680,
      "step": 6860,
      "time_per_iteration": 3.3977999687194824
    },
    {
      "auxiliary_loss_clip": 0.01133552,
      "auxiliary_loss_mlp": 0.01025781,
      "balance_loss_clip": 1.04525816,
      "balance_loss_mlp": 1.01930833,
      "epoch": 0.8249864726748031,
      "flos": 23842171866240.0,
      "grad_norm": 1.5828399546271836,
      "language_loss": 0.78141296,
      "learning_rate": 3.1264766440420177e-07,
      "loss": 0.80300629,
      "num_input_tokens_seen": 148147150,
      "step": 6861,
      "time_per_iteration": 2.5627734661102295
    },
    {
      "auxiliary_loss_clip": 0.01146372,
      "auxiliary_loss_mlp": 0.01025705,
      "balance_loss_clip": 1.04616451,
      "balance_loss_mlp": 1.01862705,
      "epoch": 0.8251067155654422,
      "flos": 20303103617280.0,
      "grad_norm": 1.896400830604076,
      "language_loss": 0.68980575,
      "learning_rate": 3.122295995836124e-07,
      "loss": 0.71152651,
      "num_input_tokens_seen": 148167020,
      "step": 6862,
      "time_per_iteration": 2.4984629154205322
    },
    {
      "auxiliary_loss_clip": 0.01153555,
      "auxiliary_loss_mlp": 0.01025379,
      "balance_loss_clip": 1.04498267,
      "balance_loss_mlp": 1.01783848,
      "epoch": 0.8252269584560813,
      "flos": 25009699536000.0,
      "grad_norm": 1.689114526578596,
      "language_loss": 0.77192575,
      "learning_rate": 3.118117907964508e-07,
      "loss": 0.79371512,
      "num_input_tokens_seen": 148188965,
      "step": 6863,
      "time_per_iteration": 2.5498530864715576
    },
    {
      "auxiliary_loss_clip": 0.01129824,
      "auxiliary_loss_mlp": 0.01027635,
      "balance_loss_clip": 1.04482639,
      "balance_loss_mlp": 1.02109015,
      "epoch": 0.8253472013467203,
      "flos": 17128564542720.0,
      "grad_norm": 1.9237896926429172,
      "language_loss": 0.80308539,
      "learning_rate": 3.1139423810609856e-07,
      "loss": 0.82465994,
      "num_input_tokens_seen": 148205660,
      "step": 6864,
      "time_per_iteration": 2.568167209625244
    },
    {
      "auxiliary_loss_clip": 0.01163011,
      "auxiliary_loss_mlp": 0.01020734,
      "balance_loss_clip": 1.04694128,
      "balance_loss_mlp": 1.01367998,
      "epoch": 0.8254674442373595,
      "flos": 22414794232320.0,
      "grad_norm": 1.9550759246208957,
      "language_loss": 0.75406456,
      "learning_rate": 3.1097694157589714e-07,
      "loss": 0.77590203,
      "num_input_tokens_seen": 148225545,
      "step": 6865,
      "time_per_iteration": 2.495314121246338
    },
    {
      "auxiliary_loss_clip": 0.01152317,
      "auxiliary_loss_mlp": 0.01029814,
      "balance_loss_clip": 1.04896069,
      "balance_loss_mlp": 1.02261066,
      "epoch": 0.8255876871279986,
      "flos": 24786765774720.0,
      "grad_norm": 2.9714693029573525,
      "language_loss": 0.76619518,
      "learning_rate": 3.105599012691511e-07,
      "loss": 0.78801644,
      "num_input_tokens_seen": 148243975,
      "step": 6866,
      "time_per_iteration": 2.510484218597412
    },
    {
      "auxiliary_loss_clip": 0.01145938,
      "auxiliary_loss_mlp": 0.0102365,
      "balance_loss_clip": 1.04548979,
      "balance_loss_mlp": 1.0168997,
      "epoch": 0.8257079300186376,
      "flos": 27455431656960.0,
      "grad_norm": 1.4312701539216148,
      "language_loss": 0.82365167,
      "learning_rate": 3.101431172491249e-07,
      "loss": 0.84534752,
      "num_input_tokens_seen": 148265520,
      "step": 6867,
      "time_per_iteration": 2.559732437133789
    },
    {
      "auxiliary_loss_clip": 0.01125139,
      "auxiliary_loss_mlp": 0.00760766,
      "balance_loss_clip": 1.04126549,
      "balance_loss_mlp": 1.00025725,
      "epoch": 0.8258281729092768,
      "flos": 16471866142080.0,
      "grad_norm": 3.08716701889091,
      "language_loss": 0.72361755,
      "learning_rate": 3.097265895790444e-07,
      "loss": 0.74247658,
      "num_input_tokens_seen": 148283730,
      "step": 6868,
      "time_per_iteration": 2.5346410274505615
    },
    {
      "auxiliary_loss_clip": 0.01124763,
      "auxiliary_loss_mlp": 0.01026462,
      "balance_loss_clip": 1.04431164,
      "balance_loss_mlp": 1.01902318,
      "epoch": 0.8259484157999158,
      "flos": 21433822824960.0,
      "grad_norm": 1.7133003405348077,
      "language_loss": 0.82990414,
      "learning_rate": 3.093103183220962e-07,
      "loss": 0.85141641,
      "num_input_tokens_seen": 148303775,
      "step": 6869,
      "time_per_iteration": 2.5667788982391357
    },
    {
      "auxiliary_loss_clip": 0.01046511,
      "auxiliary_loss_mlp": 0.01001626,
      "balance_loss_clip": 1.00897717,
      "balance_loss_mlp": 1.00047517,
      "epoch": 0.8260686586905549,
      "flos": 58322342453760.0,
      "grad_norm": 0.8201978931200202,
      "language_loss": 0.59424853,
      "learning_rate": 3.0889430354142796e-07,
      "loss": 0.61472988,
      "num_input_tokens_seen": 148365285,
      "step": 6870,
      "time_per_iteration": 3.9366321563720703
    },
    {
      "auxiliary_loss_clip": 0.01124207,
      "auxiliary_loss_mlp": 0.01026644,
      "balance_loss_clip": 1.04185557,
      "balance_loss_mlp": 1.01967573,
      "epoch": 0.826188901581194,
      "flos": 27527288814720.0,
      "grad_norm": 2.0981164144633255,
      "language_loss": 0.69977057,
      "learning_rate": 3.084785453001497e-07,
      "loss": 0.72127908,
      "num_input_tokens_seen": 148386200,
      "step": 6871,
      "time_per_iteration": 2.636141061782837
    },
    {
      "auxiliary_loss_clip": 0.01137018,
      "auxiliary_loss_mlp": 0.00760738,
      "balance_loss_clip": 1.04740727,
      "balance_loss_mlp": 1.00029433,
      "epoch": 0.8263091444718331,
      "flos": 23696051339520.0,
      "grad_norm": 2.232135199372507,
      "language_loss": 0.81604826,
      "learning_rate": 3.080630436613314e-07,
      "loss": 0.83502591,
      "num_input_tokens_seen": 148403970,
      "step": 6872,
      "time_per_iteration": 3.2671101093292236
    },
    {
      "auxiliary_loss_clip": 0.0114188,
      "auxiliary_loss_mlp": 0.01024109,
      "balance_loss_clip": 1.04434478,
      "balance_loss_mlp": 1.01761484,
      "epoch": 0.8264293873624722,
      "flos": 17165157523200.0,
      "grad_norm": 1.9390269680909522,
      "language_loss": 0.85696697,
      "learning_rate": 3.076477986880039e-07,
      "loss": 0.87862682,
      "num_input_tokens_seen": 148421765,
      "step": 6873,
      "time_per_iteration": 2.488128662109375
    },
    {
      "auxiliary_loss_clip": 0.0113611,
      "auxiliary_loss_mlp": 0.01026244,
      "balance_loss_clip": 1.04614544,
      "balance_loss_mlp": 1.01928186,
      "epoch": 0.8265496302531112,
      "flos": 24098645952000.0,
      "grad_norm": 1.968577843498901,
      "language_loss": 0.69294572,
      "learning_rate": 3.0723281044315986e-07,
      "loss": 0.71456927,
      "num_input_tokens_seen": 148443720,
      "step": 6874,
      "time_per_iteration": 2.56144118309021
    },
    {
      "auxiliary_loss_clip": 0.01162419,
      "auxiliary_loss_mlp": 0.01025503,
      "balance_loss_clip": 1.04715431,
      "balance_loss_mlp": 1.01919675,
      "epoch": 0.8266698731437504,
      "flos": 14099894599680.0,
      "grad_norm": 1.9238378969905074,
      "language_loss": 0.76437068,
      "learning_rate": 3.068180789897521e-07,
      "loss": 0.78624988,
      "num_input_tokens_seen": 148462130,
      "step": 6875,
      "time_per_iteration": 2.4342234134674072
    },
    {
      "auxiliary_loss_clip": 0.01153114,
      "auxiliary_loss_mlp": 0.01024365,
      "balance_loss_clip": 1.04554033,
      "balance_loss_mlp": 1.01724792,
      "epoch": 0.8267901160343895,
      "flos": 30777563715840.0,
      "grad_norm": 2.480386747495741,
      "language_loss": 0.81311107,
      "learning_rate": 3.064036043906966e-07,
      "loss": 0.83488584,
      "num_input_tokens_seen": 148485570,
      "step": 6876,
      "time_per_iteration": 2.5871052742004395
    },
    {
      "auxiliary_loss_clip": 0.01130077,
      "auxiliary_loss_mlp": 0.01029186,
      "balance_loss_clip": 1.04416645,
      "balance_loss_mlp": 1.02196217,
      "epoch": 0.8269103589250285,
      "flos": 40624915242240.0,
      "grad_norm": 1.9132545019362743,
      "language_loss": 0.67835331,
      "learning_rate": 3.059893867088668e-07,
      "loss": 0.69994593,
      "num_input_tokens_seen": 148509715,
      "step": 6877,
      "time_per_iteration": 2.7533822059631348
    },
    {
      "auxiliary_loss_clip": 0.01147926,
      "auxiliary_loss_mlp": 0.01025764,
      "balance_loss_clip": 1.04546595,
      "balance_loss_mlp": 1.01878381,
      "epoch": 0.8270306018156677,
      "flos": 30263645877120.0,
      "grad_norm": 1.8799802502088918,
      "language_loss": 0.66675746,
      "learning_rate": 3.055754260071004e-07,
      "loss": 0.68849444,
      "num_input_tokens_seen": 148532010,
      "step": 6878,
      "time_per_iteration": 2.5612289905548096
    },
    {
      "auxiliary_loss_clip": 0.01150996,
      "auxiliary_loss_mlp": 0.01027779,
      "balance_loss_clip": 1.0471034,
      "balance_loss_mlp": 1.02094233,
      "epoch": 0.8271508447063067,
      "flos": 25226599812480.0,
      "grad_norm": 2.0347420030660373,
      "language_loss": 0.73398846,
      "learning_rate": 3.051617223481948e-07,
      "loss": 0.75577623,
      "num_input_tokens_seen": 148553330,
      "step": 6879,
      "time_per_iteration": 2.533963441848755
    },
    {
      "auxiliary_loss_clip": 0.01131193,
      "auxiliary_loss_mlp": 0.01034432,
      "balance_loss_clip": 1.04524362,
      "balance_loss_mlp": 1.0262928,
      "epoch": 0.8272710875969458,
      "flos": 17566602900480.0,
      "grad_norm": 1.951111925215039,
      "language_loss": 0.74954289,
      "learning_rate": 3.047482757949078e-07,
      "loss": 0.77119917,
      "num_input_tokens_seen": 148570960,
      "step": 6880,
      "time_per_iteration": 2.5471670627593994
    },
    {
      "auxiliary_loss_clip": 0.01119687,
      "auxiliary_loss_mlp": 0.00759808,
      "balance_loss_clip": 1.04186463,
      "balance_loss_mlp": 1.00023389,
      "epoch": 0.827391330487585,
      "flos": 19755465886080.0,
      "grad_norm": 1.9237567424870434,
      "language_loss": 0.85930359,
      "learning_rate": 3.043350864099605e-07,
      "loss": 0.87809849,
      "num_input_tokens_seen": 148589520,
      "step": 6881,
      "time_per_iteration": 2.5626373291015625
    },
    {
      "auxiliary_loss_clip": 0.01152492,
      "auxiliary_loss_mlp": 0.01024267,
      "balance_loss_clip": 1.04589725,
      "balance_loss_mlp": 1.01732898,
      "epoch": 0.827511573378224,
      "flos": 16835174254080.0,
      "grad_norm": 2.136965144662471,
      "language_loss": 0.80607367,
      "learning_rate": 3.039221542560315e-07,
      "loss": 0.82784128,
      "num_input_tokens_seen": 148606085,
      "step": 6882,
      "time_per_iteration": 2.4555037021636963
    },
    {
      "auxiliary_loss_clip": 0.01147946,
      "auxiliary_loss_mlp": 0.01029232,
      "balance_loss_clip": 1.0468514,
      "balance_loss_mlp": 1.02206123,
      "epoch": 0.8276318162688631,
      "flos": 18369242259840.0,
      "grad_norm": 1.9765585519574091,
      "language_loss": 0.73803556,
      "learning_rate": 3.0350947939576356e-07,
      "loss": 0.75980735,
      "num_input_tokens_seen": 148625240,
      "step": 6883,
      "time_per_iteration": 2.487490177154541
    },
    {
      "auxiliary_loss_clip": 0.01153298,
      "auxiliary_loss_mlp": 0.01030552,
      "balance_loss_clip": 1.046067,
      "balance_loss_mlp": 1.02368248,
      "epoch": 0.8277520591595022,
      "flos": 19352691705600.0,
      "grad_norm": 1.5444719206469868,
      "language_loss": 0.72195923,
      "learning_rate": 3.0309706189175876e-07,
      "loss": 0.74379778,
      "num_input_tokens_seen": 148645075,
      "step": 6884,
      "time_per_iteration": 2.4765918254852295
    },
    {
      "auxiliary_loss_clip": 0.01037273,
      "auxiliary_loss_mlp": 0.01001254,
      "balance_loss_clip": 1.00892115,
      "balance_loss_mlp": 1.00016963,
      "epoch": 0.8278723020501413,
      "flos": 67918858329600.0,
      "grad_norm": 0.7549451449970025,
      "language_loss": 0.57411456,
      "learning_rate": 3.0268490180658045e-07,
      "loss": 0.59449983,
      "num_input_tokens_seen": 148707855,
      "step": 6885,
      "time_per_iteration": 3.1077535152435303
    },
    {
      "auxiliary_loss_clip": 0.01167039,
      "auxiliary_loss_mlp": 0.01026644,
      "balance_loss_clip": 1.04962456,
      "balance_loss_mlp": 1.01978099,
      "epoch": 0.8279925449407803,
      "flos": 18185738653440.0,
      "grad_norm": 2.1112361289801926,
      "language_loss": 0.79203427,
      "learning_rate": 3.0227299920275305e-07,
      "loss": 0.81397104,
      "num_input_tokens_seen": 148724170,
      "step": 6886,
      "time_per_iteration": 3.244901657104492
    },
    {
      "auxiliary_loss_clip": 0.01125859,
      "auxiliary_loss_mlp": 0.01029371,
      "balance_loss_clip": 1.04635489,
      "balance_loss_mlp": 1.02135456,
      "epoch": 0.8281127878314195,
      "flos": 20631434860800.0,
      "grad_norm": 1.8358536585927385,
      "language_loss": 0.85394937,
      "learning_rate": 3.018613541427613e-07,
      "loss": 0.87550163,
      "num_input_tokens_seen": 148743690,
      "step": 6887,
      "time_per_iteration": 2.542297601699829
    },
    {
      "auxiliary_loss_clip": 0.0115983,
      "auxiliary_loss_mlp": 0.01027034,
      "balance_loss_clip": 1.04453945,
      "balance_loss_mlp": 1.01976788,
      "epoch": 0.8282330307220586,
      "flos": 18004282122240.0,
      "grad_norm": 1.6007068669677438,
      "language_loss": 0.73481977,
      "learning_rate": 3.0144996668905243e-07,
      "loss": 0.75668842,
      "num_input_tokens_seen": 148761070,
      "step": 6888,
      "time_per_iteration": 2.4133596420288086
    },
    {
      "auxiliary_loss_clip": 0.01094927,
      "auxiliary_loss_mlp": 0.00760537,
      "balance_loss_clip": 1.03768885,
      "balance_loss_mlp": 1.00028276,
      "epoch": 0.8283532736126976,
      "flos": 20084120352000.0,
      "grad_norm": 1.9028374440675995,
      "language_loss": 0.81809688,
      "learning_rate": 3.010388369040331e-07,
      "loss": 0.83665156,
      "num_input_tokens_seen": 148779730,
      "step": 6889,
      "time_per_iteration": 2.6176748275756836
    },
    {
      "auxiliary_loss_clip": 0.01150986,
      "auxiliary_loss_mlp": 0.01024603,
      "balance_loss_clip": 1.04751623,
      "balance_loss_mlp": 1.0180589,
      "epoch": 0.8284735165033368,
      "flos": 31868421805440.0,
      "grad_norm": 1.792220357310396,
      "language_loss": 0.82707846,
      "learning_rate": 3.0062796485007156e-07,
      "loss": 0.8488344,
      "num_input_tokens_seen": 148800670,
      "step": 6890,
      "time_per_iteration": 2.5821361541748047
    },
    {
      "auxiliary_loss_clip": 0.0116457,
      "auxiliary_loss_mlp": 0.00761067,
      "balance_loss_clip": 1.04783356,
      "balance_loss_mlp": 1.00027943,
      "epoch": 0.8285937593939758,
      "flos": 26651319840000.0,
      "grad_norm": 2.2684641108547,
      "language_loss": 0.65847683,
      "learning_rate": 3.002173505894965e-07,
      "loss": 0.67773318,
      "num_input_tokens_seen": 148819820,
      "step": 6891,
      "time_per_iteration": 2.4991135597229004
    },
    {
      "auxiliary_loss_clip": 0.01154074,
      "auxiliary_loss_mlp": 0.01030438,
      "balance_loss_clip": 1.04579234,
      "balance_loss_mlp": 1.02285624,
      "epoch": 0.8287140022846149,
      "flos": 20193683811840.0,
      "grad_norm": 3.3767732423336883,
      "language_loss": 0.62769824,
      "learning_rate": 2.998069941845973e-07,
      "loss": 0.64954329,
      "num_input_tokens_seen": 148838890,
      "step": 6892,
      "time_per_iteration": 2.4892897605895996
    },
    {
      "auxiliary_loss_clip": 0.01055727,
      "auxiliary_loss_mlp": 0.01001909,
      "balance_loss_clip": 1.00843453,
      "balance_loss_mlp": 1.00085449,
      "epoch": 0.8288342451752541,
      "flos": 70755980019840.0,
      "grad_norm": 0.7200024382137162,
      "language_loss": 0.57480383,
      "learning_rate": 2.993968956976258e-07,
      "loss": 0.59538031,
      "num_input_tokens_seen": 148906635,
      "step": 6893,
      "time_per_iteration": 3.163667917251587
    },
    {
      "auxiliary_loss_clip": 0.01171216,
      "auxiliary_loss_mlp": 0.01029209,
      "balance_loss_clip": 1.0500226,
      "balance_loss_mlp": 1.0217073,
      "epoch": 0.8289544880658931,
      "flos": 24572235795840.0,
      "grad_norm": 3.091984748916916,
      "language_loss": 0.70207721,
      "learning_rate": 2.9898705519079313e-07,
      "loss": 0.7240814,
      "num_input_tokens_seen": 148925740,
      "step": 6894,
      "time_per_iteration": 2.4864370822906494
    },
    {
      "auxiliary_loss_clip": 0.01128602,
      "auxiliary_loss_mlp": 0.01027917,
      "balance_loss_clip": 1.04358435,
      "balance_loss_mlp": 1.02096665,
      "epoch": 0.8290747309565322,
      "flos": 22273378387200.0,
      "grad_norm": 1.6647851795656414,
      "language_loss": 0.74570096,
      "learning_rate": 2.985774727262715e-07,
      "loss": 0.76726609,
      "num_input_tokens_seen": 148944585,
      "step": 6895,
      "time_per_iteration": 2.5299339294433594
    },
    {
      "auxiliary_loss_clip": 0.01160823,
      "auxiliary_loss_mlp": 0.01024228,
      "balance_loss_clip": 1.04553318,
      "balance_loss_mlp": 1.0178504,
      "epoch": 0.8291949738471713,
      "flos": 23255570856960.0,
      "grad_norm": 1.9060100923375793,
      "language_loss": 0.81304991,
      "learning_rate": 2.981681483661949e-07,
      "loss": 0.8349005,
      "num_input_tokens_seen": 148964170,
      "step": 6896,
      "time_per_iteration": 3.4981472492218018
    },
    {
      "auxiliary_loss_clip": 0.01155037,
      "auxiliary_loss_mlp": 0.01027507,
      "balance_loss_clip": 1.05027318,
      "balance_loss_mlp": 1.01990116,
      "epoch": 0.8293152167378104,
      "flos": 52555768185600.0,
      "grad_norm": 1.6215569305227406,
      "language_loss": 0.71172404,
      "learning_rate": 2.9775908217265633e-07,
      "loss": 0.73354948,
      "num_input_tokens_seen": 148989405,
      "step": 6897,
      "time_per_iteration": 3.5501115322113037
    },
    {
      "auxiliary_loss_clip": 0.01003458,
      "auxiliary_loss_mlp": 0.01001412,
      "balance_loss_clip": 1.0084033,
      "balance_loss_mlp": 1.0001725,
      "epoch": 0.8294354596284494,
      "flos": 63356156294400.0,
      "grad_norm": 0.8331935919363206,
      "language_loss": 0.50378245,
      "learning_rate": 2.9735027420771253e-07,
      "loss": 0.52383113,
      "num_input_tokens_seen": 149049740,
      "step": 6898,
      "time_per_iteration": 3.91984224319458
    },
    {
      "auxiliary_loss_clip": 0.01131157,
      "auxiliary_loss_mlp": 0.01024407,
      "balance_loss_clip": 1.04752326,
      "balance_loss_mlp": 1.01839304,
      "epoch": 0.8295557025190886,
      "flos": 24827021942400.0,
      "grad_norm": 1.9590605177554137,
      "language_loss": 0.71374846,
      "learning_rate": 2.969417245333774e-07,
      "loss": 0.73530412,
      "num_input_tokens_seen": 149069120,
      "step": 6899,
      "time_per_iteration": 2.6585938930511475
    },
    {
      "auxiliary_loss_clip": 0.01117782,
      "auxiliary_loss_mlp": 0.01024332,
      "balance_loss_clip": 1.04494917,
      "balance_loss_mlp": 1.01727462,
      "epoch": 0.8296759454097277,
      "flos": 25118580637440.0,
      "grad_norm": 3.4282192890561656,
      "language_loss": 0.78123349,
      "learning_rate": 2.9653343321162915e-07,
      "loss": 0.80265462,
      "num_input_tokens_seen": 149088630,
      "step": 6900,
      "time_per_iteration": 2.577820062637329
    },
    {
      "auxiliary_loss_clip": 0.01122583,
      "auxiliary_loss_mlp": 0.01024708,
      "balance_loss_clip": 1.04525363,
      "balance_loss_mlp": 1.01737666,
      "epoch": 0.8297961883003667,
      "flos": 24132581326080.0,
      "grad_norm": 1.9838705493857713,
      "language_loss": 0.64902973,
      "learning_rate": 2.9612540030440446e-07,
      "loss": 0.67050266,
      "num_input_tokens_seen": 149109175,
      "step": 6901,
      "time_per_iteration": 2.6214756965637207
    },
    {
      "auxiliary_loss_clip": 0.01033665,
      "auxiliary_loss_mlp": 0.01000572,
      "balance_loss_clip": 1.00673175,
      "balance_loss_mlp": 0.9994694,
      "epoch": 0.8299164311910058,
      "flos": 67446561375360.0,
      "grad_norm": 0.8464807689237711,
      "language_loss": 0.64179438,
      "learning_rate": 2.9571762587360206e-07,
      "loss": 0.66213673,
      "num_input_tokens_seen": 149165560,
      "step": 6902,
      "time_per_iteration": 3.0558977127075195
    },
    {
      "auxiliary_loss_clip": 0.01103883,
      "auxiliary_loss_mlp": 0.01030223,
      "balance_loss_clip": 1.03654015,
      "balance_loss_mlp": 1.0236094,
      "epoch": 0.8300366740816449,
      "flos": 25228682801280.0,
      "grad_norm": 2.0206873145923416,
      "language_loss": 0.73894012,
      "learning_rate": 2.953101099810806e-07,
      "loss": 0.76028121,
      "num_input_tokens_seen": 149185165,
      "step": 6903,
      "time_per_iteration": 2.6441383361816406
    },
    {
      "auxiliary_loss_clip": 0.01145763,
      "auxiliary_loss_mlp": 0.01021909,
      "balance_loss_clip": 1.04888701,
      "balance_loss_mlp": 1.01503992,
      "epoch": 0.830156916972284,
      "flos": 18041018757120.0,
      "grad_norm": 2.1478537478170656,
      "language_loss": 0.82131624,
      "learning_rate": 2.9490285268865965e-07,
      "loss": 0.8429929,
      "num_input_tokens_seen": 149202655,
      "step": 6904,
      "time_per_iteration": 2.492135763168335
    },
    {
      "auxiliary_loss_clip": 0.01154189,
      "auxiliary_loss_mlp": 0.0103251,
      "balance_loss_clip": 1.04733765,
      "balance_loss_mlp": 1.02475286,
      "epoch": 0.830277159862923,
      "flos": 26322485806080.0,
      "grad_norm": 2.3107081566827485,
      "language_loss": 0.79741418,
      "learning_rate": 2.9449585405812085e-07,
      "loss": 0.8192811,
      "num_input_tokens_seen": 149220035,
      "step": 6905,
      "time_per_iteration": 2.5346853733062744
    },
    {
      "auxiliary_loss_clip": 0.01121745,
      "auxiliary_loss_mlp": 0.01028252,
      "balance_loss_clip": 1.04369259,
      "balance_loss_mlp": 1.02146554,
      "epoch": 0.8303974027535622,
      "flos": 19938861751680.0,
      "grad_norm": 1.8782173375178988,
      "language_loss": 0.73798847,
      "learning_rate": 2.940891141512043e-07,
      "loss": 0.7594884,
      "num_input_tokens_seen": 149238055,
      "step": 6906,
      "time_per_iteration": 2.5450196266174316
    },
    {
      "auxiliary_loss_clip": 0.0113324,
      "auxiliary_loss_mlp": 0.01025455,
      "balance_loss_clip": 1.04243922,
      "balance_loss_mlp": 1.01838899,
      "epoch": 0.8305176456442013,
      "flos": 17165552572800.0,
      "grad_norm": 2.6230782085655955,
      "language_loss": 0.72024798,
      "learning_rate": 2.9368263302961385e-07,
      "loss": 0.74183488,
      "num_input_tokens_seen": 149256755,
      "step": 6907,
      "time_per_iteration": 2.498473882675171
    },
    {
      "auxiliary_loss_clip": 0.0109203,
      "auxiliary_loss_mlp": 0.01026938,
      "balance_loss_clip": 1.03817129,
      "balance_loss_mlp": 1.01986027,
      "epoch": 0.8306378885348403,
      "flos": 25627614226560.0,
      "grad_norm": 1.8993583861618317,
      "language_loss": 0.7983768,
      "learning_rate": 2.9327641075501075e-07,
      "loss": 0.81956649,
      "num_input_tokens_seen": 149275745,
      "step": 6908,
      "time_per_iteration": 2.6744396686553955
    },
    {
      "auxiliary_loss_clip": 0.01129353,
      "auxiliary_loss_mlp": 0.01032008,
      "balance_loss_clip": 1.04136133,
      "balance_loss_mlp": 1.0244292,
      "epoch": 0.8307581314254795,
      "flos": 33947864985600.0,
      "grad_norm": 2.5212758564022013,
      "language_loss": 0.66277796,
      "learning_rate": 2.9287044738901866e-07,
      "loss": 0.68439162,
      "num_input_tokens_seen": 149293730,
      "step": 6909,
      "time_per_iteration": 2.6235599517822266
    },
    {
      "auxiliary_loss_clip": 0.01150842,
      "auxiliary_loss_mlp": 0.00760629,
      "balance_loss_clip": 1.04539132,
      "balance_loss_mlp": 1.00028038,
      "epoch": 0.8308783743161186,
      "flos": 17562724231680.0,
      "grad_norm": 2.2314383744752413,
      "language_loss": 0.90753531,
      "learning_rate": 2.9246474299322274e-07,
      "loss": 0.92665005,
      "num_input_tokens_seen": 149309290,
      "step": 6910,
      "time_per_iteration": 2.4742605686187744
    },
    {
      "auxiliary_loss_clip": 0.01020243,
      "auxiliary_loss_mlp": 0.01000786,
      "balance_loss_clip": 1.00860071,
      "balance_loss_mlp": 0.99966592,
      "epoch": 0.8309986172067576,
      "flos": 69412885649280.0,
      "grad_norm": 0.8873314165248043,
      "language_loss": 0.63125604,
      "learning_rate": 2.920592976291678e-07,
      "loss": 0.65146631,
      "num_input_tokens_seen": 149366620,
      "step": 6911,
      "time_per_iteration": 3.0706613063812256
    },
    {
      "auxiliary_loss_clip": 0.01149451,
      "auxiliary_loss_mlp": 0.01029055,
      "balance_loss_clip": 1.04566824,
      "balance_loss_mlp": 1.02185798,
      "epoch": 0.8311188600973968,
      "flos": 22309755886080.0,
      "grad_norm": 2.0331773594740223,
      "language_loss": 0.80727381,
      "learning_rate": 2.916541113583595e-07,
      "loss": 0.82905883,
      "num_input_tokens_seen": 149385120,
      "step": 6912,
      "time_per_iteration": 3.323946952819824
    },
    {
      "auxiliary_loss_clip": 0.01122882,
      "auxiliary_loss_mlp": 0.01024199,
      "balance_loss_clip": 1.0453825,
      "balance_loss_mlp": 1.01727307,
      "epoch": 0.8312391029880358,
      "flos": 18770077105920.0,
      "grad_norm": 1.9454694964846262,
      "language_loss": 0.66138637,
      "learning_rate": 2.912491842422642e-07,
      "loss": 0.68285716,
      "num_input_tokens_seen": 149402825,
      "step": 6913,
      "time_per_iteration": 2.528140068054199
    },
    {
      "auxiliary_loss_clip": 0.01150433,
      "auxiliary_loss_mlp": 0.01025002,
      "balance_loss_clip": 1.04592276,
      "balance_loss_mlp": 1.01850557,
      "epoch": 0.8313593458786749,
      "flos": 20376648714240.0,
      "grad_norm": 1.678789161350107,
      "language_loss": 0.71107662,
      "learning_rate": 2.9084451634230857e-07,
      "loss": 0.732831,
      "num_input_tokens_seen": 149422125,
      "step": 6914,
      "time_per_iteration": 2.498622179031372
    },
    {
      "auxiliary_loss_clip": 0.01121876,
      "auxiliary_loss_mlp": 0.01027846,
      "balance_loss_clip": 1.04239082,
      "balance_loss_mlp": 1.02012086,
      "epoch": 0.831479588769314,
      "flos": 32124069878400.0,
      "grad_norm": 2.183609824292139,
      "language_loss": 0.7159611,
      "learning_rate": 2.9044010771988125e-07,
      "loss": 0.73745829,
      "num_input_tokens_seen": 149441940,
      "step": 6915,
      "time_per_iteration": 2.6435840129852295
    },
    {
      "auxiliary_loss_clip": 0.0112886,
      "auxiliary_loss_mlp": 0.01026248,
      "balance_loss_clip": 1.04348421,
      "balance_loss_mlp": 1.01959658,
      "epoch": 0.8315998316599531,
      "flos": 45185929338240.0,
      "grad_norm": 1.9655659806882637,
      "language_loss": 0.71944344,
      "learning_rate": 2.900359584363303e-07,
      "loss": 0.74099457,
      "num_input_tokens_seen": 149465045,
      "step": 6916,
      "time_per_iteration": 2.7420504093170166
    },
    {
      "auxiliary_loss_clip": 0.01106974,
      "auxiliary_loss_mlp": 0.01036115,
      "balance_loss_clip": 1.04603755,
      "balance_loss_mlp": 1.02818155,
      "epoch": 0.8317200745505922,
      "flos": 18363747479040.0,
      "grad_norm": 2.1619751453930878,
      "language_loss": 0.84580094,
      "learning_rate": 2.8963206855296494e-07,
      "loss": 0.86723185,
      "num_input_tokens_seen": 149481285,
      "step": 6917,
      "time_per_iteration": 2.5486273765563965
    },
    {
      "auxiliary_loss_clip": 0.01149279,
      "auxiliary_loss_mlp": 0.01025599,
      "balance_loss_clip": 1.04451215,
      "balance_loss_mlp": 1.01896191,
      "epoch": 0.8318403174412313,
      "flos": 24206557386240.0,
      "grad_norm": 2.097963353352098,
      "language_loss": 0.76986897,
      "learning_rate": 2.892284381310548e-07,
      "loss": 0.79161775,
      "num_input_tokens_seen": 149502700,
      "step": 6918,
      "time_per_iteration": 2.5238144397735596
    },
    {
      "auxiliary_loss_clip": 0.0113235,
      "auxiliary_loss_mlp": 0.01027978,
      "balance_loss_clip": 1.04495144,
      "balance_loss_mlp": 1.02018166,
      "epoch": 0.8319605603318704,
      "flos": 22418780641920.0,
      "grad_norm": 2.408191634394838,
      "language_loss": 0.72620606,
      "learning_rate": 2.888250672318302e-07,
      "loss": 0.74780935,
      "num_input_tokens_seen": 149520100,
      "step": 6919,
      "time_per_iteration": 2.520975112915039
    },
    {
      "auxiliary_loss_clip": 0.01166618,
      "auxiliary_loss_mlp": 0.0102371,
      "balance_loss_clip": 1.04920876,
      "balance_loss_mlp": 1.01633358,
      "epoch": 0.8320808032225094,
      "flos": 37414501459200.0,
      "grad_norm": 1.8791478253207499,
      "language_loss": 0.68471563,
      "learning_rate": 2.884219559164831e-07,
      "loss": 0.70661896,
      "num_input_tokens_seen": 149543245,
      "step": 6920,
      "time_per_iteration": 2.6202375888824463
    },
    {
      "auxiliary_loss_clip": 0.01149717,
      "auxiliary_loss_mlp": 0.01027005,
      "balance_loss_clip": 1.04734325,
      "balance_loss_mlp": 1.0198462,
      "epoch": 0.8322010461131486,
      "flos": 12787395638400.0,
      "grad_norm": 1.7738973017407877,
      "language_loss": 0.81305653,
      "learning_rate": 2.880191042461635e-07,
      "loss": 0.83482373,
      "num_input_tokens_seen": 149559185,
      "step": 6921,
      "time_per_iteration": 2.4635233879089355
    },
    {
      "auxiliary_loss_clip": 0.01114242,
      "auxiliary_loss_mlp": 0.01024882,
      "balance_loss_clip": 1.0418452,
      "balance_loss_mlp": 1.01843584,
      "epoch": 0.8323212890037877,
      "flos": 15815455050240.0,
      "grad_norm": 1.8041463115212717,
      "language_loss": 0.80084991,
      "learning_rate": 2.876165122819849e-07,
      "loss": 0.82224119,
      "num_input_tokens_seen": 149577165,
      "step": 6922,
      "time_per_iteration": 4.1805949211120605
    },
    {
      "auxiliary_loss_clip": 0.01161583,
      "auxiliary_loss_mlp": 0.01025222,
      "balance_loss_clip": 1.04801238,
      "balance_loss_mlp": 1.01795888,
      "epoch": 0.8324415318944267,
      "flos": 21719276208000.0,
      "grad_norm": 1.504507128361725,
      "language_loss": 0.791767,
      "learning_rate": 2.872141800850201e-07,
      "loss": 0.81363499,
      "num_input_tokens_seen": 149594340,
      "step": 6923,
      "time_per_iteration": 2.4639930725097656
    },
    {
      "auxiliary_loss_clip": 0.0116218,
      "auxiliary_loss_mlp": 0.01026709,
      "balance_loss_clip": 1.0470798,
      "balance_loss_mlp": 1.02017963,
      "epoch": 0.8325617747850659,
      "flos": 34198700636160.0,
      "grad_norm": 1.5905806187787586,
      "language_loss": 0.72978544,
      "learning_rate": 2.868121077163024e-07,
      "loss": 0.75167429,
      "num_input_tokens_seen": 149613895,
      "step": 6924,
      "time_per_iteration": 3.2682645320892334
    },
    {
      "auxiliary_loss_clip": 0.01150252,
      "auxiliary_loss_mlp": 0.01027764,
      "balance_loss_clip": 1.04374492,
      "balance_loss_mlp": 1.02046287,
      "epoch": 0.8326820176757049,
      "flos": 18369457741440.0,
      "grad_norm": 1.906763391193866,
      "language_loss": 0.71792132,
      "learning_rate": 2.864102952368257e-07,
      "loss": 0.73970151,
      "num_input_tokens_seen": 149631820,
      "step": 6925,
      "time_per_iteration": 2.4722516536712646
    },
    {
      "auxiliary_loss_clip": 0.01098295,
      "auxiliary_loss_mlp": 0.01027673,
      "balance_loss_clip": 1.03768682,
      "balance_loss_mlp": 1.02039528,
      "epoch": 0.832802260566344,
      "flos": 35991325716480.0,
      "grad_norm": 1.2155282795429059,
      "language_loss": 0.5930047,
      "learning_rate": 2.860087427075444e-07,
      "loss": 0.61426443,
      "num_input_tokens_seen": 149656070,
      "step": 6926,
      "time_per_iteration": 2.7123942375183105
    },
    {
      "auxiliary_loss_clip": 0.01132716,
      "auxiliary_loss_mlp": 0.01030122,
      "balance_loss_clip": 1.04466772,
      "balance_loss_mlp": 1.02339244,
      "epoch": 0.8329225034569832,
      "flos": 14244434928000.0,
      "grad_norm": 3.957038101064704,
      "language_loss": 0.86509615,
      "learning_rate": 2.856074501893744e-07,
      "loss": 0.88672453,
      "num_input_tokens_seen": 149671270,
      "step": 6927,
      "time_per_iteration": 2.4870588779449463
    },
    {
      "auxiliary_loss_clip": 0.0115197,
      "auxiliary_loss_mlp": 0.01025658,
      "balance_loss_clip": 1.04890585,
      "balance_loss_mlp": 1.01857352,
      "epoch": 0.8330427463476222,
      "flos": 18077468083200.0,
      "grad_norm": 1.7068402110295868,
      "language_loss": 0.81339824,
      "learning_rate": 2.8520641774319054e-07,
      "loss": 0.8351745,
      "num_input_tokens_seen": 149689360,
      "step": 6928,
      "time_per_iteration": 2.469676971435547
    },
    {
      "auxiliary_loss_clip": 0.0113349,
      "auxiliary_loss_mlp": 0.01024716,
      "balance_loss_clip": 1.03932095,
      "balance_loss_mlp": 1.01715195,
      "epoch": 0.8331629892382613,
      "flos": 18040839189120.0,
      "grad_norm": 2.1214569376378263,
      "language_loss": 0.76454782,
      "learning_rate": 2.848056454298309e-07,
      "loss": 0.78612989,
      "num_input_tokens_seen": 149706685,
      "step": 6929,
      "time_per_iteration": 2.490370035171509
    },
    {
      "auxiliary_loss_clip": 0.0113642,
      "auxiliary_loss_mlp": 0.01034846,
      "balance_loss_clip": 1.04678035,
      "balance_loss_mlp": 1.02801251,
      "epoch": 0.8332832321289004,
      "flos": 17457398576640.0,
      "grad_norm": 2.1437969134455073,
      "language_loss": 0.654728,
      "learning_rate": 2.844051333100905e-07,
      "loss": 0.6764406,
      "num_input_tokens_seen": 149724230,
      "step": 6930,
      "time_per_iteration": 2.507584571838379
    },
    {
      "auxiliary_loss_clip": 0.01137791,
      "auxiliary_loss_mlp": 0.01029014,
      "balance_loss_clip": 1.04812276,
      "balance_loss_mlp": 1.02288985,
      "epoch": 0.8334034750195395,
      "flos": 15084852416640.0,
      "grad_norm": 1.784572527459324,
      "language_loss": 0.83658791,
      "learning_rate": 2.840048814447269e-07,
      "loss": 0.85825598,
      "num_input_tokens_seen": 149742395,
      "step": 6931,
      "time_per_iteration": 2.5012803077697754
    },
    {
      "auxiliary_loss_clip": 0.01127364,
      "auxiliary_loss_mlp": 0.010278,
      "balance_loss_clip": 1.04226565,
      "balance_loss_mlp": 1.02040315,
      "epoch": 0.8335237179101785,
      "flos": 19427170556160.0,
      "grad_norm": 2.2468238284084556,
      "language_loss": 0.74125546,
      "learning_rate": 2.836048898944587e-07,
      "loss": 0.76280713,
      "num_input_tokens_seen": 149760820,
      "step": 6932,
      "time_per_iteration": 2.5041511058807373
    },
    {
      "auxiliary_loss_clip": 0.01135174,
      "auxiliary_loss_mlp": 0.01023014,
      "balance_loss_clip": 1.04357815,
      "balance_loss_mlp": 1.01635885,
      "epoch": 0.8336439608008177,
      "flos": 21762046327680.0,
      "grad_norm": 2.9272970814453,
      "language_loss": 0.72563815,
      "learning_rate": 2.832051587199642e-07,
      "loss": 0.74721992,
      "num_input_tokens_seen": 149778075,
      "step": 6933,
      "time_per_iteration": 2.53576922416687
    },
    {
      "auxiliary_loss_clip": 0.01046519,
      "auxiliary_loss_mlp": 0.01000502,
      "balance_loss_clip": 1.00814819,
      "balance_loss_mlp": 0.99932194,
      "epoch": 0.8337642036914568,
      "flos": 59702783990400.0,
      "grad_norm": 0.8606813897775577,
      "language_loss": 0.57772636,
      "learning_rate": 2.828056879818821e-07,
      "loss": 0.59819663,
      "num_input_tokens_seen": 149837150,
      "step": 6934,
      "time_per_iteration": 3.035710573196411
    },
    {
      "auxiliary_loss_clip": 0.01119193,
      "auxiliary_loss_mlp": 0.01028614,
      "balance_loss_clip": 1.0387013,
      "balance_loss_mlp": 1.0221616,
      "epoch": 0.8338844465820958,
      "flos": 27162185022720.0,
      "grad_norm": 1.9893297660813498,
      "language_loss": 0.83579826,
      "learning_rate": 2.824064777408117e-07,
      "loss": 0.85727632,
      "num_input_tokens_seen": 149856940,
      "step": 6935,
      "time_per_iteration": 2.5999209880828857
    },
    {
      "auxiliary_loss_clip": 0.01150146,
      "auxiliary_loss_mlp": 0.01029275,
      "balance_loss_clip": 1.04775047,
      "balance_loss_mlp": 1.02205062,
      "epoch": 0.8340046894727349,
      "flos": 30481264425600.0,
      "grad_norm": 1.8010123581805222,
      "language_loss": 0.75971317,
      "learning_rate": 2.8200752805731263e-07,
      "loss": 0.78150737,
      "num_input_tokens_seen": 149879930,
      "step": 6936,
      "time_per_iteration": 2.582643747329712
    },
    {
      "auxiliary_loss_clip": 0.01150903,
      "auxiliary_loss_mlp": 0.01034754,
      "balance_loss_clip": 1.04791331,
      "balance_loss_mlp": 1.02744341,
      "epoch": 0.834124932363374,
      "flos": 27126166659840.0,
      "grad_norm": 1.4657558330022697,
      "language_loss": 0.80977416,
      "learning_rate": 2.8160883899190625e-07,
      "loss": 0.83163083,
      "num_input_tokens_seen": 149903200,
      "step": 6937,
      "time_per_iteration": 3.3680598735809326
    },
    {
      "auxiliary_loss_clip": 0.01113268,
      "auxiliary_loss_mlp": 0.01028677,
      "balance_loss_clip": 1.04452181,
      "balance_loss_mlp": 1.02194142,
      "epoch": 0.8342451752540131,
      "flos": 24569865498240.0,
      "grad_norm": 1.9791017051973514,
      "language_loss": 0.73281473,
      "learning_rate": 2.8121041060507234e-07,
      "loss": 0.75423419,
      "num_input_tokens_seen": 149922230,
      "step": 6938,
      "time_per_iteration": 2.5991439819335938
    },
    {
      "auxiliary_loss_clip": 0.01152498,
      "auxiliary_loss_mlp": 0.01022477,
      "balance_loss_clip": 1.04510832,
      "balance_loss_mlp": 1.01560163,
      "epoch": 0.8343654181446521,
      "flos": 26615085995520.0,
      "grad_norm": 1.6533816019507215,
      "language_loss": 0.7149142,
      "learning_rate": 2.808122429572528e-07,
      "loss": 0.73666394,
      "num_input_tokens_seen": 149942435,
      "step": 6939,
      "time_per_iteration": 2.525315284729004
    },
    {
      "auxiliary_loss_clip": 0.01131314,
      "auxiliary_loss_mlp": 0.01024999,
      "balance_loss_clip": 1.04516673,
      "balance_loss_mlp": 1.01838017,
      "epoch": 0.8344856610352913,
      "flos": 20777268078720.0,
      "grad_norm": 3.002769587587452,
      "language_loss": 0.75834548,
      "learning_rate": 2.804143361088489e-07,
      "loss": 0.77990866,
      "num_input_tokens_seen": 149961615,
      "step": 6940,
      "time_per_iteration": 2.5601181983947754
    },
    {
      "auxiliary_loss_clip": 0.0113066,
      "auxiliary_loss_mlp": 0.01023569,
      "balance_loss_clip": 1.04403853,
      "balance_loss_mlp": 1.01682746,
      "epoch": 0.8346059039259304,
      "flos": 26095960684800.0,
      "grad_norm": 2.1184543043852875,
      "language_loss": 0.77812815,
      "learning_rate": 2.8001669012022277e-07,
      "loss": 0.7996704,
      "num_input_tokens_seen": 149979585,
      "step": 6941,
      "time_per_iteration": 2.5500922203063965
    },
    {
      "auxiliary_loss_clip": 0.0115069,
      "auxiliary_loss_mlp": 0.01027115,
      "balance_loss_clip": 1.04904294,
      "balance_loss_mlp": 1.02050233,
      "epoch": 0.8347261468165694,
      "flos": 29027708755200.0,
      "grad_norm": 2.88573513958506,
      "language_loss": 0.69206262,
      "learning_rate": 2.7961930505169795e-07,
      "loss": 0.71384066,
      "num_input_tokens_seen": 150003830,
      "step": 6942,
      "time_per_iteration": 2.574700355529785
    },
    {
      "auxiliary_loss_clip": 0.01152558,
      "auxiliary_loss_mlp": 0.00760741,
      "balance_loss_clip": 1.04690421,
      "balance_loss_mlp": 1.00028253,
      "epoch": 0.8348463897072086,
      "flos": 26396461866240.0,
      "grad_norm": 1.813701949470426,
      "language_loss": 0.76458108,
      "learning_rate": 2.792221809635558e-07,
      "loss": 0.78371412,
      "num_input_tokens_seen": 150024460,
      "step": 6943,
      "time_per_iteration": 2.531498908996582
    },
    {
      "auxiliary_loss_clip": 0.01081154,
      "auxiliary_loss_mlp": 0.01024117,
      "balance_loss_clip": 1.04009283,
      "balance_loss_mlp": 1.01641011,
      "epoch": 0.8349666325978476,
      "flos": 23367720096000.0,
      "grad_norm": 2.0243868426321003,
      "language_loss": 0.75070429,
      "learning_rate": 2.788253179160411e-07,
      "loss": 0.77175701,
      "num_input_tokens_seen": 150045620,
      "step": 6944,
      "time_per_iteration": 2.6595914363861084
    },
    {
      "auxiliary_loss_clip": 0.011373,
      "auxiliary_loss_mlp": 0.01028937,
      "balance_loss_clip": 1.04593992,
      "balance_loss_mlp": 1.02230597,
      "epoch": 0.8350868754884867,
      "flos": 12896528135040.0,
      "grad_norm": 1.9969684922585191,
      "language_loss": 0.6487689,
      "learning_rate": 2.7842871596935725e-07,
      "loss": 0.6704312,
      "num_input_tokens_seen": 150064135,
      "step": 6945,
      "time_per_iteration": 2.4938418865203857
    },
    {
      "auxiliary_loss_clip": 0.01150638,
      "auxiliary_loss_mlp": 0.01027064,
      "balance_loss_clip": 1.0451777,
      "balance_loss_mlp": 1.01998329,
      "epoch": 0.8352071183791259,
      "flos": 26505522535680.0,
      "grad_norm": 1.5891884532658413,
      "language_loss": 0.69336224,
      "learning_rate": 2.780323751836682e-07,
      "loss": 0.71513927,
      "num_input_tokens_seen": 150085350,
      "step": 6946,
      "time_per_iteration": 2.5338375568389893
    },
    {
      "auxiliary_loss_clip": 0.01134334,
      "auxiliary_loss_mlp": 0.00760414,
      "balance_loss_clip": 1.04177403,
      "balance_loss_mlp": 1.00030375,
      "epoch": 0.8353273612697649,
      "flos": 20668063754880.0,
      "grad_norm": 1.418877767900213,
      "language_loss": 0.78774226,
      "learning_rate": 2.7763629561909876e-07,
      "loss": 0.80668974,
      "num_input_tokens_seen": 150106180,
      "step": 6947,
      "time_per_iteration": 2.5455806255340576
    },
    {
      "auxiliary_loss_clip": 0.01161892,
      "auxiliary_loss_mlp": 0.0102449,
      "balance_loss_clip": 1.04652619,
      "balance_loss_mlp": 1.01733696,
      "epoch": 0.835447604160404,
      "flos": 19754137082880.0,
      "grad_norm": 1.9011594481347391,
      "language_loss": 0.77043235,
      "learning_rate": 2.772404773357335e-07,
      "loss": 0.79229623,
      "num_input_tokens_seen": 150125585,
      "step": 6948,
      "time_per_iteration": 4.042009353637695
    },
    {
      "auxiliary_loss_clip": 0.01113059,
      "auxiliary_loss_mlp": 0.01023524,
      "balance_loss_clip": 1.04204798,
      "balance_loss_mlp": 1.0166868,
      "epoch": 0.8355678470510431,
      "flos": 23435842239360.0,
      "grad_norm": 2.310157871185444,
      "language_loss": 0.78437167,
      "learning_rate": 2.7684492039361853e-07,
      "loss": 0.8057375,
      "num_input_tokens_seen": 150144810,
      "step": 6949,
      "time_per_iteration": 2.573755979537964
    },
    {
      "auxiliary_loss_clip": 0.01166374,
      "auxiliary_loss_mlp": 0.01026772,
      "balance_loss_clip": 1.05042684,
      "balance_loss_mlp": 1.02005124,
      "epoch": 0.8356880899416822,
      "flos": 21214588164480.0,
      "grad_norm": 1.7395999244525835,
      "language_loss": 0.83807564,
      "learning_rate": 2.764496248527586e-07,
      "loss": 0.86000705,
      "num_input_tokens_seen": 150163785,
      "step": 6950,
      "time_per_iteration": 3.2029471397399902
    },
    {
      "auxiliary_loss_clip": 0.01125033,
      "auxiliary_loss_mlp": 0.01024412,
      "balance_loss_clip": 1.04242051,
      "balance_loss_mlp": 1.01739645,
      "epoch": 0.8358083328323213,
      "flos": 28037543466240.0,
      "grad_norm": 1.7428694759786143,
      "language_loss": 0.7845186,
      "learning_rate": 2.760545907731211e-07,
      "loss": 0.80601299,
      "num_input_tokens_seen": 150184360,
      "step": 6951,
      "time_per_iteration": 2.5964114665985107
    },
    {
      "auxiliary_loss_clip": 0.01149679,
      "auxiliary_loss_mlp": 0.01031819,
      "balance_loss_clip": 1.04467058,
      "balance_loss_mlp": 1.02465999,
      "epoch": 0.8359285757229604,
      "flos": 27783655159680.0,
      "grad_norm": 1.8009366962808282,
      "language_loss": 0.67856836,
      "learning_rate": 2.75659818214631e-07,
      "loss": 0.70038342,
      "num_input_tokens_seen": 150205465,
      "step": 6952,
      "time_per_iteration": 2.5559446811676025
    },
    {
      "auxiliary_loss_clip": 0.01138956,
      "auxiliary_loss_mlp": 0.01025753,
      "balance_loss_clip": 1.04455447,
      "balance_loss_mlp": 1.019238,
      "epoch": 0.8360488186135995,
      "flos": 21435115714560.0,
      "grad_norm": 2.6500068444593965,
      "language_loss": 0.77958357,
      "learning_rate": 2.752653072371749e-07,
      "loss": 0.80123061,
      "num_input_tokens_seen": 150224900,
      "step": 6953,
      "time_per_iteration": 2.530442476272583
    },
    {
      "auxiliary_loss_clip": 0.0112091,
      "auxiliary_loss_mlp": 0.01028468,
      "balance_loss_clip": 1.04623795,
      "balance_loss_mlp": 1.02167916,
      "epoch": 0.8361690615042385,
      "flos": 27632327160960.0,
      "grad_norm": 1.788053440864496,
      "language_loss": 0.74744362,
      "learning_rate": 2.7487105790060105e-07,
      "loss": 0.76893741,
      "num_input_tokens_seen": 150244310,
      "step": 6954,
      "time_per_iteration": 2.600433588027954
    },
    {
      "auxiliary_loss_clip": 0.0115145,
      "auxiliary_loss_mlp": 0.01021378,
      "balance_loss_clip": 1.0458554,
      "balance_loss_mlp": 1.01520872,
      "epoch": 0.8362893043948777,
      "flos": 39202529598720.0,
      "grad_norm": 1.818640304444648,
      "language_loss": 0.69312578,
      "learning_rate": 2.7447707026471587e-07,
      "loss": 0.714854,
      "num_input_tokens_seen": 150267285,
      "step": 6955,
      "time_per_iteration": 2.6609978675842285
    },
    {
      "auxiliary_loss_clip": 0.01123124,
      "auxiliary_loss_mlp": 0.01023147,
      "balance_loss_clip": 1.04177511,
      "balance_loss_mlp": 1.01642418,
      "epoch": 0.8364095472855168,
      "flos": 24785329230720.0,
      "grad_norm": 1.8959171807103095,
      "language_loss": 0.79701591,
      "learning_rate": 2.740833443892874e-07,
      "loss": 0.81847858,
      "num_input_tokens_seen": 150285455,
      "step": 6956,
      "time_per_iteration": 2.5894551277160645
    },
    {
      "auxiliary_loss_clip": 0.01136616,
      "auxiliary_loss_mlp": 0.01028376,
      "balance_loss_clip": 1.044541,
      "balance_loss_mlp": 1.02161694,
      "epoch": 0.8365297901761558,
      "flos": 22743412784640.0,
      "grad_norm": 2.1104840003381518,
      "language_loss": 0.79624593,
      "learning_rate": 2.7368988033404327e-07,
      "loss": 0.81789583,
      "num_input_tokens_seen": 150302970,
      "step": 6957,
      "time_per_iteration": 2.541539430618286
    },
    {
      "auxiliary_loss_clip": 0.0112583,
      "auxiliary_loss_mlp": 0.01028207,
      "balance_loss_clip": 1.04454756,
      "balance_loss_mlp": 1.02158189,
      "epoch": 0.836650033066795,
      "flos": 28396003242240.0,
      "grad_norm": 1.6385348365862804,
      "language_loss": 0.84315002,
      "learning_rate": 2.732966781586712e-07,
      "loss": 0.8646903,
      "num_input_tokens_seen": 150322715,
      "step": 6958,
      "time_per_iteration": 2.6140875816345215
    },
    {
      "auxiliary_loss_clip": 0.01145139,
      "auxiliary_loss_mlp": 0.01027795,
      "balance_loss_clip": 1.04406095,
      "balance_loss_mlp": 1.02091074,
      "epoch": 0.836770275957434,
      "flos": 22236857233920.0,
      "grad_norm": 1.7950866027382386,
      "language_loss": 0.66726971,
      "learning_rate": 2.729037379228205e-07,
      "loss": 0.688999,
      "num_input_tokens_seen": 150342900,
      "step": 6959,
      "time_per_iteration": 2.5265390872955322
    },
    {
      "auxiliary_loss_clip": 0.01138429,
      "auxiliary_loss_mlp": 0.01027031,
      "balance_loss_clip": 1.04806042,
      "balance_loss_mlp": 1.02020037,
      "epoch": 0.8368905188480731,
      "flos": 22491930689280.0,
      "grad_norm": 1.4139741463418498,
      "language_loss": 0.80470526,
      "learning_rate": 2.725110596860998e-07,
      "loss": 0.82635981,
      "num_input_tokens_seen": 150363580,
      "step": 6960,
      "time_per_iteration": 2.570114850997925
    },
    {
      "auxiliary_loss_clip": 0.01106071,
      "auxiliary_loss_mlp": 0.01025792,
      "balance_loss_clip": 1.04348159,
      "balance_loss_mlp": 1.0191133,
      "epoch": 0.8370107617387123,
      "flos": 13370405287680.0,
      "grad_norm": 1.972801501213052,
      "language_loss": 0.69863671,
      "learning_rate": 2.7211864350807776e-07,
      "loss": 0.71995533,
      "num_input_tokens_seen": 150381780,
      "step": 6961,
      "time_per_iteration": 2.6092357635498047
    },
    {
      "auxiliary_loss_clip": 0.01163251,
      "auxiliary_loss_mlp": 0.01025626,
      "balance_loss_clip": 1.04637194,
      "balance_loss_mlp": 1.01835775,
      "epoch": 0.8371310046293513,
      "flos": 25261289372160.0,
      "grad_norm": 1.5544537257518154,
      "language_loss": 0.73609275,
      "learning_rate": 2.717264894482836e-07,
      "loss": 0.75798148,
      "num_input_tokens_seen": 150402120,
      "step": 6962,
      "time_per_iteration": 2.4989330768585205
    },
    {
      "auxiliary_loss_clip": 0.01151094,
      "auxiliary_loss_mlp": 0.01024668,
      "balance_loss_clip": 1.04618549,
      "balance_loss_mlp": 1.01760197,
      "epoch": 0.8372512475199904,
      "flos": 19792705311360.0,
      "grad_norm": 1.918830587723856,
      "language_loss": 0.81022978,
      "learning_rate": 2.7133459756620646e-07,
      "loss": 0.83198744,
      "num_input_tokens_seen": 150419315,
      "step": 6963,
      "time_per_iteration": 3.310995578765869
    },
    {
      "auxiliary_loss_clip": 0.01143135,
      "auxiliary_loss_mlp": 0.01027953,
      "balance_loss_clip": 1.04398012,
      "balance_loss_mlp": 1.02059793,
      "epoch": 0.8373714904106295,
      "flos": 19391224020480.0,
      "grad_norm": 1.6041692642114038,
      "language_loss": 0.73805106,
      "learning_rate": 2.7094296792129733e-07,
      "loss": 0.75976199,
      "num_input_tokens_seen": 150438915,
      "step": 6964,
      "time_per_iteration": 2.486720323562622
    },
    {
      "auxiliary_loss_clip": 0.01151637,
      "auxiliary_loss_mlp": 0.01028177,
      "balance_loss_clip": 1.04652119,
      "balance_loss_mlp": 1.02155757,
      "epoch": 0.8374917333012686,
      "flos": 14975935401600.0,
      "grad_norm": 1.9026981064982034,
      "language_loss": 0.75457311,
      "learning_rate": 2.7055160057296424e-07,
      "loss": 0.77637124,
      "num_input_tokens_seen": 150456155,
      "step": 6965,
      "time_per_iteration": 2.4502720832824707
    },
    {
      "auxiliary_loss_clip": 0.01123433,
      "auxiliary_loss_mlp": 0.01028203,
      "balance_loss_clip": 1.04403269,
      "balance_loss_mlp": 1.02098513,
      "epoch": 0.8376119761919076,
      "flos": 30331839847680.0,
      "grad_norm": 1.5840589112968788,
      "language_loss": 0.72228789,
      "learning_rate": 2.7016049558057896e-07,
      "loss": 0.74380422,
      "num_input_tokens_seen": 150478115,
      "step": 6966,
      "time_per_iteration": 2.6322102546691895
    },
    {
      "auxiliary_loss_clip": 0.01150453,
      "auxiliary_loss_mlp": 0.01021675,
      "balance_loss_clip": 1.04693413,
      "balance_loss_mlp": 1.01480246,
      "epoch": 0.8377322190825467,
      "flos": 29423336129280.0,
      "grad_norm": 1.7121631329994285,
      "language_loss": 0.70734161,
      "learning_rate": 2.6976965300347074e-07,
      "loss": 0.72906291,
      "num_input_tokens_seen": 150500725,
      "step": 6967,
      "time_per_iteration": 2.5537824630737305
    },
    {
      "auxiliary_loss_clip": 0.01130853,
      "auxiliary_loss_mlp": 0.01027544,
      "balance_loss_clip": 1.04211807,
      "balance_loss_mlp": 1.02081454,
      "epoch": 0.8378524619731859,
      "flos": 26687086807680.0,
      "grad_norm": 2.623288530221996,
      "language_loss": 0.69551104,
      "learning_rate": 2.693790729009309e-07,
      "loss": 0.71709502,
      "num_input_tokens_seen": 150522335,
      "step": 6968,
      "time_per_iteration": 2.5764079093933105
    },
    {
      "auxiliary_loss_clip": 0.01136774,
      "auxiliary_loss_mlp": 0.01024481,
      "balance_loss_clip": 1.04501462,
      "balance_loss_mlp": 1.01781166,
      "epoch": 0.8379727048638249,
      "flos": 20703866636160.0,
      "grad_norm": 1.8445608458355607,
      "language_loss": 0.88643444,
      "learning_rate": 2.6898875533220946e-07,
      "loss": 0.90804702,
      "num_input_tokens_seen": 150541640,
      "step": 6969,
      "time_per_iteration": 2.537450075149536
    },
    {
      "auxiliary_loss_clip": 0.0115787,
      "auxiliary_loss_mlp": 0.01023598,
      "balance_loss_clip": 1.04652369,
      "balance_loss_mlp": 1.01798296,
      "epoch": 0.838092947754464,
      "flos": 20084084438400.0,
      "grad_norm": 1.6668767143367138,
      "language_loss": 0.81733429,
      "learning_rate": 2.685987003565171e-07,
      "loss": 0.839149,
      "num_input_tokens_seen": 150559680,
      "step": 6970,
      "time_per_iteration": 2.467628240585327
    },
    {
      "auxiliary_loss_clip": 0.0111473,
      "auxiliary_loss_mlp": 0.01025713,
      "balance_loss_clip": 1.04528964,
      "balance_loss_mlp": 1.01885259,
      "epoch": 0.8382131906451031,
      "flos": 18113270964480.0,
      "grad_norm": 2.2374621660147356,
      "language_loss": 0.74824572,
      "learning_rate": 2.6820890803302566e-07,
      "loss": 0.76965016,
      "num_input_tokens_seen": 150575205,
      "step": 6971,
      "time_per_iteration": 2.5271248817443848
    },
    {
      "auxiliary_loss_clip": 0.01131001,
      "auxiliary_loss_mlp": 0.01028671,
      "balance_loss_clip": 1.04591012,
      "balance_loss_mlp": 1.02188444,
      "epoch": 0.8383334335357422,
      "flos": 17092653920640.0,
      "grad_norm": 2.1670283431241186,
      "language_loss": 0.81884503,
      "learning_rate": 2.6781937842086557e-07,
      "loss": 0.84044176,
      "num_input_tokens_seen": 150593995,
      "step": 6972,
      "time_per_iteration": 2.507235288619995
    },
    {
      "auxiliary_loss_clip": 0.01151762,
      "auxiliary_loss_mlp": 0.01025713,
      "balance_loss_clip": 1.04600441,
      "balance_loss_mlp": 1.01951706,
      "epoch": 0.8384536764263812,
      "flos": 20704728562560.0,
      "grad_norm": 1.7386135731704802,
      "language_loss": 0.670784,
      "learning_rate": 2.6743011157912933e-07,
      "loss": 0.69255871,
      "num_input_tokens_seen": 150613715,
      "step": 6973,
      "time_per_iteration": 2.4929585456848145
    },
    {
      "auxiliary_loss_clip": 0.01105293,
      "auxiliary_loss_mlp": 0.01029168,
      "balance_loss_clip": 1.03767157,
      "balance_loss_mlp": 1.02221262,
      "epoch": 0.8385739193170204,
      "flos": 28986842056320.0,
      "grad_norm": 1.625313833064709,
      "language_loss": 0.65178227,
      "learning_rate": 2.6704110756686725e-07,
      "loss": 0.67312694,
      "num_input_tokens_seen": 150634540,
      "step": 6974,
      "time_per_iteration": 4.276062726974487
    },
    {
      "auxiliary_loss_clip": 0.01131385,
      "auxiliary_loss_mlp": 0.00760849,
      "balance_loss_clip": 1.04269838,
      "balance_loss_mlp": 1.00028884,
      "epoch": 0.8386941622076595,
      "flos": 23438068882560.0,
      "grad_norm": 2.0240438074934057,
      "language_loss": 0.83553338,
      "learning_rate": 2.6665236644309085e-07,
      "loss": 0.85445571,
      "num_input_tokens_seen": 150654850,
      "step": 6975,
      "time_per_iteration": 2.564086675643921
    },
    {
      "auxiliary_loss_clip": 0.01151874,
      "auxiliary_loss_mlp": 0.01028927,
      "balance_loss_clip": 1.0463798,
      "balance_loss_mlp": 1.02205777,
      "epoch": 0.8388144050982985,
      "flos": 23002724044800.0,
      "grad_norm": 3.0469221625996585,
      "language_loss": 0.79535228,
      "learning_rate": 2.662638882667727e-07,
      "loss": 0.81716025,
      "num_input_tokens_seen": 150673790,
      "step": 6976,
      "time_per_iteration": 3.259716272354126
    },
    {
      "auxiliary_loss_clip": 0.01166989,
      "auxiliary_loss_mlp": 0.01028922,
      "balance_loss_clip": 1.04767907,
      "balance_loss_mlp": 1.02187395,
      "epoch": 0.8389346479889377,
      "flos": 24280353878400.0,
      "grad_norm": 1.986111163350732,
      "language_loss": 0.72994471,
      "learning_rate": 2.658756730968443e-07,
      "loss": 0.75190377,
      "num_input_tokens_seen": 150692255,
      "step": 6977,
      "time_per_iteration": 2.4829554557800293
    },
    {
      "auxiliary_loss_clip": 0.01139097,
      "auxiliary_loss_mlp": 0.01027462,
      "balance_loss_clip": 1.0458889,
      "balance_loss_mlp": 1.02038097,
      "epoch": 0.8390548908795767,
      "flos": 21215019127680.0,
      "grad_norm": 3.259294049439094,
      "language_loss": 0.88360351,
      "learning_rate": 2.654877209921975e-07,
      "loss": 0.90526903,
      "num_input_tokens_seen": 150709790,
      "step": 6978,
      "time_per_iteration": 2.5188701152801514
    },
    {
      "auxiliary_loss_clip": 0.01114379,
      "auxiliary_loss_mlp": 0.01025455,
      "balance_loss_clip": 1.04080653,
      "balance_loss_mlp": 1.01832891,
      "epoch": 0.8391751337702158,
      "flos": 35627299332480.0,
      "grad_norm": 2.3642599401974844,
      "language_loss": 0.62424564,
      "learning_rate": 2.651000320116843e-07,
      "loss": 0.64564395,
      "num_input_tokens_seen": 150730675,
      "step": 6979,
      "time_per_iteration": 2.707624673843384
    },
    {
      "auxiliary_loss_clip": 0.01119548,
      "auxiliary_loss_mlp": 0.00761361,
      "balance_loss_clip": 1.04291844,
      "balance_loss_mlp": 1.00027096,
      "epoch": 0.839295376660855,
      "flos": 21325229032320.0,
      "grad_norm": 2.135165426897522,
      "language_loss": 0.75812829,
      "learning_rate": 2.647126062141163e-07,
      "loss": 0.77693737,
      "num_input_tokens_seen": 150749750,
      "step": 6980,
      "time_per_iteration": 2.5651581287384033
    },
    {
      "auxiliary_loss_clip": 0.01136044,
      "auxiliary_loss_mlp": 0.01020847,
      "balance_loss_clip": 1.03957915,
      "balance_loss_mlp": 1.01393247,
      "epoch": 0.839415619551494,
      "flos": 18442535961600.0,
      "grad_norm": 2.039692373166702,
      "language_loss": 0.83960247,
      "learning_rate": 2.643254436582669e-07,
      "loss": 0.86117136,
      "num_input_tokens_seen": 150769240,
      "step": 6981,
      "time_per_iteration": 2.527170419692993
    },
    {
      "auxiliary_loss_clip": 0.01110751,
      "auxiliary_loss_mlp": 0.01028875,
      "balance_loss_clip": 1.04404676,
      "balance_loss_mlp": 1.02208889,
      "epoch": 0.8395358624421331,
      "flos": 23221958705280.0,
      "grad_norm": 1.7970901710996285,
      "language_loss": 0.822752,
      "learning_rate": 2.6393854440286743e-07,
      "loss": 0.84414822,
      "num_input_tokens_seen": 150788410,
      "step": 6982,
      "time_per_iteration": 2.6289656162261963
    },
    {
      "auxiliary_loss_clip": 0.01165747,
      "auxiliary_loss_mlp": 0.01026242,
      "balance_loss_clip": 1.05024433,
      "balance_loss_mlp": 1.01934004,
      "epoch": 0.8396561053327722,
      "flos": 24381657210240.0,
      "grad_norm": 1.9319821018857597,
      "language_loss": 0.70904201,
      "learning_rate": 2.6355190850661045e-07,
      "loss": 0.73096192,
      "num_input_tokens_seen": 150805245,
      "step": 6983,
      "time_per_iteration": 2.476008176803589
    },
    {
      "auxiliary_loss_clip": 0.01136832,
      "auxiliary_loss_mlp": 0.01027976,
      "balance_loss_clip": 1.04785824,
      "balance_loss_mlp": 1.02049005,
      "epoch": 0.8397763482234113,
      "flos": 22237755073920.0,
      "grad_norm": 1.5001502678213448,
      "language_loss": 0.86568415,
      "learning_rate": 2.631655360281486e-07,
      "loss": 0.88733226,
      "num_input_tokens_seen": 150824920,
      "step": 6984,
      "time_per_iteration": 2.5529427528381348
    },
    {
      "auxiliary_loss_clip": 0.01150487,
      "auxiliary_loss_mlp": 0.00760554,
      "balance_loss_clip": 1.04569066,
      "balance_loss_mlp": 1.00028276,
      "epoch": 0.8398965911140504,
      "flos": 22163743100160.0,
      "grad_norm": 1.8001417429298758,
      "language_loss": 0.65791178,
      "learning_rate": 2.6277942702609323e-07,
      "loss": 0.67702222,
      "num_input_tokens_seen": 150844400,
      "step": 6985,
      "time_per_iteration": 2.507512092590332
    },
    {
      "auxiliary_loss_clip": 0.01122638,
      "auxiliary_loss_mlp": 0.01026048,
      "balance_loss_clip": 1.04323506,
      "balance_loss_mlp": 1.01913953,
      "epoch": 0.8400168340046895,
      "flos": 21542775753600.0,
      "grad_norm": 2.2901404884610743,
      "language_loss": 0.87467116,
      "learning_rate": 2.623935815590186e-07,
      "loss": 0.89615798,
      "num_input_tokens_seen": 150862780,
      "step": 6986,
      "time_per_iteration": 2.560955047607422
    },
    {
      "auxiliary_loss_clip": 0.01137066,
      "auxiliary_loss_mlp": 0.01027455,
      "balance_loss_clip": 1.04619503,
      "balance_loss_mlp": 1.02033782,
      "epoch": 0.8401370768953286,
      "flos": 22491966602880.0,
      "grad_norm": 1.6815694427809096,
      "language_loss": 0.80693913,
      "learning_rate": 2.6200799968545516e-07,
      "loss": 0.82858431,
      "num_input_tokens_seen": 150883075,
      "step": 6987,
      "time_per_iteration": 2.537477970123291
    },
    {
      "auxiliary_loss_clip": 0.01031338,
      "auxiliary_loss_mlp": 0.01002311,
      "balance_loss_clip": 1.00913513,
      "balance_loss_mlp": 1.00133312,
      "epoch": 0.8402573197859676,
      "flos": 59238890818560.0,
      "grad_norm": 0.7852626339366039,
      "language_loss": 0.56480241,
      "learning_rate": 2.616226814638969e-07,
      "loss": 0.58513892,
      "num_input_tokens_seen": 150948180,
      "step": 6988,
      "time_per_iteration": 3.164156675338745
    },
    {
      "auxiliary_loss_clip": 0.0113507,
      "auxiliary_loss_mlp": 0.01024887,
      "balance_loss_clip": 1.04570425,
      "balance_loss_mlp": 1.0181576,
      "epoch": 0.8403775626766068,
      "flos": 22674608282880.0,
      "grad_norm": 1.8976502562124775,
      "language_loss": 0.77533293,
      "learning_rate": 2.612376269527954e-07,
      "loss": 0.79693246,
      "num_input_tokens_seen": 150967885,
      "step": 6989,
      "time_per_iteration": 2.537059783935547
    },
    {
      "auxiliary_loss_clip": 0.01135679,
      "auxiliary_loss_mlp": 0.01025841,
      "balance_loss_clip": 1.04640055,
      "balance_loss_mlp": 1.0184319,
      "epoch": 0.8404978055672458,
      "flos": 19609704495360.0,
      "grad_norm": 1.812978426605329,
      "language_loss": 0.67621547,
      "learning_rate": 2.608528362105635e-07,
      "loss": 0.69783068,
      "num_input_tokens_seen": 150987255,
      "step": 6990,
      "time_per_iteration": 3.2498533725738525
    },
    {
      "auxiliary_loss_clip": 0.0112281,
      "auxiliary_loss_mlp": 0.01022424,
      "balance_loss_clip": 1.04112363,
      "balance_loss_mlp": 1.01577485,
      "epoch": 0.8406180484578849,
      "flos": 27526929678720.0,
      "grad_norm": 1.7529014520591946,
      "language_loss": 0.72991693,
      "learning_rate": 2.6046830929557374e-07,
      "loss": 0.75136924,
      "num_input_tokens_seen": 151006905,
      "step": 6991,
      "time_per_iteration": 2.6162219047546387
    },
    {
      "auxiliary_loss_clip": 0.01117562,
      "auxiliary_loss_mlp": 0.01032111,
      "balance_loss_clip": 1.04320049,
      "balance_loss_mlp": 1.02544379,
      "epoch": 0.8407382913485241,
      "flos": 22127473342080.0,
      "grad_norm": 1.853753502122538,
      "language_loss": 0.84970617,
      "learning_rate": 2.6008404626615776e-07,
      "loss": 0.87120283,
      "num_input_tokens_seen": 151025405,
      "step": 6992,
      "time_per_iteration": 2.6020123958587646
    },
    {
      "auxiliary_loss_clip": 0.01152202,
      "auxiliary_loss_mlp": 0.01029416,
      "balance_loss_clip": 1.0466435,
      "balance_loss_mlp": 1.02219808,
      "epoch": 0.8408585342391631,
      "flos": 13918473982080.0,
      "grad_norm": 17.47952559738663,
      "language_loss": 0.74090397,
      "learning_rate": 2.597000471806092e-07,
      "loss": 0.76272011,
      "num_input_tokens_seen": 151041970,
      "step": 6993,
      "time_per_iteration": 2.483903169631958
    },
    {
      "auxiliary_loss_clip": 0.01132104,
      "auxiliary_loss_mlp": 0.01029893,
      "balance_loss_clip": 1.04742682,
      "balance_loss_mlp": 1.02279747,
      "epoch": 0.8409787771298022,
      "flos": 20187865808640.0,
      "grad_norm": 1.8195560310574774,
      "language_loss": 0.73008859,
      "learning_rate": 2.593163120971793e-07,
      "loss": 0.75170851,
      "num_input_tokens_seen": 151060835,
      "step": 6994,
      "time_per_iteration": 2.5459322929382324
    },
    {
      "auxiliary_loss_clip": 0.01099597,
      "auxiliary_loss_mlp": 0.01024029,
      "balance_loss_clip": 1.03830075,
      "balance_loss_mlp": 1.01717114,
      "epoch": 0.8410990200204413,
      "flos": 23142523777920.0,
      "grad_norm": 1.825510693413091,
      "language_loss": 0.69104493,
      "learning_rate": 2.5893284107408165e-07,
      "loss": 0.71228111,
      "num_input_tokens_seen": 151078205,
      "step": 6995,
      "time_per_iteration": 2.6083624362945557
    },
    {
      "auxiliary_loss_clip": 0.0110601,
      "auxiliary_loss_mlp": 0.01025883,
      "balance_loss_clip": 1.04232466,
      "balance_loss_mlp": 1.01879609,
      "epoch": 0.8412192629110804,
      "flos": 24027219757440.0,
      "grad_norm": 1.7384383751861623,
      "language_loss": 0.77665859,
      "learning_rate": 2.5854963416948726e-07,
      "loss": 0.79797745,
      "num_input_tokens_seen": 151100470,
      "step": 6996,
      "time_per_iteration": 2.657032012939453
    },
    {
      "auxiliary_loss_clip": 0.01103076,
      "auxiliary_loss_mlp": 0.01029306,
      "balance_loss_clip": 1.03765547,
      "balance_loss_mlp": 1.0225563,
      "epoch": 0.8413395058017195,
      "flos": 25591703604480.0,
      "grad_norm": 1.5539344652599512,
      "language_loss": 0.69218338,
      "learning_rate": 2.5816669144152816e-07,
      "loss": 0.71350718,
      "num_input_tokens_seen": 151121650,
      "step": 6997,
      "time_per_iteration": 2.64845871925354
    },
    {
      "auxiliary_loss_clip": 0.01054625,
      "auxiliary_loss_mlp": 0.01002603,
      "balance_loss_clip": 1.00730872,
      "balance_loss_mlp": 1.00153065,
      "epoch": 0.8414597486923585,
      "flos": 63635396624640.0,
      "grad_norm": 0.8510225124165085,
      "language_loss": 0.66283178,
      "learning_rate": 2.5778401294829777e-07,
      "loss": 0.68340409,
      "num_input_tokens_seen": 151180390,
      "step": 6998,
      "time_per_iteration": 3.1462180614471436
    },
    {
      "auxiliary_loss_clip": 0.01146527,
      "auxiliary_loss_mlp": 0.00760768,
      "balance_loss_clip": 1.04504287,
      "balance_loss_mlp": 1.00027204,
      "epoch": 0.8415799915829977,
      "flos": 19098731571840.0,
      "grad_norm": 1.6032223285574714,
      "language_loss": 0.64799929,
      "learning_rate": 2.574015987478473e-07,
      "loss": 0.66707218,
      "num_input_tokens_seen": 151198520,
      "step": 6999,
      "time_per_iteration": 2.479705333709717
    },
    {
      "auxiliary_loss_clip": 0.01140776,
      "auxiliary_loss_mlp": 0.01029307,
      "balance_loss_clip": 1.04542947,
      "balance_loss_mlp": 1.02169847,
      "epoch": 0.8417002344736367,
      "flos": 19821612781440.0,
      "grad_norm": 1.979499922254669,
      "language_loss": 0.867064,
      "learning_rate": 2.570194488981887e-07,
      "loss": 0.8887648,
      "num_input_tokens_seen": 151215065,
      "step": 7000,
      "time_per_iteration": 4.164418697357178
    },
    {
      "auxiliary_loss_clip": 0.01055353,
      "auxiliary_loss_mlp": 0.01002059,
      "balance_loss_clip": 1.00797796,
      "balance_loss_mlp": 1.00099802,
      "epoch": 0.8418204773642758,
      "flos": 62161516834560.0,
      "grad_norm": 0.8381690011365832,
      "language_loss": 0.60351789,
      "learning_rate": 2.566375634572939e-07,
      "loss": 0.62409204,
      "num_input_tokens_seen": 151275705,
      "step": 7001,
      "time_per_iteration": 3.010477066040039
    },
    {
      "auxiliary_loss_clip": 0.01124733,
      "auxiliary_loss_mlp": 0.01026018,
      "balance_loss_clip": 1.04122198,
      "balance_loss_mlp": 1.01847243,
      "epoch": 0.841940720254915,
      "flos": 17092905315840.0,
      "grad_norm": 1.737000420887874,
      "language_loss": 0.75868994,
      "learning_rate": 2.562559424830943e-07,
      "loss": 0.78019744,
      "num_input_tokens_seen": 151293665,
      "step": 7002,
      "time_per_iteration": 3.2193753719329834
    },
    {
      "auxiliary_loss_clip": 0.01130812,
      "auxiliary_loss_mlp": 0.01027336,
      "balance_loss_clip": 1.04252434,
      "balance_loss_mlp": 1.02025175,
      "epoch": 0.842060963145554,
      "flos": 16283586026880.0,
      "grad_norm": 3.9456816207422594,
      "language_loss": 0.70684135,
      "learning_rate": 2.5587458603348256e-07,
      "loss": 0.72842282,
      "num_input_tokens_seen": 151310955,
      "step": 7003,
      "time_per_iteration": 2.498051881790161
    },
    {
      "auxiliary_loss_clip": 0.0111562,
      "auxiliary_loss_mlp": 0.01028601,
      "balance_loss_clip": 1.04244077,
      "balance_loss_mlp": 1.02151728,
      "epoch": 0.8421812060361931,
      "flos": 21908238681600.0,
      "grad_norm": 1.6954764554503827,
      "language_loss": 0.83847845,
      "learning_rate": 2.554934941663085e-07,
      "loss": 0.85992062,
      "num_input_tokens_seen": 151328490,
      "step": 7004,
      "time_per_iteration": 2.5597798824310303
    },
    {
      "auxiliary_loss_clip": 0.01119595,
      "auxiliary_loss_mlp": 0.0102821,
      "balance_loss_clip": 1.04224598,
      "balance_loss_mlp": 1.02104819,
      "epoch": 0.8423014489268322,
      "flos": 27777693502080.0,
      "grad_norm": 2.554371009659951,
      "language_loss": 0.73230577,
      "learning_rate": 2.5511266693938484e-07,
      "loss": 0.75378382,
      "num_input_tokens_seen": 151346950,
      "step": 7005,
      "time_per_iteration": 2.5962836742401123
    },
    {
      "auxiliary_loss_clip": 0.01133269,
      "auxiliary_loss_mlp": 0.01031067,
      "balance_loss_clip": 1.04686713,
      "balance_loss_mlp": 1.02380085,
      "epoch": 0.8424216918174713,
      "flos": 25117610970240.0,
      "grad_norm": 4.548872087613753,
      "language_loss": 0.78021085,
      "learning_rate": 2.547321044104822e-07,
      "loss": 0.80185419,
      "num_input_tokens_seen": 151368445,
      "step": 7006,
      "time_per_iteration": 2.5657424926757812
    },
    {
      "auxiliary_loss_clip": 0.01166368,
      "auxiliary_loss_mlp": 0.01025259,
      "balance_loss_clip": 1.04693484,
      "balance_loss_mlp": 1.01806521,
      "epoch": 0.8425419347081103,
      "flos": 24748448941440.0,
      "grad_norm": 1.9297870857787616,
      "language_loss": 0.76751721,
      "learning_rate": 2.5435180663733113e-07,
      "loss": 0.78943348,
      "num_input_tokens_seen": 151388745,
      "step": 7007,
      "time_per_iteration": 2.4974164962768555
    },
    {
      "auxiliary_loss_clip": 0.01114552,
      "auxiliary_loss_mlp": 0.01026378,
      "balance_loss_clip": 1.04221153,
      "balance_loss_mlp": 1.01952338,
      "epoch": 0.8426621775987495,
      "flos": 24820916630400.0,
      "grad_norm": 2.239193693315536,
      "language_loss": 0.71564305,
      "learning_rate": 2.539717736776241e-07,
      "loss": 0.73705238,
      "num_input_tokens_seen": 151404970,
      "step": 7008,
      "time_per_iteration": 2.6135783195495605
    },
    {
      "auxiliary_loss_clip": 0.0114543,
      "auxiliary_loss_mlp": 0.01024085,
      "balance_loss_clip": 1.04641604,
      "balance_loss_mlp": 1.01737928,
      "epoch": 0.8427824204893886,
      "flos": 23550074467200.0,
      "grad_norm": 1.3747878315919682,
      "language_loss": 0.76398373,
      "learning_rate": 2.535920055890097e-07,
      "loss": 0.78567892,
      "num_input_tokens_seen": 151426265,
      "step": 7009,
      "time_per_iteration": 2.495159387588501
    },
    {
      "auxiliary_loss_clip": 0.01100293,
      "auxiliary_loss_mlp": 0.01031369,
      "balance_loss_clip": 1.0378902,
      "balance_loss_mlp": 1.02379322,
      "epoch": 0.8429026633800276,
      "flos": 16143858120960.0,
      "grad_norm": 6.839989645619045,
      "language_loss": 0.64721787,
      "learning_rate": 2.5321250242910006e-07,
      "loss": 0.66853452,
      "num_input_tokens_seen": 151444180,
      "step": 7010,
      "time_per_iteration": 2.570758581161499
    },
    {
      "auxiliary_loss_clip": 0.01166276,
      "auxiliary_loss_mlp": 0.01030225,
      "balance_loss_clip": 1.04984272,
      "balance_loss_mlp": 1.02300739,
      "epoch": 0.8430229062706668,
      "flos": 22198540400640.0,
      "grad_norm": 1.6143572718132144,
      "language_loss": 0.86411989,
      "learning_rate": 2.5283326425546493e-07,
      "loss": 0.88608491,
      "num_input_tokens_seen": 151463290,
      "step": 7011,
      "time_per_iteration": 2.4638795852661133
    },
    {
      "auxiliary_loss_clip": 0.01114978,
      "auxiliary_loss_mlp": 0.01024973,
      "balance_loss_clip": 1.04535556,
      "balance_loss_mlp": 1.01855075,
      "epoch": 0.8431431491613058,
      "flos": 35330317683840.0,
      "grad_norm": 1.8584907344756423,
      "language_loss": 0.69150162,
      "learning_rate": 2.5245429112563443e-07,
      "loss": 0.71290123,
      "num_input_tokens_seen": 151483965,
      "step": 7012,
      "time_per_iteration": 2.706101179122925
    },
    {
      "auxiliary_loss_clip": 0.01150682,
      "auxiliary_loss_mlp": 0.01023798,
      "balance_loss_clip": 1.0474236,
      "balance_loss_mlp": 1.01654673,
      "epoch": 0.8432633920519449,
      "flos": 25812374808960.0,
      "grad_norm": 1.731550835977976,
      "language_loss": 0.81836396,
      "learning_rate": 2.5207558309709865e-07,
      "loss": 0.84010875,
      "num_input_tokens_seen": 151503700,
      "step": 7013,
      "time_per_iteration": 2.5519516468048096
    },
    {
      "auxiliary_loss_clip": 0.01029451,
      "auxiliary_loss_mlp": 0.00751231,
      "balance_loss_clip": 1.00794518,
      "balance_loss_mlp": 1.00008631,
      "epoch": 0.8433836349425841,
      "flos": 64959531592320.0,
      "grad_norm": 0.6555151056694961,
      "language_loss": 0.56296849,
      "learning_rate": 2.516971402273065e-07,
      "loss": 0.58077532,
      "num_input_tokens_seen": 151569765,
      "step": 7014,
      "time_per_iteration": 3.1411688327789307
    },
    {
      "auxiliary_loss_clip": 0.01134718,
      "auxiliary_loss_mlp": 0.01023173,
      "balance_loss_clip": 1.04217124,
      "balance_loss_mlp": 1.01653337,
      "epoch": 0.8435038778332231,
      "flos": 20229989483520.0,
      "grad_norm": 1.8006871294143818,
      "language_loss": 0.67680132,
      "learning_rate": 2.513189625736687e-07,
      "loss": 0.69838023,
      "num_input_tokens_seen": 151586660,
      "step": 7015,
      "time_per_iteration": 3.345205783843994
    },
    {
      "auxiliary_loss_clip": 0.01126466,
      "auxiliary_loss_mlp": 0.01029034,
      "balance_loss_clip": 1.04349232,
      "balance_loss_mlp": 1.02222729,
      "epoch": 0.8436241207238622,
      "flos": 20992229020800.0,
      "grad_norm": 2.1454323822711396,
      "language_loss": 0.71690416,
      "learning_rate": 2.509410501935534e-07,
      "loss": 0.73845911,
      "num_input_tokens_seen": 151602295,
      "step": 7016,
      "time_per_iteration": 2.5729012489318848
    },
    {
      "auxiliary_loss_clip": 0.01138581,
      "auxiliary_loss_mlp": 0.01030538,
      "balance_loss_clip": 1.04524183,
      "balance_loss_mlp": 1.02317429,
      "epoch": 0.8437443636145013,
      "flos": 14682257804160.0,
      "grad_norm": 2.1114331325008644,
      "language_loss": 0.75183892,
      "learning_rate": 2.5056340314429116e-07,
      "loss": 0.77353007,
      "num_input_tokens_seen": 151619760,
      "step": 7017,
      "time_per_iteration": 2.506105661392212
    },
    {
      "auxiliary_loss_clip": 0.01109604,
      "auxiliary_loss_mlp": 0.01027024,
      "balance_loss_clip": 1.04052138,
      "balance_loss_mlp": 1.01945376,
      "epoch": 0.8438646065051404,
      "flos": 21608814908160.0,
      "grad_norm": 2.3062927031070513,
      "language_loss": 0.80303901,
      "learning_rate": 2.5018602148316904e-07,
      "loss": 0.82440531,
      "num_input_tokens_seen": 151635795,
      "step": 7018,
      "time_per_iteration": 2.5797815322875977
    },
    {
      "auxiliary_loss_clip": 0.01163124,
      "auxiliary_loss_mlp": 0.01026754,
      "balance_loss_clip": 1.04898906,
      "balance_loss_mlp": 1.02003682,
      "epoch": 0.8439848493957794,
      "flos": 23289937194240.0,
      "grad_norm": 1.6929395366675075,
      "language_loss": 0.80287713,
      "learning_rate": 2.498089052674359e-07,
      "loss": 0.82477587,
      "num_input_tokens_seen": 151653770,
      "step": 7019,
      "time_per_iteration": 2.48478627204895
    },
    {
      "auxiliary_loss_clip": 0.01153389,
      "auxiliary_loss_mlp": 0.01029652,
      "balance_loss_clip": 1.04929554,
      "balance_loss_mlp": 1.02255869,
      "epoch": 0.8441050922864186,
      "flos": 19719339782400.0,
      "grad_norm": 2.0045510755493563,
      "language_loss": 0.74957693,
      "learning_rate": 2.494320545543007e-07,
      "loss": 0.77140731,
      "num_input_tokens_seen": 151673340,
      "step": 7020,
      "time_per_iteration": 2.4795982837677
    },
    {
      "auxiliary_loss_clip": 0.01165762,
      "auxiliary_loss_mlp": 0.0102778,
      "balance_loss_clip": 1.04704046,
      "balance_loss_mlp": 1.02024007,
      "epoch": 0.8442253351770577,
      "flos": 21835268202240.0,
      "grad_norm": 1.6386687250603638,
      "language_loss": 0.6634106,
      "learning_rate": 2.490554694009308e-07,
      "loss": 0.68534601,
      "num_input_tokens_seen": 151694205,
      "step": 7021,
      "time_per_iteration": 2.4592936038970947
    },
    {
      "auxiliary_loss_clip": 0.01152867,
      "auxiliary_loss_mlp": 0.0102132,
      "balance_loss_clip": 1.04588926,
      "balance_loss_mlp": 1.01412022,
      "epoch": 0.8443455780676967,
      "flos": 34346365447680.0,
      "grad_norm": 1.587720690628415,
      "language_loss": 0.78538877,
      "learning_rate": 2.4867914986445426e-07,
      "loss": 0.80713063,
      "num_input_tokens_seen": 151716595,
      "step": 7022,
      "time_per_iteration": 2.609590530395508
    },
    {
      "auxiliary_loss_clip": 0.01137732,
      "auxiliary_loss_mlp": 0.01026439,
      "balance_loss_clip": 1.04227877,
      "balance_loss_mlp": 1.02004385,
      "epoch": 0.8444658209583359,
      "flos": 48214599281280.0,
      "grad_norm": 1.8194855038893754,
      "language_loss": 0.71578121,
      "learning_rate": 2.483030960019581e-07,
      "loss": 0.73742294,
      "num_input_tokens_seen": 151740525,
      "step": 7023,
      "time_per_iteration": 2.7668371200561523
    },
    {
      "auxiliary_loss_clip": 0.01013014,
      "auxiliary_loss_mlp": 0.01001497,
      "balance_loss_clip": 1.00719965,
      "balance_loss_mlp": 1.00043571,
      "epoch": 0.8445860638489749,
      "flos": 68484773105280.0,
      "grad_norm": 0.7280118113912966,
      "language_loss": 0.55500335,
      "learning_rate": 2.479273078704891e-07,
      "loss": 0.57514846,
      "num_input_tokens_seen": 151793890,
      "step": 7024,
      "time_per_iteration": 3.0091805458068848
    },
    {
      "auxiliary_loss_clip": 0.01010277,
      "auxiliary_loss_mlp": 0.01001638,
      "balance_loss_clip": 1.01106954,
      "balance_loss_mlp": 1.00061846,
      "epoch": 0.844706306739614,
      "flos": 62833331882880.0,
      "grad_norm": 0.8348432571624139,
      "language_loss": 0.64742815,
      "learning_rate": 2.475517855270552e-07,
      "loss": 0.66754723,
      "num_input_tokens_seen": 151853970,
      "step": 7025,
      "time_per_iteration": 3.148796319961548
    },
    {
      "auxiliary_loss_clip": 0.01162287,
      "auxiliary_loss_mlp": 0.01020633,
      "balance_loss_clip": 1.04814935,
      "balance_loss_mlp": 1.01378798,
      "epoch": 0.8448265496302532,
      "flos": 14976114969600.0,
      "grad_norm": 2.0738376566496264,
      "language_loss": 0.7261945,
      "learning_rate": 2.4717652902862143e-07,
      "loss": 0.74802375,
      "num_input_tokens_seen": 151872945,
      "step": 7026,
      "time_per_iteration": 4.201322555541992
    },
    {
      "auxiliary_loss_clip": 0.01136213,
      "auxiliary_loss_mlp": 0.01025281,
      "balance_loss_clip": 1.04452872,
      "balance_loss_mlp": 1.01866794,
      "epoch": 0.8449467925208922,
      "flos": 23441265192960.0,
      "grad_norm": 1.7041507385838286,
      "language_loss": 0.81385761,
      "learning_rate": 2.4680153843211495e-07,
      "loss": 0.83547258,
      "num_input_tokens_seen": 151892875,
      "step": 7027,
      "time_per_iteration": 2.564302921295166
    },
    {
      "auxiliary_loss_clip": 0.01138272,
      "auxiliary_loss_mlp": 0.01025404,
      "balance_loss_clip": 1.04995883,
      "balance_loss_mlp": 1.01816821,
      "epoch": 0.8450670354115313,
      "flos": 22748045639040.0,
      "grad_norm": 2.0057053338902007,
      "language_loss": 0.72523695,
      "learning_rate": 2.464268137944212e-07,
      "loss": 0.74687374,
      "num_input_tokens_seen": 151914170,
      "step": 7028,
      "time_per_iteration": 3.2892391681671143
    },
    {
      "auxiliary_loss_clip": 0.01097653,
      "auxiliary_loss_mlp": 0.0103043,
      "balance_loss_clip": 1.04070318,
      "balance_loss_mlp": 1.02284515,
      "epoch": 0.8451872783021703,
      "flos": 29825571605760.0,
      "grad_norm": 1.9288888085736424,
      "language_loss": 0.77964038,
      "learning_rate": 2.46052355172385e-07,
      "loss": 0.8009212,
      "num_input_tokens_seen": 151932210,
      "step": 7029,
      "time_per_iteration": 2.6318960189819336
    },
    {
      "auxiliary_loss_clip": 0.01165006,
      "auxiliary_loss_mlp": 0.01023845,
      "balance_loss_clip": 1.04849279,
      "balance_loss_mlp": 1.01617408,
      "epoch": 0.8453075211928095,
      "flos": 21870029589120.0,
      "grad_norm": 1.7383339214581182,
      "language_loss": 0.74746686,
      "learning_rate": 2.456781626228128e-07,
      "loss": 0.7693553,
      "num_input_tokens_seen": 151951715,
      "step": 7030,
      "time_per_iteration": 2.4987714290618896
    },
    {
      "auxiliary_loss_clip": 0.01012842,
      "auxiliary_loss_mlp": 0.00751457,
      "balance_loss_clip": 1.00582576,
      "balance_loss_mlp": 1.00016582,
      "epoch": 0.8454277640834486,
      "flos": 58751869288320.0,
      "grad_norm": 0.9131311815935615,
      "language_loss": 0.66400921,
      "learning_rate": 2.453042362024675e-07,
      "loss": 0.68165219,
      "num_input_tokens_seen": 152004960,
      "step": 7031,
      "time_per_iteration": 3.1902053356170654
    },
    {
      "auxiliary_loss_clip": 0.01161224,
      "auxiliary_loss_mlp": 0.01025387,
      "balance_loss_clip": 1.04605722,
      "balance_loss_mlp": 1.0185684,
      "epoch": 0.8455480069740876,
      "flos": 27090076469760.0,
      "grad_norm": 1.4084345818115724,
      "language_loss": 0.73261273,
      "learning_rate": 2.449305759680751e-07,
      "loss": 0.75447881,
      "num_input_tokens_seen": 152026285,
      "step": 7032,
      "time_per_iteration": 2.5161941051483154
    },
    {
      "auxiliary_loss_clip": 0.01121454,
      "auxiliary_loss_mlp": 0.0102942,
      "balance_loss_clip": 1.04635882,
      "balance_loss_mlp": 1.02243781,
      "epoch": 0.8456682498647268,
      "flos": 27198670262400.0,
      "grad_norm": 1.4357030782437141,
      "language_loss": 0.74947762,
      "learning_rate": 2.445571819763188e-07,
      "loss": 0.77098644,
      "num_input_tokens_seen": 152048585,
      "step": 7033,
      "time_per_iteration": 2.61283802986145
    },
    {
      "auxiliary_loss_clip": 0.01162799,
      "auxiliary_loss_mlp": 0.01021848,
      "balance_loss_clip": 1.04870331,
      "balance_loss_mlp": 1.01515985,
      "epoch": 0.8457884927553658,
      "flos": 20631901737600.0,
      "grad_norm": 1.5358165758947169,
      "language_loss": 0.58248508,
      "learning_rate": 2.4418405428384227e-07,
      "loss": 0.60433155,
      "num_input_tokens_seen": 152068795,
      "step": 7034,
      "time_per_iteration": 2.4701952934265137
    },
    {
      "auxiliary_loss_clip": 0.01162756,
      "auxiliary_loss_mlp": 0.00761086,
      "balance_loss_clip": 1.04741859,
      "balance_loss_mlp": 1.00031829,
      "epoch": 0.8459087356460049,
      "flos": 15299023259520.0,
      "grad_norm": 2.045551206684969,
      "language_loss": 0.7159878,
      "learning_rate": 2.4381119294724864e-07,
      "loss": 0.73522627,
      "num_input_tokens_seen": 152086240,
      "step": 7035,
      "time_per_iteration": 2.443713426589966
    },
    {
      "auxiliary_loss_clip": 0.01162138,
      "auxiliary_loss_mlp": 0.01030113,
      "balance_loss_clip": 1.04654229,
      "balance_loss_mlp": 1.02339518,
      "epoch": 0.846028978536644,
      "flos": 18843155326080.0,
      "grad_norm": 2.1927478290687445,
      "language_loss": 0.5371564,
      "learning_rate": 2.434385980231004e-07,
      "loss": 0.55907887,
      "num_input_tokens_seen": 152105080,
      "step": 7036,
      "time_per_iteration": 2.434847116470337
    },
    {
      "auxiliary_loss_clip": 0.01148984,
      "auxiliary_loss_mlp": 0.01025758,
      "balance_loss_clip": 1.04554188,
      "balance_loss_mlp": 1.01894808,
      "epoch": 0.8461492214272831,
      "flos": 52661740285440.0,
      "grad_norm": 1.4537488929315034,
      "language_loss": 0.65519053,
      "learning_rate": 2.4306626956792043e-07,
      "loss": 0.67693794,
      "num_input_tokens_seen": 152130025,
      "step": 7037,
      "time_per_iteration": 2.768519639968872
    },
    {
      "auxiliary_loss_clip": 0.01148924,
      "auxiliary_loss_mlp": 0.01026475,
      "balance_loss_clip": 1.04461312,
      "balance_loss_mlp": 1.01970053,
      "epoch": 0.8462694643179222,
      "flos": 18588405093120.0,
      "grad_norm": 1.659560725816226,
      "language_loss": 0.75380051,
      "learning_rate": 2.4269420763819017e-07,
      "loss": 0.77555448,
      "num_input_tokens_seen": 152148070,
      "step": 7038,
      "time_per_iteration": 2.4677011966705322
    },
    {
      "auxiliary_loss_clip": 0.01147479,
      "auxiliary_loss_mlp": 0.01023134,
      "balance_loss_clip": 1.04598999,
      "balance_loss_mlp": 1.0165863,
      "epoch": 0.8463897072085613,
      "flos": 24387080163840.0,
      "grad_norm": 3.2460956279533297,
      "language_loss": 0.83223504,
      "learning_rate": 2.4232241229035223e-07,
      "loss": 0.8539412,
      "num_input_tokens_seen": 152165825,
      "step": 7039,
      "time_per_iteration": 2.499114990234375
    },
    {
      "auxiliary_loss_clip": 0.01045384,
      "auxiliary_loss_mlp": 0.01001906,
      "balance_loss_clip": 1.00716627,
      "balance_loss_mlp": 1.00082719,
      "epoch": 0.8465099500992004,
      "flos": 68702140258560.0,
      "grad_norm": 0.7534895916985822,
      "language_loss": 0.5677762,
      "learning_rate": 2.419508835808064e-07,
      "loss": 0.58824909,
      "num_input_tokens_seen": 152222380,
      "step": 7040,
      "time_per_iteration": 3.0388760566711426
    },
    {
      "auxiliary_loss_clip": 0.01135424,
      "auxiliary_loss_mlp": 0.01026383,
      "balance_loss_clip": 1.04432189,
      "balance_loss_mlp": 1.01904607,
      "epoch": 0.8466301929898394,
      "flos": 13735724561280.0,
      "grad_norm": 2.1848797883762376,
      "language_loss": 0.62629986,
      "learning_rate": 2.415796215659134e-07,
      "loss": 0.64791787,
      "num_input_tokens_seen": 152239085,
      "step": 7041,
      "time_per_iteration": 3.2734925746917725
    },
    {
      "auxiliary_loss_clip": 0.01122624,
      "auxiliary_loss_mlp": 0.01027326,
      "balance_loss_clip": 1.0389502,
      "balance_loss_mlp": 1.01997137,
      "epoch": 0.8467504358804786,
      "flos": 19241260738560.0,
      "grad_norm": 1.9983476178858195,
      "language_loss": 0.77301615,
      "learning_rate": 2.412086263019939e-07,
      "loss": 0.79451561,
      "num_input_tokens_seen": 152257110,
      "step": 7042,
      "time_per_iteration": 2.5249366760253906
    },
    {
      "auxiliary_loss_clip": 0.01160432,
      "auxiliary_loss_mlp": 0.01022487,
      "balance_loss_clip": 1.04791713,
      "balance_loss_mlp": 1.01578152,
      "epoch": 0.8468706787711177,
      "flos": 21324115710720.0,
      "grad_norm": 1.669404455405343,
      "language_loss": 0.7988373,
      "learning_rate": 2.408378978453276e-07,
      "loss": 0.82066655,
      "num_input_tokens_seen": 152277230,
      "step": 7043,
      "time_per_iteration": 2.4815549850463867
    },
    {
      "auxiliary_loss_clip": 0.01046007,
      "auxiliary_loss_mlp": 0.01000569,
      "balance_loss_clip": 1.00787294,
      "balance_loss_mlp": 0.99939519,
      "epoch": 0.8469909216617567,
      "flos": 64877439058560.0,
      "grad_norm": 0.8087358708888047,
      "language_loss": 0.63986182,
      "learning_rate": 2.404674362521533e-07,
      "loss": 0.66032761,
      "num_input_tokens_seen": 152335725,
      "step": 7044,
      "time_per_iteration": 2.985466957092285
    },
    {
      "auxiliary_loss_clip": 0.01150483,
      "auxiliary_loss_mlp": 0.01029657,
      "balance_loss_clip": 1.04832315,
      "balance_loss_mlp": 1.02287388,
      "epoch": 0.8471111645523959,
      "flos": 19280583152640.0,
      "grad_norm": 2.256584762813342,
      "language_loss": 0.74957973,
      "learning_rate": 2.4009724157866997e-07,
      "loss": 0.77138114,
      "num_input_tokens_seen": 152352785,
      "step": 7045,
      "time_per_iteration": 2.480224370956421
    },
    {
      "auxiliary_loss_clip": 0.01161367,
      "auxiliary_loss_mlp": 0.01022086,
      "balance_loss_clip": 1.04672194,
      "balance_loss_mlp": 1.01561034,
      "epoch": 0.8472314074430349,
      "flos": 22015826893440.0,
      "grad_norm": 1.7770821919531326,
      "language_loss": 0.76037455,
      "learning_rate": 2.3972731388103564e-07,
      "loss": 0.78220916,
      "num_input_tokens_seen": 152371265,
      "step": 7046,
      "time_per_iteration": 2.462667465209961
    },
    {
      "auxiliary_loss_clip": 0.00999339,
      "auxiliary_loss_mlp": 0.01000575,
      "balance_loss_clip": 1.00755239,
      "balance_loss_mlp": 0.9994548,
      "epoch": 0.847351650333674,
      "flos": 57882580243200.0,
      "grad_norm": 0.7982031210553313,
      "language_loss": 0.62391883,
      "learning_rate": 2.393576532153687e-07,
      "loss": 0.64391798,
      "num_input_tokens_seen": 152435050,
      "step": 7047,
      "time_per_iteration": 3.3935883045196533
    },
    {
      "auxiliary_loss_clip": 0.01043749,
      "auxiliary_loss_mlp": 0.01002015,
      "balance_loss_clip": 1.00769734,
      "balance_loss_mlp": 1.0008173,
      "epoch": 0.8474718932243132,
      "flos": 41284238313600.0,
      "grad_norm": 0.9258930230809921,
      "language_loss": 0.57809734,
      "learning_rate": 2.389882596377453e-07,
      "loss": 0.59855509,
      "num_input_tokens_seen": 152489315,
      "step": 7048,
      "time_per_iteration": 3.1417064666748047
    },
    {
      "auxiliary_loss_clip": 0.01160985,
      "auxiliary_loss_mlp": 0.01032963,
      "balance_loss_clip": 1.04558516,
      "balance_loss_mlp": 1.02586985,
      "epoch": 0.8475921361149522,
      "flos": 38180906974080.0,
      "grad_norm": 3.301794662411009,
      "language_loss": 0.7636615,
      "learning_rate": 2.386191332042031e-07,
      "loss": 0.78560102,
      "num_input_tokens_seen": 152511210,
      "step": 7049,
      "time_per_iteration": 2.6277542114257812
    },
    {
      "auxiliary_loss_clip": 0.01166713,
      "auxiliary_loss_mlp": 0.0103711,
      "balance_loss_clip": 1.04808497,
      "balance_loss_mlp": 1.02935863,
      "epoch": 0.8477123790055913,
      "flos": 25375054723200.0,
      "grad_norm": 1.6855541986123743,
      "language_loss": 0.7289989,
      "learning_rate": 2.3825027397073794e-07,
      "loss": 0.75103712,
      "num_input_tokens_seen": 152531685,
      "step": 7050,
      "time_per_iteration": 2.483851909637451
    },
    {
      "auxiliary_loss_clip": 0.01147056,
      "auxiliary_loss_mlp": 0.01029891,
      "balance_loss_clip": 1.04834414,
      "balance_loss_mlp": 1.02319181,
      "epoch": 0.8478326218962304,
      "flos": 30225185389440.0,
      "grad_norm": 1.972068967879117,
      "language_loss": 0.67121673,
      "learning_rate": 2.3788168199330515e-07,
      "loss": 0.69298625,
      "num_input_tokens_seen": 152553245,
      "step": 7051,
      "time_per_iteration": 2.540411949157715
    },
    {
      "auxiliary_loss_clip": 0.01120541,
      "auxiliary_loss_mlp": 0.01025456,
      "balance_loss_clip": 1.03946936,
      "balance_loss_mlp": 1.01873851,
      "epoch": 0.8479528647868695,
      "flos": 38213800853760.0,
      "grad_norm": 1.521415416204912,
      "language_loss": 0.72545052,
      "learning_rate": 2.3751335732782074e-07,
      "loss": 0.74691051,
      "num_input_tokens_seen": 152574505,
      "step": 7052,
      "time_per_iteration": 4.297223806381226
    },
    {
      "auxiliary_loss_clip": 0.01147807,
      "auxiliary_loss_mlp": 0.01033388,
      "balance_loss_clip": 1.0464921,
      "balance_loss_mlp": 1.02627754,
      "epoch": 0.8480731076775085,
      "flos": 20957790856320.0,
      "grad_norm": 1.7981743767137541,
      "language_loss": 0.7978012,
      "learning_rate": 2.371453000301582e-07,
      "loss": 0.8196131,
      "num_input_tokens_seen": 152593190,
      "step": 7053,
      "time_per_iteration": 2.476506233215332
    },
    {
      "auxiliary_loss_clip": 0.01116333,
      "auxiliary_loss_mlp": 0.01023961,
      "balance_loss_clip": 1.0426774,
      "balance_loss_mlp": 1.01759243,
      "epoch": 0.8481933505681477,
      "flos": 32596510487040.0,
      "grad_norm": 1.8707400271106658,
      "language_loss": 0.74233961,
      "learning_rate": 2.3677751015615222e-07,
      "loss": 0.76374257,
      "num_input_tokens_seen": 152615265,
      "step": 7054,
      "time_per_iteration": 3.3688371181488037
    },
    {
      "auxiliary_loss_clip": 0.01123615,
      "auxiliary_loss_mlp": 0.0103091,
      "balance_loss_clip": 1.04208183,
      "balance_loss_mlp": 1.02358425,
      "epoch": 0.8483135934587868,
      "flos": 20741177888640.0,
      "grad_norm": 1.729974346393647,
      "language_loss": 0.85372764,
      "learning_rate": 2.3640998776159593e-07,
      "loss": 0.87527287,
      "num_input_tokens_seen": 152632770,
      "step": 7055,
      "time_per_iteration": 2.5133745670318604
    },
    {
      "auxiliary_loss_clip": 0.0113756,
      "auxiliary_loss_mlp": 0.01027302,
      "balance_loss_clip": 1.04639328,
      "balance_loss_mlp": 1.02125764,
      "epoch": 0.8484338363494258,
      "flos": 21653057485440.0,
      "grad_norm": 1.7749461873710313,
      "language_loss": 0.80992484,
      "learning_rate": 2.3604273290224253e-07,
      "loss": 0.83157349,
      "num_input_tokens_seen": 152653485,
      "step": 7056,
      "time_per_iteration": 2.536184549331665
    },
    {
      "auxiliary_loss_clip": 0.01137878,
      "auxiliary_loss_mlp": 0.01030906,
      "balance_loss_clip": 1.04687786,
      "balance_loss_mlp": 1.02319288,
      "epoch": 0.848554079240065,
      "flos": 15013964926080.0,
      "grad_norm": 1.8422821079901068,
      "language_loss": 0.74555367,
      "learning_rate": 2.356757456338039e-07,
      "loss": 0.76724154,
      "num_input_tokens_seen": 152670970,
      "step": 7057,
      "time_per_iteration": 2.5079658031463623
    },
    {
      "auxiliary_loss_clip": 0.01035205,
      "auxiliary_loss_mlp": 0.01000904,
      "balance_loss_clip": 1.01215971,
      "balance_loss_mlp": 0.99990863,
      "epoch": 0.848674322130704,
      "flos": 68060453742720.0,
      "grad_norm": 0.7519293817279145,
      "language_loss": 0.59058273,
      "learning_rate": 2.3530902601195147e-07,
      "loss": 0.61094391,
      "num_input_tokens_seen": 152739460,
      "step": 7058,
      "time_per_iteration": 3.2415740489959717
    },
    {
      "auxiliary_loss_clip": 0.01148962,
      "auxiliary_loss_mlp": 0.01031584,
      "balance_loss_clip": 1.04750443,
      "balance_loss_mlp": 1.02274418,
      "epoch": 0.8487945650213431,
      "flos": 18475788977280.0,
      "grad_norm": 2.2207044979595563,
      "language_loss": 0.78925228,
      "learning_rate": 2.34942574092317e-07,
      "loss": 0.81105775,
      "num_input_tokens_seen": 152754710,
      "step": 7059,
      "time_per_iteration": 2.4910688400268555
    },
    {
      "auxiliary_loss_clip": 0.01155461,
      "auxiliary_loss_mlp": 0.01029793,
      "balance_loss_clip": 1.04897904,
      "balance_loss_mlp": 1.02297425,
      "epoch": 0.8489148079119821,
      "flos": 23473189405440.0,
      "grad_norm": 2.121203528887361,
      "language_loss": 0.7660414,
      "learning_rate": 2.3457638993049045e-07,
      "loss": 0.78789395,
      "num_input_tokens_seen": 152772700,
      "step": 7060,
      "time_per_iteration": 2.5215368270874023
    },
    {
      "auxiliary_loss_clip": 0.01095489,
      "auxiliary_loss_mlp": 0.01027832,
      "balance_loss_clip": 1.04227948,
      "balance_loss_mlp": 1.02031589,
      "epoch": 0.8490350508026213,
      "flos": 19937604775680.0,
      "grad_norm": 1.9749433250007153,
      "language_loss": 0.64293337,
      "learning_rate": 2.3421047358202252e-07,
      "loss": 0.66416663,
      "num_input_tokens_seen": 152791550,
      "step": 7061,
      "time_per_iteration": 2.6150450706481934
    },
    {
      "auxiliary_loss_clip": 0.01154046,
      "auxiliary_loss_mlp": 0.01027205,
      "balance_loss_clip": 1.04828858,
      "balance_loss_mlp": 1.01988888,
      "epoch": 0.8491552936932604,
      "flos": 24279958828800.0,
      "grad_norm": 2.242554804561456,
      "language_loss": 0.83037889,
      "learning_rate": 2.3384482510242144e-07,
      "loss": 0.85219139,
      "num_input_tokens_seen": 152809410,
      "step": 7062,
      "time_per_iteration": 2.5155465602874756
    },
    {
      "auxiliary_loss_clip": 0.0116319,
      "auxiliary_loss_mlp": 0.01028625,
      "balance_loss_clip": 1.04504943,
      "balance_loss_mlp": 1.02116513,
      "epoch": 0.8492755365838994,
      "flos": 22522526098560.0,
      "grad_norm": 2.167255765375438,
      "language_loss": 0.77254295,
      "learning_rate": 2.3347944454715575e-07,
      "loss": 0.79446113,
      "num_input_tokens_seen": 152825800,
      "step": 7063,
      "time_per_iteration": 2.466905117034912
    },
    {
      "auxiliary_loss_clip": 0.0116379,
      "auxiliary_loss_mlp": 0.01027141,
      "balance_loss_clip": 1.04659986,
      "balance_loss_mlp": 1.01984859,
      "epoch": 0.8493957794745386,
      "flos": 26980441182720.0,
      "grad_norm": 1.8733654165639655,
      "language_loss": 0.66928327,
      "learning_rate": 2.331143319716542e-07,
      "loss": 0.69119257,
      "num_input_tokens_seen": 152845330,
      "step": 7064,
      "time_per_iteration": 2.513002395629883
    },
    {
      "auxiliary_loss_clip": 0.01120894,
      "auxiliary_loss_mlp": 0.01021942,
      "balance_loss_clip": 1.04270148,
      "balance_loss_mlp": 1.01462865,
      "epoch": 0.8495160223651776,
      "flos": 29861985018240.0,
      "grad_norm": 2.2038569048485135,
      "language_loss": 0.66241032,
      "learning_rate": 2.3274948743130363e-07,
      "loss": 0.68383873,
      "num_input_tokens_seen": 152865165,
      "step": 7065,
      "time_per_iteration": 2.6240925788879395
    },
    {
      "auxiliary_loss_clip": 0.01161425,
      "auxiliary_loss_mlp": 0.0102316,
      "balance_loss_clip": 1.04571021,
      "balance_loss_mlp": 1.01671994,
      "epoch": 0.8496362652558167,
      "flos": 23075443128960.0,
      "grad_norm": 1.5622522385818285,
      "language_loss": 0.79262614,
      "learning_rate": 2.3238491098145085e-07,
      "loss": 0.81447196,
      "num_input_tokens_seen": 152884695,
      "step": 7066,
      "time_per_iteration": 2.478912591934204
    },
    {
      "auxiliary_loss_clip": 0.01152458,
      "auxiliary_loss_mlp": 0.01027923,
      "balance_loss_clip": 1.04833734,
      "balance_loss_mlp": 1.0211103,
      "epoch": 0.8497565081464559,
      "flos": 14609107756800.0,
      "grad_norm": 2.148362165666817,
      "language_loss": 0.73359048,
      "learning_rate": 2.3202060267740141e-07,
      "loss": 0.75539428,
      "num_input_tokens_seen": 152902220,
      "step": 7067,
      "time_per_iteration": 3.5590946674346924
    },
    {
      "auxiliary_loss_clip": 0.0110167,
      "auxiliary_loss_mlp": 0.01023786,
      "balance_loss_clip": 1.03814054,
      "balance_loss_mlp": 1.01677692,
      "epoch": 0.8498767510370949,
      "flos": 21136446126720.0,
      "grad_norm": 2.3594405136862386,
      "language_loss": 0.76918495,
      "learning_rate": 2.3165656257442044e-07,
      "loss": 0.79043955,
      "num_input_tokens_seen": 152920740,
      "step": 7068,
      "time_per_iteration": 2.6745221614837646
    },
    {
      "auxiliary_loss_clip": 0.01146005,
      "auxiliary_loss_mlp": 0.01027064,
      "balance_loss_clip": 1.04670024,
      "balance_loss_mlp": 1.02051377,
      "epoch": 0.849996993927734,
      "flos": 23654538195840.0,
      "grad_norm": 1.8208785741318407,
      "language_loss": 0.90143144,
      "learning_rate": 2.31292790727734e-07,
      "loss": 0.9231621,
      "num_input_tokens_seen": 152938305,
      "step": 7069,
      "time_per_iteration": 2.496387243270874
    },
    {
      "auxiliary_loss_clip": 0.01159175,
      "auxiliary_loss_mlp": 0.01024799,
      "balance_loss_clip": 1.04519796,
      "balance_loss_mlp": 1.01798582,
      "epoch": 0.8501172368183731,
      "flos": 20558069331840.0,
      "grad_norm": 2.5900098484439718,
      "language_loss": 0.79984736,
      "learning_rate": 2.3092928719252392e-07,
      "loss": 0.8216871,
      "num_input_tokens_seen": 152956705,
      "step": 7070,
      "time_per_iteration": 2.464017629623413
    },
    {
      "auxiliary_loss_clip": 0.0114647,
      "auxiliary_loss_mlp": 0.0102457,
      "balance_loss_clip": 1.04431915,
      "balance_loss_mlp": 1.0178194,
      "epoch": 0.8502374797090122,
      "flos": 22272624201600.0,
      "grad_norm": 2.0437908963875056,
      "language_loss": 0.78396118,
      "learning_rate": 2.3056605202393475e-07,
      "loss": 0.80567157,
      "num_input_tokens_seen": 152974265,
      "step": 7071,
      "time_per_iteration": 2.483010768890381
    },
    {
      "auxiliary_loss_clip": 0.01142267,
      "auxiliary_loss_mlp": 0.00761239,
      "balance_loss_clip": 1.04024696,
      "balance_loss_mlp": 1.00029922,
      "epoch": 0.8503577225996513,
      "flos": 23659817495040.0,
      "grad_norm": 2.5387385051966485,
      "language_loss": 0.66886067,
      "learning_rate": 2.3020308527706888e-07,
      "loss": 0.68789577,
      "num_input_tokens_seen": 152993680,
      "step": 7072,
      "time_per_iteration": 2.516066551208496
    },
    {
      "auxiliary_loss_clip": 0.01138337,
      "auxiliary_loss_mlp": 0.01024961,
      "balance_loss_clip": 1.04293633,
      "balance_loss_mlp": 1.01808906,
      "epoch": 0.8504779654902904,
      "flos": 26758513002240.0,
      "grad_norm": 1.9609204586111524,
      "language_loss": 0.89144981,
      "learning_rate": 2.2984038700698715e-07,
      "loss": 0.91308278,
      "num_input_tokens_seen": 153012990,
      "step": 7073,
      "time_per_iteration": 2.579803705215454
    },
    {
      "auxiliary_loss_clip": 0.01147215,
      "auxiliary_loss_mlp": 0.01026394,
      "balance_loss_clip": 1.04654384,
      "balance_loss_mlp": 1.01992428,
      "epoch": 0.8505982083809295,
      "flos": 26468247196800.0,
      "grad_norm": 1.8273253395274969,
      "language_loss": 0.78899539,
      "learning_rate": 2.2947795726871222e-07,
      "loss": 0.81073141,
      "num_input_tokens_seen": 153034015,
      "step": 7074,
      "time_per_iteration": 2.5389277935028076
    },
    {
      "auxiliary_loss_clip": 0.01148552,
      "auxiliary_loss_mlp": 0.00760953,
      "balance_loss_clip": 1.04927707,
      "balance_loss_mlp": 1.00033545,
      "epoch": 0.8507184512715685,
      "flos": 20303390926080.0,
      "grad_norm": 1.7106387109087806,
      "language_loss": 0.85518283,
      "learning_rate": 2.2911579611722253e-07,
      "loss": 0.87427789,
      "num_input_tokens_seen": 153053160,
      "step": 7075,
      "time_per_iteration": 2.486285448074341
    },
    {
      "auxiliary_loss_clip": 0.01129282,
      "auxiliary_loss_mlp": 0.01027109,
      "balance_loss_clip": 1.04268467,
      "balance_loss_mlp": 1.02067518,
      "epoch": 0.8508386941622077,
      "flos": 19025186474880.0,
      "grad_norm": 2.2527955360846206,
      "language_loss": 0.87279129,
      "learning_rate": 2.2875390360745905e-07,
      "loss": 0.89435518,
      "num_input_tokens_seen": 153072565,
      "step": 7076,
      "time_per_iteration": 2.512643575668335
    },
    {
      "auxiliary_loss_clip": 0.01124152,
      "auxiliary_loss_mlp": 0.01022839,
      "balance_loss_clip": 1.04208386,
      "balance_loss_mlp": 1.01567459,
      "epoch": 0.8509589370528468,
      "flos": 16433405654400.0,
      "grad_norm": 1.6312882088282665,
      "language_loss": 0.77329135,
      "learning_rate": 2.2839227979432008e-07,
      "loss": 0.7947613,
      "num_input_tokens_seen": 153090215,
      "step": 7077,
      "time_per_iteration": 2.531562328338623
    },
    {
      "auxiliary_loss_clip": 0.01138474,
      "auxiliary_loss_mlp": 0.01025521,
      "balance_loss_clip": 1.04448974,
      "balance_loss_mlp": 1.01831222,
      "epoch": 0.8510791799434858,
      "flos": 18259714713600.0,
      "grad_norm": 1.693106886776018,
      "language_loss": 0.84886175,
      "learning_rate": 2.2803092473266373e-07,
      "loss": 0.8705017,
      "num_input_tokens_seen": 153107740,
      "step": 7078,
      "time_per_iteration": 4.158874988555908
    },
    {
      "auxiliary_loss_clip": 0.01165967,
      "auxiliary_loss_mlp": 0.01030359,
      "balance_loss_clip": 1.04950404,
      "balance_loss_mlp": 1.02342975,
      "epoch": 0.851199422834125,
      "flos": 23441372933760.0,
      "grad_norm": 2.332604473455492,
      "language_loss": 0.86490464,
      "learning_rate": 2.2766983847730724e-07,
      "loss": 0.88686794,
      "num_input_tokens_seen": 153127410,
      "step": 7079,
      "time_per_iteration": 2.4785001277923584
    },
    {
      "auxiliary_loss_clip": 0.01130147,
      "auxiliary_loss_mlp": 0.01031964,
      "balance_loss_clip": 1.04254985,
      "balance_loss_mlp": 1.02457345,
      "epoch": 0.851319665724764,
      "flos": 16289404030080.0,
      "grad_norm": 2.3082065367252187,
      "language_loss": 0.66663706,
      "learning_rate": 2.2730902108302663e-07,
      "loss": 0.68825817,
      "num_input_tokens_seen": 153144325,
      "step": 7080,
      "time_per_iteration": 3.2927205562591553
    },
    {
      "auxiliary_loss_clip": 0.01127792,
      "auxiliary_loss_mlp": 0.01026088,
      "balance_loss_clip": 1.04105759,
      "balance_loss_mlp": 1.0186646,
      "epoch": 0.8514399086154031,
      "flos": 18989347680000.0,
      "grad_norm": 1.6027284782879994,
      "language_loss": 0.68491662,
      "learning_rate": 2.269484726045583e-07,
      "loss": 0.70645547,
      "num_input_tokens_seen": 153163240,
      "step": 7081,
      "time_per_iteration": 2.5224485397338867
    },
    {
      "auxiliary_loss_clip": 0.01128441,
      "auxiliary_loss_mlp": 0.01023868,
      "balance_loss_clip": 1.04659975,
      "balance_loss_mlp": 1.01729345,
      "epoch": 0.8515601515060423,
      "flos": 24571194301440.0,
      "grad_norm": 1.6613215328222168,
      "language_loss": 0.78881681,
      "learning_rate": 2.2658819309659672e-07,
      "loss": 0.81033993,
      "num_input_tokens_seen": 153183440,
      "step": 7082,
      "time_per_iteration": 2.6303181648254395
    },
    {
      "auxiliary_loss_clip": 0.01132763,
      "auxiliary_loss_mlp": 0.01023381,
      "balance_loss_clip": 1.0466038,
      "balance_loss_mlp": 1.01682174,
      "epoch": 0.8516803943966813,
      "flos": 19529443555200.0,
      "grad_norm": 1.8145613117354946,
      "language_loss": 0.84845209,
      "learning_rate": 2.2622818261379706e-07,
      "loss": 0.87001348,
      "num_input_tokens_seen": 153200460,
      "step": 7083,
      "time_per_iteration": 2.5482494831085205
    },
    {
      "auxiliary_loss_clip": 0.01133199,
      "auxiliary_loss_mlp": 0.01027624,
      "balance_loss_clip": 1.04413331,
      "balance_loss_mlp": 1.01994705,
      "epoch": 0.8518006372873204,
      "flos": 20265792364800.0,
      "grad_norm": 1.7940183998586234,
      "language_loss": 0.74710262,
      "learning_rate": 2.2586844121077142e-07,
      "loss": 0.76871085,
      "num_input_tokens_seen": 153218970,
      "step": 7084,
      "time_per_iteration": 2.521526575088501
    },
    {
      "auxiliary_loss_clip": 0.01104388,
      "auxiliary_loss_mlp": 0.01025956,
      "balance_loss_clip": 1.03960371,
      "balance_loss_mlp": 1.01886296,
      "epoch": 0.8519208801779595,
      "flos": 24133227770880.0,
      "grad_norm": 2.0112045670666387,
      "language_loss": 0.71774584,
      "learning_rate": 2.2550896894209215e-07,
      "loss": 0.73904932,
      "num_input_tokens_seen": 153238485,
      "step": 7085,
      "time_per_iteration": 2.6175196170806885
    },
    {
      "auxiliary_loss_clip": 0.01008525,
      "auxiliary_loss_mlp": 0.01001505,
      "balance_loss_clip": 1.01063967,
      "balance_loss_mlp": 1.00031304,
      "epoch": 0.8520411230685986,
      "flos": 63035223252480.0,
      "grad_norm": 0.6779881122009657,
      "language_loss": 0.56654203,
      "learning_rate": 2.2514976586229184e-07,
      "loss": 0.58664232,
      "num_input_tokens_seen": 153306430,
      "step": 7086,
      "time_per_iteration": 3.3455471992492676
    },
    {
      "auxiliary_loss_clip": 0.0104627,
      "auxiliary_loss_mlp": 0.01002574,
      "balance_loss_clip": 1.00864959,
      "balance_loss_mlp": 1.00147772,
      "epoch": 0.8521613659592376,
      "flos": 65836865283840.0,
      "grad_norm": 0.7508151229877704,
      "language_loss": 0.54663694,
      "learning_rate": 2.247908320258609e-07,
      "loss": 0.56712544,
      "num_input_tokens_seen": 153366520,
      "step": 7087,
      "time_per_iteration": 3.085184335708618
    },
    {
      "auxiliary_loss_clip": 0.01099618,
      "auxiliary_loss_mlp": 0.01024735,
      "balance_loss_clip": 1.04253721,
      "balance_loss_mlp": 1.01709962,
      "epoch": 0.8522816088498768,
      "flos": 23112323418240.0,
      "grad_norm": 2.000702376488285,
      "language_loss": 0.79355013,
      "learning_rate": 2.2443216748724914e-07,
      "loss": 0.81479371,
      "num_input_tokens_seen": 153387230,
      "step": 7088,
      "time_per_iteration": 2.6145870685577393
    },
    {
      "auxiliary_loss_clip": 0.01152943,
      "auxiliary_loss_mlp": 0.00760592,
      "balance_loss_clip": 1.04770005,
      "balance_loss_mlp": 1.00026321,
      "epoch": 0.8524018517405159,
      "flos": 31758140073600.0,
      "grad_norm": 2.246860344815966,
      "language_loss": 0.74306846,
      "learning_rate": 2.2407377230086588e-07,
      "loss": 0.76220381,
      "num_input_tokens_seen": 153409585,
      "step": 7089,
      "time_per_iteration": 2.584655523300171
    },
    {
      "auxiliary_loss_clip": 0.01118506,
      "auxiliary_loss_mlp": 0.01031128,
      "balance_loss_clip": 1.04352224,
      "balance_loss_mlp": 1.0240972,
      "epoch": 0.8525220946311549,
      "flos": 18690318956160.0,
      "grad_norm": 1.9773833379981034,
      "language_loss": 0.8327986,
      "learning_rate": 2.23715646521079e-07,
      "loss": 0.85429502,
      "num_input_tokens_seen": 153427105,
      "step": 7090,
      "time_per_iteration": 2.5695416927337646
    },
    {
      "auxiliary_loss_clip": 0.01154161,
      "auxiliary_loss_mlp": 0.00761258,
      "balance_loss_clip": 1.0465126,
      "balance_loss_mlp": 1.00030243,
      "epoch": 0.852642337521794,
      "flos": 21793216354560.0,
      "grad_norm": 1.946998432856179,
      "language_loss": 0.83912277,
      "learning_rate": 2.2335779020221724e-07,
      "loss": 0.85827696,
      "num_input_tokens_seen": 153443725,
      "step": 7091,
      "time_per_iteration": 2.4974160194396973
    },
    {
      "auxiliary_loss_clip": 0.01038967,
      "auxiliary_loss_mlp": 0.01000296,
      "balance_loss_clip": 1.00859499,
      "balance_loss_mlp": 0.99926513,
      "epoch": 0.8527625804124331,
      "flos": 69040132260480.0,
      "grad_norm": 0.8072700180855711,
      "language_loss": 0.56415737,
      "learning_rate": 2.2300020339856497e-07,
      "loss": 0.58455002,
      "num_input_tokens_seen": 153506410,
      "step": 7092,
      "time_per_iteration": 3.138139009475708
    },
    {
      "auxiliary_loss_clip": 0.01134368,
      "auxiliary_loss_mlp": 0.01023438,
      "balance_loss_clip": 1.04419136,
      "balance_loss_mlp": 1.01691103,
      "epoch": 0.8528828233030722,
      "flos": 26979399688320.0,
      "grad_norm": 2.2935103499693184,
      "language_loss": 0.78152347,
      "learning_rate": 2.2264288616436966e-07,
      "loss": 0.8031016,
      "num_input_tokens_seen": 153526665,
      "step": 7093,
      "time_per_iteration": 3.5478947162628174
    },
    {
      "auxiliary_loss_clip": 0.01130921,
      "auxiliary_loss_mlp": 0.01027187,
      "balance_loss_clip": 1.04675484,
      "balance_loss_mlp": 1.02034163,
      "epoch": 0.8530030661937112,
      "flos": 17487598936320.0,
      "grad_norm": 1.8855149353758696,
      "language_loss": 0.72376919,
      "learning_rate": 2.222858385538351e-07,
      "loss": 0.74535024,
      "num_input_tokens_seen": 153543465,
      "step": 7094,
      "time_per_iteration": 2.50673508644104
    },
    {
      "auxiliary_loss_clip": 0.01147935,
      "auxiliary_loss_mlp": 0.01028807,
      "balance_loss_clip": 1.04537678,
      "balance_loss_mlp": 1.02173185,
      "epoch": 0.8531233090843504,
      "flos": 22160798184960.0,
      "grad_norm": 1.6001851121193664,
      "language_loss": 0.6819452,
      "learning_rate": 2.2192906062112527e-07,
      "loss": 0.70371258,
      "num_input_tokens_seen": 153563340,
      "step": 7095,
      "time_per_iteration": 2.501384735107422
    },
    {
      "auxiliary_loss_clip": 0.01161928,
      "auxiliary_loss_mlp": 0.01025397,
      "balance_loss_clip": 1.04602659,
      "balance_loss_mlp": 1.01860821,
      "epoch": 0.8532435519749895,
      "flos": 37635388145280.0,
      "grad_norm": 1.5481216993896698,
      "language_loss": 0.70452368,
      "learning_rate": 2.2157255242036377e-07,
      "loss": 0.72639692,
      "num_input_tokens_seen": 153587005,
      "step": 7096,
      "time_per_iteration": 2.607110023498535
    },
    {
      "auxiliary_loss_clip": 0.0111865,
      "auxiliary_loss_mlp": 0.01026879,
      "balance_loss_clip": 1.04224229,
      "balance_loss_mlp": 1.02019143,
      "epoch": 0.8533637948656285,
      "flos": 21398163598080.0,
      "grad_norm": 1.6059794880448326,
      "language_loss": 0.73915172,
      "learning_rate": 2.2121631400563135e-07,
      "loss": 0.760607,
      "num_input_tokens_seen": 153606835,
      "step": 7097,
      "time_per_iteration": 2.5775718688964844
    },
    {
      "auxiliary_loss_clip": 0.01044558,
      "auxiliary_loss_mlp": 0.01001481,
      "balance_loss_clip": 1.0115453,
      "balance_loss_mlp": 1.00049198,
      "epoch": 0.8534840377562677,
      "flos": 53345122490880.0,
      "grad_norm": 0.7783484311923864,
      "language_loss": 0.53025484,
      "learning_rate": 2.208603454309701e-07,
      "loss": 0.55071527,
      "num_input_tokens_seen": 153664925,
      "step": 7098,
      "time_per_iteration": 3.0725367069244385
    },
    {
      "auxiliary_loss_clip": 0.01106779,
      "auxiliary_loss_mlp": 0.01026255,
      "balance_loss_clip": 1.04305255,
      "balance_loss_mlp": 1.0191704,
      "epoch": 0.8536042806469067,
      "flos": 20814148368000.0,
      "grad_norm": 1.8889054053657617,
      "language_loss": 0.70997918,
      "learning_rate": 2.2050464675037994e-07,
      "loss": 0.73130953,
      "num_input_tokens_seen": 153683550,
      "step": 7099,
      "time_per_iteration": 2.6063196659088135
    },
    {
      "auxiliary_loss_clip": 0.01135709,
      "auxiliary_loss_mlp": 0.01028817,
      "balance_loss_clip": 1.04454231,
      "balance_loss_mlp": 1.02163458,
      "epoch": 0.8537245235375458,
      "flos": 24681368292480.0,
      "grad_norm": 2.0834455857019663,
      "language_loss": 0.72423565,
      "learning_rate": 2.2014921801782016e-07,
      "loss": 0.74588084,
      "num_input_tokens_seen": 153703040,
      "step": 7100,
      "time_per_iteration": 2.5610036849975586
    },
    {
      "auxiliary_loss_clip": 0.01135974,
      "auxiliary_loss_mlp": 0.01024929,
      "balance_loss_clip": 1.04060292,
      "balance_loss_mlp": 1.01802945,
      "epoch": 0.853844766428185,
      "flos": 24384817607040.0,
      "grad_norm": 1.8246362742583777,
      "language_loss": 0.73818231,
      "learning_rate": 2.1979405928720872e-07,
      "loss": 0.75979137,
      "num_input_tokens_seen": 153722695,
      "step": 7101,
      "time_per_iteration": 2.564816951751709
    },
    {
      "auxiliary_loss_clip": 0.01137723,
      "auxiliary_loss_mlp": 0.0102667,
      "balance_loss_clip": 1.04341316,
      "balance_loss_mlp": 1.02005386,
      "epoch": 0.853965009318824,
      "flos": 20955707867520.0,
      "grad_norm": 1.409318191139175,
      "language_loss": 0.792858,
      "learning_rate": 2.1943917061242257e-07,
      "loss": 0.81450194,
      "num_input_tokens_seen": 153742550,
      "step": 7102,
      "time_per_iteration": 2.5368049144744873
    },
    {
      "auxiliary_loss_clip": 0.01156628,
      "auxiliary_loss_mlp": 0.00761409,
      "balance_loss_clip": 1.04778934,
      "balance_loss_mlp": 1.00027108,
      "epoch": 0.8540852522094631,
      "flos": 24201816791040.0,
      "grad_norm": 1.5657498606566191,
      "language_loss": 0.66541851,
      "learning_rate": 2.1908455204729903e-07,
      "loss": 0.68459892,
      "num_input_tokens_seen": 153761700,
      "step": 7103,
      "time_per_iteration": 3.312654495239258
    },
    {
      "auxiliary_loss_clip": 0.01134026,
      "auxiliary_loss_mlp": 0.01024995,
      "balance_loss_clip": 1.04283011,
      "balance_loss_mlp": 1.01808095,
      "epoch": 0.8542054951001022,
      "flos": 25082921410560.0,
      "grad_norm": 1.8388413769805714,
      "language_loss": 0.7832495,
      "learning_rate": 2.1873020364563265e-07,
      "loss": 0.80483973,
      "num_input_tokens_seen": 153780765,
      "step": 7104,
      "time_per_iteration": 3.403931140899658
    },
    {
      "auxiliary_loss_clip": 0.01145734,
      "auxiliary_loss_mlp": 0.01018123,
      "balance_loss_clip": 1.04601431,
      "balance_loss_mlp": 1.01103604,
      "epoch": 0.8543257379907413,
      "flos": 24316551809280.0,
      "grad_norm": 2.1154769325732605,
      "language_loss": 0.75910103,
      "learning_rate": 2.183761254611789e-07,
      "loss": 0.78073955,
      "num_input_tokens_seen": 153801090,
      "step": 7105,
      "time_per_iteration": 2.5172483921051025
    },
    {
      "auxiliary_loss_clip": 0.0114899,
      "auxiliary_loss_mlp": 0.01024504,
      "balance_loss_clip": 1.04837799,
      "balance_loss_mlp": 1.0179565,
      "epoch": 0.8544459808813804,
      "flos": 55286630467200.0,
      "grad_norm": 2.0624387480063238,
      "language_loss": 0.70188475,
      "learning_rate": 2.1802231754764987e-07,
      "loss": 0.7236197,
      "num_input_tokens_seen": 153826530,
      "step": 7106,
      "time_per_iteration": 3.4550530910491943
    },
    {
      "auxiliary_loss_clip": 0.01137044,
      "auxiliary_loss_mlp": 0.01026885,
      "balance_loss_clip": 1.04333031,
      "balance_loss_mlp": 1.01915431,
      "epoch": 0.8545662237720195,
      "flos": 25776248705280.0,
      "grad_norm": 1.8284435046589689,
      "language_loss": 0.76550949,
      "learning_rate": 2.17668779958718e-07,
      "loss": 0.78714871,
      "num_input_tokens_seen": 153849110,
      "step": 7107,
      "time_per_iteration": 2.5585248470306396
    },
    {
      "auxiliary_loss_clip": 0.01164182,
      "auxiliary_loss_mlp": 0.01027634,
      "balance_loss_clip": 1.04839826,
      "balance_loss_mlp": 1.02042508,
      "epoch": 0.8546864666626586,
      "flos": 11108320427520.0,
      "grad_norm": 2.1980495197941514,
      "language_loss": 0.80796874,
      "learning_rate": 2.1731551274801553e-07,
      "loss": 0.82988691,
      "num_input_tokens_seen": 153865550,
      "step": 7108,
      "time_per_iteration": 2.4387567043304443
    },
    {
      "auxiliary_loss_clip": 0.01134685,
      "auxiliary_loss_mlp": 0.01026142,
      "balance_loss_clip": 1.04409838,
      "balance_loss_mlp": 1.01890635,
      "epoch": 0.8548067095532976,
      "flos": 25520169669120.0,
      "grad_norm": 2.0299040063634495,
      "language_loss": 0.61442196,
      "learning_rate": 2.169625159691324e-07,
      "loss": 0.63603026,
      "num_input_tokens_seen": 153885425,
      "step": 7109,
      "time_per_iteration": 2.543661117553711
    },
    {
      "auxiliary_loss_clip": 0.01118034,
      "auxiliary_loss_mlp": 0.01025039,
      "balance_loss_clip": 1.04342449,
      "balance_loss_mlp": 1.01798165,
      "epoch": 0.8549269524439368,
      "flos": 24717853532160.0,
      "grad_norm": 2.228441603914061,
      "language_loss": 0.73923558,
      "learning_rate": 2.1660978967561784e-07,
      "loss": 0.76066631,
      "num_input_tokens_seen": 153904760,
      "step": 7110,
      "time_per_iteration": 2.6260414123535156
    },
    {
      "auxiliary_loss_clip": 0.01158299,
      "auxiliary_loss_mlp": 0.0102648,
      "balance_loss_clip": 1.04264927,
      "balance_loss_mlp": 1.01943183,
      "epoch": 0.8550471953345758,
      "flos": 19825599191040.0,
      "grad_norm": 2.445236501993734,
      "language_loss": 0.79220951,
      "learning_rate": 2.1625733392098035e-07,
      "loss": 0.81405723,
      "num_input_tokens_seen": 153920370,
      "step": 7111,
      "time_per_iteration": 2.4345145225524902
    },
    {
      "auxiliary_loss_clip": 0.01161436,
      "auxiliary_loss_mlp": 0.01025306,
      "balance_loss_clip": 1.04635286,
      "balance_loss_mlp": 1.01879692,
      "epoch": 0.8551674382252149,
      "flos": 22820441500800.0,
      "grad_norm": 1.692352145656544,
      "language_loss": 0.79617375,
      "learning_rate": 2.159051487586867e-07,
      "loss": 0.81804121,
      "num_input_tokens_seen": 153940500,
      "step": 7112,
      "time_per_iteration": 2.478616714477539
    },
    {
      "auxiliary_loss_clip": 0.01143697,
      "auxiliary_loss_mlp": 0.01026497,
      "balance_loss_clip": 1.04888487,
      "balance_loss_mlp": 1.01922226,
      "epoch": 0.8552876811158541,
      "flos": 20631255292800.0,
      "grad_norm": 2.5653553296721414,
      "language_loss": 0.72672504,
      "learning_rate": 2.155532342421642e-07,
      "loss": 0.74842697,
      "num_input_tokens_seen": 153958500,
      "step": 7113,
      "time_per_iteration": 2.560652494430542
    },
    {
      "auxiliary_loss_clip": 0.01150284,
      "auxiliary_loss_mlp": 0.01026434,
      "balance_loss_clip": 1.04474068,
      "balance_loss_mlp": 1.01908803,
      "epoch": 0.8554079240064931,
      "flos": 23112359331840.0,
      "grad_norm": 1.8298395763940394,
      "language_loss": 0.78571332,
      "learning_rate": 2.1520159042479636e-07,
      "loss": 0.80748045,
      "num_input_tokens_seen": 153976790,
      "step": 7114,
      "time_per_iteration": 2.5248165130615234
    },
    {
      "auxiliary_loss_clip": 0.01147671,
      "auxiliary_loss_mlp": 0.01024466,
      "balance_loss_clip": 1.04663479,
      "balance_loss_mlp": 1.01787663,
      "epoch": 0.8555281668971322,
      "flos": 22128047959680.0,
      "grad_norm": 1.9894412486379829,
      "language_loss": 0.71075153,
      "learning_rate": 2.148502173599287e-07,
      "loss": 0.7324729,
      "num_input_tokens_seen": 153994930,
      "step": 7115,
      "time_per_iteration": 2.503873348236084
    },
    {
      "auxiliary_loss_clip": 0.01129259,
      "auxiliary_loss_mlp": 0.01026942,
      "balance_loss_clip": 1.04357839,
      "balance_loss_mlp": 1.01964986,
      "epoch": 0.8556484097877713,
      "flos": 31139040234240.0,
      "grad_norm": 2.130948885000986,
      "language_loss": 0.66043174,
      "learning_rate": 2.1449911510086372e-07,
      "loss": 0.68199378,
      "num_input_tokens_seen": 154014400,
      "step": 7116,
      "time_per_iteration": 2.5927248001098633
    },
    {
      "auxiliary_loss_clip": 0.0114727,
      "auxiliary_loss_mlp": 0.01024881,
      "balance_loss_clip": 1.04553294,
      "balance_loss_mlp": 1.01835763,
      "epoch": 0.8557686526784104,
      "flos": 24316551809280.0,
      "grad_norm": 1.9316531468792806,
      "language_loss": 0.76900971,
      "learning_rate": 2.141482837008628e-07,
      "loss": 0.79073119,
      "num_input_tokens_seen": 154034940,
      "step": 7117,
      "time_per_iteration": 2.519806385040283
    },
    {
      "auxiliary_loss_clip": 0.01142201,
      "auxiliary_loss_mlp": 0.01025752,
      "balance_loss_clip": 1.04438066,
      "balance_loss_mlp": 1.01830101,
      "epoch": 0.8558888955690495,
      "flos": 17712723427200.0,
      "grad_norm": 1.865499358594141,
      "language_loss": 0.71863925,
      "learning_rate": 2.1379772321314826e-07,
      "loss": 0.74031878,
      "num_input_tokens_seen": 154052985,
      "step": 7118,
      "time_per_iteration": 2.4904842376708984
    },
    {
      "auxiliary_loss_clip": 0.01084942,
      "auxiliary_loss_mlp": 0.01028299,
      "balance_loss_clip": 1.03977692,
      "balance_loss_mlp": 1.02137613,
      "epoch": 0.8560091384596886,
      "flos": 19171702051200.0,
      "grad_norm": 2.074005724409962,
      "language_loss": 0.8119117,
      "learning_rate": 2.1344743369089802e-07,
      "loss": 0.83304417,
      "num_input_tokens_seen": 154068765,
      "step": 7119,
      "time_per_iteration": 3.3647048473358154
    },
    {
      "auxiliary_loss_clip": 0.01133237,
      "auxiliary_loss_mlp": 0.01029185,
      "balance_loss_clip": 1.04423213,
      "balance_loss_mlp": 1.02222908,
      "epoch": 0.8561293813503277,
      "flos": 23914855036800.0,
      "grad_norm": 1.5915827003851792,
      "language_loss": 0.82098871,
      "learning_rate": 2.130974151872522e-07,
      "loss": 0.84261298,
      "num_input_tokens_seen": 154089100,
      "step": 7120,
      "time_per_iteration": 2.5654871463775635
    },
    {
      "auxiliary_loss_clip": 0.01119248,
      "auxiliary_loss_mlp": 0.01023563,
      "balance_loss_clip": 1.04499924,
      "balance_loss_mlp": 1.01620793,
      "epoch": 0.8562496242409667,
      "flos": 22529206028160.0,
      "grad_norm": 2.040549387611809,
      "language_loss": 0.78584158,
      "learning_rate": 2.1274766775530773e-07,
      "loss": 0.80726969,
      "num_input_tokens_seen": 154108965,
      "step": 7121,
      "time_per_iteration": 2.5666699409484863
    },
    {
      "auxiliary_loss_clip": 0.01165694,
      "auxiliary_loss_mlp": 0.01025501,
      "balance_loss_clip": 1.04737842,
      "balance_loss_mlp": 1.01837826,
      "epoch": 0.8563698671316058,
      "flos": 14712745472640.0,
      "grad_norm": 2.0914246848158475,
      "language_loss": 0.79474115,
      "learning_rate": 2.1239819144812077e-07,
      "loss": 0.81665313,
      "num_input_tokens_seen": 154123425,
      "step": 7122,
      "time_per_iteration": 2.4316513538360596
    },
    {
      "auxiliary_loss_clip": 0.01114485,
      "auxiliary_loss_mlp": 0.01031054,
      "balance_loss_clip": 1.0404768,
      "balance_loss_mlp": 1.02385962,
      "epoch": 0.856490110022245,
      "flos": 39167768211840.0,
      "grad_norm": 1.5894299410531016,
      "language_loss": 0.69870245,
      "learning_rate": 2.1204898631870716e-07,
      "loss": 0.72015786,
      "num_input_tokens_seen": 154148315,
      "step": 7123,
      "time_per_iteration": 2.746953248977661
    },
    {
      "auxiliary_loss_clip": 0.01139052,
      "auxiliary_loss_mlp": 0.01021725,
      "balance_loss_clip": 1.04929316,
      "balance_loss_mlp": 1.01527548,
      "epoch": 0.856610352912884,
      "flos": 29059345658880.0,
      "grad_norm": 1.928569654537518,
      "language_loss": 0.75881845,
      "learning_rate": 2.1170005242004006e-07,
      "loss": 0.78042614,
      "num_input_tokens_seen": 154169665,
      "step": 7124,
      "time_per_iteration": 2.594882011413574
    },
    {
      "auxiliary_loss_clip": 0.01135234,
      "auxiliary_loss_mlp": 0.01025945,
      "balance_loss_clip": 1.04352331,
      "balance_loss_mlp": 1.01943922,
      "epoch": 0.8567305958035231,
      "flos": 23878333883520.0,
      "grad_norm": 3.3863640845531027,
      "language_loss": 0.7778604,
      "learning_rate": 2.1135138980505384e-07,
      "loss": 0.79947221,
      "num_input_tokens_seen": 154190335,
      "step": 7125,
      "time_per_iteration": 2.573951482772827
    },
    {
      "auxiliary_loss_clip": 0.01131535,
      "auxiliary_loss_mlp": 0.010251,
      "balance_loss_clip": 1.04565728,
      "balance_loss_mlp": 1.01852882,
      "epoch": 0.8568508386941622,
      "flos": 22200120599040.0,
      "grad_norm": 1.8436497252679986,
      "language_loss": 0.71927702,
      "learning_rate": 2.110029985266395e-07,
      "loss": 0.74084342,
      "num_input_tokens_seen": 154210040,
      "step": 7126,
      "time_per_iteration": 2.5380666255950928
    },
    {
      "auxiliary_loss_clip": 0.01134813,
      "auxiliary_loss_mlp": 0.01027905,
      "balance_loss_clip": 1.04209256,
      "balance_loss_mlp": 1.02104497,
      "epoch": 0.8569710815848013,
      "flos": 17307507121920.0,
      "grad_norm": 1.5628647635365924,
      "language_loss": 0.73796213,
      "learning_rate": 2.1065487863764787e-07,
      "loss": 0.75958925,
      "num_input_tokens_seen": 154228385,
      "step": 7127,
      "time_per_iteration": 2.5043489933013916
    },
    {
      "auxiliary_loss_clip": 0.01096582,
      "auxiliary_loss_mlp": 0.01025983,
      "balance_loss_clip": 1.03591514,
      "balance_loss_mlp": 1.01878858,
      "epoch": 0.8570913244754403,
      "flos": 23732285184000.0,
      "grad_norm": 1.5067117999831743,
      "language_loss": 0.85849237,
      "learning_rate": 2.1030703019088846e-07,
      "loss": 0.87971801,
      "num_input_tokens_seen": 154249015,
      "step": 7128,
      "time_per_iteration": 2.621246814727783
    },
    {
      "auxiliary_loss_clip": 0.01145217,
      "auxiliary_loss_mlp": 0.01023368,
      "balance_loss_clip": 1.04568398,
      "balance_loss_mlp": 1.01700521,
      "epoch": 0.8572115673660795,
      "flos": 20048748433920.0,
      "grad_norm": 1.9673981870137627,
      "language_loss": 0.70281518,
      "learning_rate": 2.099594532391291e-07,
      "loss": 0.72450101,
      "num_input_tokens_seen": 154267700,
      "step": 7129,
      "time_per_iteration": 3.2773783206939697
    },
    {
      "auxiliary_loss_clip": 0.01139667,
      "auxiliary_loss_mlp": 0.01026365,
      "balance_loss_clip": 1.04317117,
      "balance_loss_mlp": 1.0194869,
      "epoch": 0.8573318102567186,
      "flos": 27160389342720.0,
      "grad_norm": 1.5917602929765993,
      "language_loss": 0.78984076,
      "learning_rate": 2.0961214783509806e-07,
      "loss": 0.81150109,
      "num_input_tokens_seen": 154290580,
      "step": 7130,
      "time_per_iteration": 3.3909220695495605
    },
    {
      "auxiliary_loss_clip": 0.01138973,
      "auxiliary_loss_mlp": 0.01025985,
      "balance_loss_clip": 1.04312313,
      "balance_loss_mlp": 1.01932716,
      "epoch": 0.8574520531473576,
      "flos": 24936585402240.0,
      "grad_norm": 1.7295013111161173,
      "language_loss": 0.74488258,
      "learning_rate": 2.0926511403148051e-07,
      "loss": 0.76653218,
      "num_input_tokens_seen": 154309545,
      "step": 7131,
      "time_per_iteration": 2.5754666328430176
    },
    {
      "auxiliary_loss_clip": 0.01126065,
      "auxiliary_loss_mlp": 0.01024752,
      "balance_loss_clip": 1.04527652,
      "balance_loss_mlp": 1.01785576,
      "epoch": 0.8575722960379968,
      "flos": 18771154513920.0,
      "grad_norm": 2.366629418189321,
      "language_loss": 0.75903332,
      "learning_rate": 2.0891835188092143e-07,
      "loss": 0.78054148,
      "num_input_tokens_seen": 154326545,
      "step": 7132,
      "time_per_iteration": 3.2505407333374023
    },
    {
      "auxiliary_loss_clip": 0.01122528,
      "auxiliary_loss_mlp": 0.01030818,
      "balance_loss_clip": 1.04231679,
      "balance_loss_mlp": 1.02390361,
      "epoch": 0.8576925389286358,
      "flos": 22200300167040.0,
      "grad_norm": 1.8328221872097745,
      "language_loss": 0.81166589,
      "learning_rate": 2.0857186143602434e-07,
      "loss": 0.83319938,
      "num_input_tokens_seen": 154345190,
      "step": 7133,
      "time_per_iteration": 2.5705645084381104
    },
    {
      "auxiliary_loss_clip": 0.01112901,
      "auxiliary_loss_mlp": 0.01028832,
      "balance_loss_clip": 1.04051077,
      "balance_loss_mlp": 1.02184939,
      "epoch": 0.8578127818192749,
      "flos": 22894345733760.0,
      "grad_norm": 1.7158710048723405,
      "language_loss": 0.67678046,
      "learning_rate": 2.0822564274935094e-07,
      "loss": 0.69819784,
      "num_input_tokens_seen": 154364615,
      "step": 7134,
      "time_per_iteration": 2.614271402359009
    },
    {
      "auxiliary_loss_clip": 0.01133974,
      "auxiliary_loss_mlp": 0.01025745,
      "balance_loss_clip": 1.04660535,
      "balance_loss_mlp": 1.01848531,
      "epoch": 0.8579330247099141,
      "flos": 34824839541120.0,
      "grad_norm": 1.8615872723854747,
      "language_loss": 0.6697405,
      "learning_rate": 2.078796958734239e-07,
      "loss": 0.69133765,
      "num_input_tokens_seen": 154387335,
      "step": 7135,
      "time_per_iteration": 2.666271209716797
    },
    {
      "auxiliary_loss_clip": 0.01149946,
      "auxiliary_loss_mlp": 0.01023013,
      "balance_loss_clip": 1.04750943,
      "balance_loss_mlp": 1.01553857,
      "epoch": 0.8580532676005531,
      "flos": 19755681367680.0,
      "grad_norm": 2.406507315281437,
      "language_loss": 0.74623871,
      "learning_rate": 2.0753402086072124e-07,
      "loss": 0.7679683,
      "num_input_tokens_seen": 154405965,
      "step": 7136,
      "time_per_iteration": 2.499075412750244
    },
    {
      "auxiliary_loss_clip": 0.01088025,
      "auxiliary_loss_mlp": 0.01024756,
      "balance_loss_clip": 1.04157615,
      "balance_loss_mlp": 1.01823533,
      "epoch": 0.8581735104911922,
      "flos": 22739318634240.0,
      "grad_norm": 2.077459825572958,
      "language_loss": 0.75259149,
      "learning_rate": 2.071886177636828e-07,
      "loss": 0.77371931,
      "num_input_tokens_seen": 154422750,
      "step": 7137,
      "time_per_iteration": 2.671675682067871
    },
    {
      "auxiliary_loss_clip": 0.01148004,
      "auxiliary_loss_mlp": 0.01025204,
      "balance_loss_clip": 1.04689789,
      "balance_loss_mlp": 1.01775932,
      "epoch": 0.8582937533818313,
      "flos": 23149131880320.0,
      "grad_norm": 1.7337138732927413,
      "language_loss": 0.83015734,
      "learning_rate": 2.0684348663470575e-07,
      "loss": 0.85188937,
      "num_input_tokens_seen": 154442930,
      "step": 7138,
      "time_per_iteration": 2.5330188274383545
    },
    {
      "auxiliary_loss_clip": 0.01133406,
      "auxiliary_loss_mlp": 0.01022961,
      "balance_loss_clip": 1.04053998,
      "balance_loss_mlp": 1.01565957,
      "epoch": 0.8584139962724704,
      "flos": 19498668577920.0,
      "grad_norm": 1.8442432114910252,
      "language_loss": 0.617109,
      "learning_rate": 2.0649862752614555e-07,
      "loss": 0.63867271,
      "num_input_tokens_seen": 154461640,
      "step": 7139,
      "time_per_iteration": 2.532924175262451
    },
    {
      "auxiliary_loss_clip": 0.01036272,
      "auxiliary_loss_mlp": 0.01000487,
      "balance_loss_clip": 1.00809002,
      "balance_loss_mlp": 0.99937803,
      "epoch": 0.8585342391631094,
      "flos": 71276577788160.0,
      "grad_norm": 0.7534611323741901,
      "language_loss": 0.57017314,
      "learning_rate": 2.0615404049031838e-07,
      "loss": 0.59054077,
      "num_input_tokens_seen": 154518610,
      "step": 7140,
      "time_per_iteration": 3.1308369636535645
    },
    {
      "auxiliary_loss_clip": 0.01149417,
      "auxiliary_loss_mlp": 0.01030282,
      "balance_loss_clip": 1.04635549,
      "balance_loss_mlp": 1.02277768,
      "epoch": 0.8586544820537486,
      "flos": 10815432929280.0,
      "grad_norm": 3.7116436035198097,
      "language_loss": 0.78028667,
      "learning_rate": 2.0580972557949616e-07,
      "loss": 0.80208361,
      "num_input_tokens_seen": 154533700,
      "step": 7141,
      "time_per_iteration": 2.475665330886841
    },
    {
      "auxiliary_loss_clip": 0.01046147,
      "auxiliary_loss_mlp": 0.01002556,
      "balance_loss_clip": 1.00756681,
      "balance_loss_mlp": 1.0013814,
      "epoch": 0.8587747249443877,
      "flos": 64811184422400.0,
      "grad_norm": 0.7944029044256636,
      "language_loss": 0.54260665,
      "learning_rate": 2.054656828459125e-07,
      "loss": 0.56309366,
      "num_input_tokens_seen": 154597810,
      "step": 7142,
      "time_per_iteration": 3.133776903152466
    },
    {
      "auxiliary_loss_clip": 0.01105441,
      "auxiliary_loss_mlp": 0.01031308,
      "balance_loss_clip": 1.04197574,
      "balance_loss_mlp": 1.02430773,
      "epoch": 0.8588949678350267,
      "flos": 26834607964800.0,
      "grad_norm": 1.7529046913003,
      "language_loss": 0.77421844,
      "learning_rate": 2.051219123417578e-07,
      "loss": 0.79558587,
      "num_input_tokens_seen": 154617870,
      "step": 7143,
      "time_per_iteration": 2.662224292755127
    },
    {
      "auxiliary_loss_clip": 0.01163305,
      "auxiliary_loss_mlp": 0.0102702,
      "balance_loss_clip": 1.04620028,
      "balance_loss_mlp": 1.01994205,
      "epoch": 0.8590152107256659,
      "flos": 26104256726400.0,
      "grad_norm": 2.11316025719277,
      "language_loss": 0.60273933,
      "learning_rate": 2.0477841411918196e-07,
      "loss": 0.62464261,
      "num_input_tokens_seen": 154637395,
      "step": 7144,
      "time_per_iteration": 2.5115489959716797
    },
    {
      "auxiliary_loss_clip": 0.01143798,
      "auxiliary_loss_mlp": 0.01023313,
      "balance_loss_clip": 1.04434276,
      "balance_loss_mlp": 1.01639009,
      "epoch": 0.859135453616305,
      "flos": 26140885620480.0,
      "grad_norm": 2.1324422584551685,
      "language_loss": 0.7459327,
      "learning_rate": 2.0443518823029326e-07,
      "loss": 0.76760381,
      "num_input_tokens_seen": 154657935,
      "step": 7145,
      "time_per_iteration": 3.3944649696350098
    },
    {
      "auxiliary_loss_clip": 0.01115936,
      "auxiliary_loss_mlp": 0.01025259,
      "balance_loss_clip": 1.04212904,
      "balance_loss_mlp": 1.01859534,
      "epoch": 0.859255696506944,
      "flos": 12969319046400.0,
      "grad_norm": 1.970716338473445,
      "language_loss": 0.76516694,
      "learning_rate": 2.0409223472715854e-07,
      "loss": 0.78657889,
      "num_input_tokens_seen": 154675080,
      "step": 7146,
      "time_per_iteration": 2.541827440261841
    },
    {
      "auxiliary_loss_clip": 0.01120478,
      "auxiliary_loss_mlp": 0.00760422,
      "balance_loss_clip": 1.04346418,
      "balance_loss_mlp": 1.00031924,
      "epoch": 0.8593759393975832,
      "flos": 18475753063680.0,
      "grad_norm": 1.8965006049407658,
      "language_loss": 0.74906409,
      "learning_rate": 2.0374955366180434e-07,
      "loss": 0.76787305,
      "num_input_tokens_seen": 154692720,
      "step": 7147,
      "time_per_iteration": 2.5774054527282715
    },
    {
      "auxiliary_loss_clip": 0.01124535,
      "auxiliary_loss_mlp": 0.01023958,
      "balance_loss_clip": 1.04115152,
      "balance_loss_mlp": 1.01673687,
      "epoch": 0.8594961822882222,
      "flos": 22200156512640.0,
      "grad_norm": 1.6816025136425539,
      "language_loss": 0.72729921,
      "learning_rate": 2.034071450862147e-07,
      "loss": 0.74878418,
      "num_input_tokens_seen": 154710190,
      "step": 7148,
      "time_per_iteration": 2.5772314071655273
    },
    {
      "auxiliary_loss_clip": 0.01137763,
      "auxiliary_loss_mlp": 0.01031603,
      "balance_loss_clip": 1.04227948,
      "balance_loss_mlp": 1.02411664,
      "epoch": 0.8596164251788613,
      "flos": 23294749616640.0,
      "grad_norm": 1.6261222756730747,
      "language_loss": 0.76445735,
      "learning_rate": 2.030650090523327e-07,
      "loss": 0.78615105,
      "num_input_tokens_seen": 154729380,
      "step": 7149,
      "time_per_iteration": 2.559288501739502
    },
    {
      "auxiliary_loss_clip": 0.01117664,
      "auxiliary_loss_mlp": 0.01025041,
      "balance_loss_clip": 1.04130793,
      "balance_loss_mlp": 1.01757252,
      "epoch": 0.8597366680695004,
      "flos": 31649905416960.0,
      "grad_norm": 1.6305425960348283,
      "language_loss": 0.59612447,
      "learning_rate": 2.0272314561205995e-07,
      "loss": 0.61755157,
      "num_input_tokens_seen": 154749775,
      "step": 7150,
      "time_per_iteration": 2.6614830493927
    },
    {
      "auxiliary_loss_clip": 0.01114892,
      "auxiliary_loss_mlp": 0.01027519,
      "balance_loss_clip": 1.04010403,
      "balance_loss_mlp": 1.0208348,
      "epoch": 0.8598569109601395,
      "flos": 21287738211840.0,
      "grad_norm": 1.7353291012423602,
      "language_loss": 0.72873402,
      "learning_rate": 2.023815548172567e-07,
      "loss": 0.75015807,
      "num_input_tokens_seen": 154769845,
      "step": 7151,
      "time_per_iteration": 2.6153972148895264
    },
    {
      "auxiliary_loss_clip": 0.01147916,
      "auxiliary_loss_mlp": 0.01029167,
      "balance_loss_clip": 1.04471421,
      "balance_loss_mlp": 1.02218127,
      "epoch": 0.8599771538507786,
      "flos": 25447809720960.0,
      "grad_norm": 1.6880678954411976,
      "language_loss": 0.65823734,
      "learning_rate": 2.0204023671974267e-07,
      "loss": 0.68000817,
      "num_input_tokens_seen": 154789230,
      "step": 7152,
      "time_per_iteration": 2.5435783863067627
    },
    {
      "auxiliary_loss_clip": 0.01144195,
      "auxiliary_loss_mlp": 0.01025976,
      "balance_loss_clip": 1.04437792,
      "balance_loss_mlp": 1.01927686,
      "epoch": 0.8600973967414177,
      "flos": 16723958768640.0,
      "grad_norm": 2.0465524089375666,
      "language_loss": 0.80985057,
      "learning_rate": 2.0169919137129532e-07,
      "loss": 0.83155227,
      "num_input_tokens_seen": 154807670,
      "step": 7153,
      "time_per_iteration": 2.4748005867004395
    },
    {
      "auxiliary_loss_clip": 0.01153599,
      "auxiliary_loss_mlp": 0.01028036,
      "balance_loss_clip": 1.04918921,
      "balance_loss_mlp": 1.02031088,
      "epoch": 0.8602176396320568,
      "flos": 25227928615680.0,
      "grad_norm": 2.057025864587331,
      "language_loss": 0.70629394,
      "learning_rate": 2.013584188236508e-07,
      "loss": 0.72811031,
      "num_input_tokens_seen": 154825575,
      "step": 7154,
      "time_per_iteration": 2.539083957672119
    },
    {
      "auxiliary_loss_clip": 0.01165331,
      "auxiliary_loss_mlp": 0.01025339,
      "balance_loss_clip": 1.04790735,
      "balance_loss_mlp": 1.01880956,
      "epoch": 0.8603378825226958,
      "flos": 20412236113920.0,
      "grad_norm": 1.645221947012553,
      "language_loss": 0.79536462,
      "learning_rate": 2.0101791912850396e-07,
      "loss": 0.81727135,
      "num_input_tokens_seen": 154845115,
      "step": 7155,
      "time_per_iteration": 3.2552435398101807
    },
    {
      "auxiliary_loss_clip": 0.01137741,
      "auxiliary_loss_mlp": 0.01023919,
      "balance_loss_clip": 1.04589856,
      "balance_loss_mlp": 1.01667988,
      "epoch": 0.8604581254133349,
      "flos": 34930201109760.0,
      "grad_norm": 1.8505876380047412,
      "language_loss": 0.63898826,
      "learning_rate": 2.006776923375082e-07,
      "loss": 0.66060489,
      "num_input_tokens_seen": 154866770,
      "step": 7156,
      "time_per_iteration": 3.4595329761505127
    },
    {
      "auxiliary_loss_clip": 0.01161807,
      "auxiliary_loss_mlp": 0.01023509,
      "balance_loss_clip": 1.04643798,
      "balance_loss_mlp": 1.01657653,
      "epoch": 0.860578368303974,
      "flos": 22596538072320.0,
      "grad_norm": 1.7182304424495394,
      "language_loss": 0.71041155,
      "learning_rate": 2.003377385022764e-07,
      "loss": 0.73226476,
      "num_input_tokens_seen": 154885595,
      "step": 7157,
      "time_per_iteration": 2.4753329753875732
    },
    {
      "auxiliary_loss_clip": 0.01138003,
      "auxiliary_loss_mlp": 0.01027736,
      "balance_loss_clip": 1.04505718,
      "balance_loss_mlp": 1.02117944,
      "epoch": 0.8606986111946131,
      "flos": 21324331192320.0,
      "grad_norm": 2.189964845654228,
      "language_loss": 0.77302706,
      "learning_rate": 1.9999805767437826e-07,
      "loss": 0.79468441,
      "num_input_tokens_seen": 154904485,
      "step": 7158,
      "time_per_iteration": 3.230210304260254
    },
    {
      "auxiliary_loss_clip": 0.01126757,
      "auxiliary_loss_mlp": 0.01023766,
      "balance_loss_clip": 1.0406692,
      "balance_loss_mlp": 1.01650059,
      "epoch": 0.8608188540852522,
      "flos": 28877206769280.0,
      "grad_norm": 1.718491510576672,
      "language_loss": 0.71505129,
      "learning_rate": 1.9965864990534386e-07,
      "loss": 0.73655653,
      "num_input_tokens_seen": 154925010,
      "step": 7159,
      "time_per_iteration": 2.568971633911133
    },
    {
      "auxiliary_loss_clip": 0.01113702,
      "auxiliary_loss_mlp": 0.01025179,
      "balance_loss_clip": 1.03934407,
      "balance_loss_mlp": 1.01833916,
      "epoch": 0.8609390969758913,
      "flos": 29716187713920.0,
      "grad_norm": 1.8088948711517434,
      "language_loss": 0.77564454,
      "learning_rate": 1.9931951524666092e-07,
      "loss": 0.79703337,
      "num_input_tokens_seen": 154946100,
      "step": 7160,
      "time_per_iteration": 2.66259765625
    },
    {
      "auxiliary_loss_clip": 0.01151101,
      "auxiliary_loss_mlp": 0.00760674,
      "balance_loss_clip": 1.0458535,
      "balance_loss_mlp": 1.00031376,
      "epoch": 0.8610593398665304,
      "flos": 21249349551360.0,
      "grad_norm": 1.9462646814776654,
      "language_loss": 0.81033158,
      "learning_rate": 1.9898065374977534e-07,
      "loss": 0.82944936,
      "num_input_tokens_seen": 154966305,
      "step": 7161,
      "time_per_iteration": 2.5024962425231934
    },
    {
      "auxiliary_loss_clip": 0.01114793,
      "auxiliary_loss_mlp": 0.01027037,
      "balance_loss_clip": 1.04261041,
      "balance_loss_mlp": 1.02126765,
      "epoch": 0.8611795827571694,
      "flos": 14830102183680.0,
      "grad_norm": 1.9129355287467098,
      "language_loss": 0.73095334,
      "learning_rate": 1.9864206546609342e-07,
      "loss": 0.75237167,
      "num_input_tokens_seen": 154985145,
      "step": 7162,
      "time_per_iteration": 2.5485823154449463
    },
    {
      "auxiliary_loss_clip": 0.01160705,
      "auxiliary_loss_mlp": 0.01024433,
      "balance_loss_clip": 1.04544783,
      "balance_loss_mlp": 1.01780248,
      "epoch": 0.8612998256478086,
      "flos": 24243258107520.0,
      "grad_norm": 1.7484356839992945,
      "language_loss": 0.84227335,
      "learning_rate": 1.983037504469771e-07,
      "loss": 0.86412477,
      "num_input_tokens_seen": 155003855,
      "step": 7163,
      "time_per_iteration": 2.485396385192871
    },
    {
      "auxiliary_loss_clip": 0.0114997,
      "auxiliary_loss_mlp": 0.01031015,
      "balance_loss_clip": 1.04729593,
      "balance_loss_mlp": 1.02391887,
      "epoch": 0.8614200685384477,
      "flos": 21252653602560.0,
      "grad_norm": 1.624706848745363,
      "language_loss": 0.6614362,
      "learning_rate": 1.9796570874374984e-07,
      "loss": 0.68324602,
      "num_input_tokens_seen": 155023960,
      "step": 7164,
      "time_per_iteration": 2.5103588104248047
    },
    {
      "auxiliary_loss_clip": 0.01139432,
      "auxiliary_loss_mlp": 0.01024299,
      "balance_loss_clip": 1.04523063,
      "balance_loss_mlp": 1.01716483,
      "epoch": 0.8615403114290867,
      "flos": 20007738080640.0,
      "grad_norm": 1.6307874990096964,
      "language_loss": 0.77630866,
      "learning_rate": 1.976279404076917e-07,
      "loss": 0.79794598,
      "num_input_tokens_seen": 155043360,
      "step": 7165,
      "time_per_iteration": 2.554715633392334
    },
    {
      "auxiliary_loss_clip": 0.01124318,
      "auxiliary_loss_mlp": 0.01025132,
      "balance_loss_clip": 1.0472374,
      "balance_loss_mlp": 1.0178659,
      "epoch": 0.8616605543197259,
      "flos": 29789373674880.0,
      "grad_norm": 2.357895460210842,
      "language_loss": 0.76363385,
      "learning_rate": 1.9729044549004193e-07,
      "loss": 0.78512836,
      "num_input_tokens_seen": 155064745,
      "step": 7166,
      "time_per_iteration": 2.6374943256378174
    },
    {
      "auxiliary_loss_clip": 0.01148962,
      "auxiliary_loss_mlp": 0.01023188,
      "balance_loss_clip": 1.04756629,
      "balance_loss_mlp": 1.01684594,
      "epoch": 0.8617807972103649,
      "flos": 28911609020160.0,
      "grad_norm": 1.6244525444004287,
      "language_loss": 0.70402241,
      "learning_rate": 1.9695322404199822e-07,
      "loss": 0.72574383,
      "num_input_tokens_seen": 155086790,
      "step": 7167,
      "time_per_iteration": 2.563080310821533
    },
    {
      "auxiliary_loss_clip": 0.01132401,
      "auxiliary_loss_mlp": 0.01027166,
      "balance_loss_clip": 1.04343009,
      "balance_loss_mlp": 1.02015662,
      "epoch": 0.861901040101004,
      "flos": 27673804391040.0,
      "grad_norm": 1.7686423999246503,
      "language_loss": 0.82133794,
      "learning_rate": 1.9661627611471654e-07,
      "loss": 0.84293365,
      "num_input_tokens_seen": 155106585,
      "step": 7168,
      "time_per_iteration": 2.589808940887451
    },
    {
      "auxiliary_loss_clip": 0.01136262,
      "auxiliary_loss_mlp": 0.01027777,
      "balance_loss_clip": 1.04253578,
      "balance_loss_mlp": 1.02052867,
      "epoch": 0.8620212829916432,
      "flos": 49748056755840.0,
      "grad_norm": 1.7436370851333989,
      "language_loss": 0.70214409,
      "learning_rate": 1.9627960175931246e-07,
      "loss": 0.72378445,
      "num_input_tokens_seen": 155131285,
      "step": 7169,
      "time_per_iteration": 2.778090238571167
    },
    {
      "auxiliary_loss_clip": 0.01149953,
      "auxiliary_loss_mlp": 0.01032786,
      "balance_loss_clip": 1.04799402,
      "balance_loss_mlp": 1.02640569,
      "epoch": 0.8621415258822822,
      "flos": 21138672769920.0,
      "grad_norm": 1.7519167904296022,
      "language_loss": 0.74137694,
      "learning_rate": 1.9594320102685847e-07,
      "loss": 0.76320434,
      "num_input_tokens_seen": 155150555,
      "step": 7170,
      "time_per_iteration": 2.523455858230591
    },
    {
      "auxiliary_loss_clip": 0.01126196,
      "auxiliary_loss_mlp": 0.00760415,
      "balance_loss_clip": 1.04237628,
      "balance_loss_mlp": 1.00030947,
      "epoch": 0.8622617687729213,
      "flos": 21689039934720.0,
      "grad_norm": 1.9912862530119098,
      "language_loss": 0.63854975,
      "learning_rate": 1.956070739683864e-07,
      "loss": 0.65741587,
      "num_input_tokens_seen": 155169890,
      "step": 7171,
      "time_per_iteration": 3.30953311920166
    },
    {
      "auxiliary_loss_clip": 0.01102564,
      "auxiliary_loss_mlp": 0.01025597,
      "balance_loss_clip": 1.03730941,
      "balance_loss_mlp": 1.01909995,
      "epoch": 0.8623820116635604,
      "flos": 26250592734720.0,
      "grad_norm": 1.450767929302588,
      "language_loss": 0.74171996,
      "learning_rate": 1.9527122063488678e-07,
      "loss": 0.76300162,
      "num_input_tokens_seen": 155191005,
      "step": 7172,
      "time_per_iteration": 2.596179723739624
    },
    {
      "auxiliary_loss_clip": 0.0113314,
      "auxiliary_loss_mlp": 0.01030063,
      "balance_loss_clip": 1.04101849,
      "balance_loss_mlp": 1.02302372,
      "epoch": 0.8625022545541995,
      "flos": 19647554451840.0,
      "grad_norm": 1.5313159819419662,
      "language_loss": 0.80136704,
      "learning_rate": 1.9493564107730755e-07,
      "loss": 0.822999,
      "num_input_tokens_seen": 155211005,
      "step": 7173,
      "time_per_iteration": 2.554286241531372
    },
    {
      "auxiliary_loss_clip": 0.01128331,
      "auxiliary_loss_mlp": 0.01029172,
      "balance_loss_clip": 1.04055667,
      "balance_loss_mlp": 1.02266896,
      "epoch": 0.8626224974448385,
      "flos": 21908382336000.0,
      "grad_norm": 1.8678988506650676,
      "language_loss": 0.60632885,
      "learning_rate": 1.9460033534655684e-07,
      "loss": 0.62790382,
      "num_input_tokens_seen": 155230365,
      "step": 7174,
      "time_per_iteration": 2.536045789718628
    },
    {
      "auxiliary_loss_clip": 0.01128965,
      "auxiliary_loss_mlp": 0.01022836,
      "balance_loss_clip": 1.03986394,
      "balance_loss_mlp": 1.0160799,
      "epoch": 0.8627427403354777,
      "flos": 23331198942720.0,
      "grad_norm": 1.5551753517803542,
      "language_loss": 0.84248042,
      "learning_rate": 1.9426530349349978e-07,
      "loss": 0.86399841,
      "num_input_tokens_seen": 155250815,
      "step": 7175,
      "time_per_iteration": 2.5564353466033936
    },
    {
      "auxiliary_loss_clip": 0.01147658,
      "auxiliary_loss_mlp": 0.00760563,
      "balance_loss_clip": 1.04448867,
      "balance_loss_mlp": 1.00027418,
      "epoch": 0.8628629832261168,
      "flos": 16362877299840.0,
      "grad_norm": 1.7960932922775064,
      "language_loss": 0.64517534,
      "learning_rate": 1.9393054556896038e-07,
      "loss": 0.66425753,
      "num_input_tokens_seen": 155268515,
      "step": 7176,
      "time_per_iteration": 2.4777421951293945
    },
    {
      "auxiliary_loss_clip": 0.01111922,
      "auxiliary_loss_mlp": 0.01026713,
      "balance_loss_clip": 1.03892016,
      "balance_loss_mlp": 1.01940227,
      "epoch": 0.8629832261167558,
      "flos": 28103941756800.0,
      "grad_norm": 2.4597074888009947,
      "language_loss": 0.69052738,
      "learning_rate": 1.9359606162372133e-07,
      "loss": 0.7119137,
      "num_input_tokens_seen": 155290120,
      "step": 7177,
      "time_per_iteration": 2.6304659843444824
    },
    {
      "auxiliary_loss_clip": 0.011631,
      "auxiliary_loss_mlp": 0.01024432,
      "balance_loss_clip": 1.04791057,
      "balance_loss_mlp": 1.01792622,
      "epoch": 0.863103469007395,
      "flos": 20230061310720.0,
      "grad_norm": 1.6427269285059034,
      "language_loss": 0.70653886,
      "learning_rate": 1.9326185170852293e-07,
      "loss": 0.72841418,
      "num_input_tokens_seen": 155309085,
      "step": 7178,
      "time_per_iteration": 2.450962543487549
    },
    {
      "auxiliary_loss_clip": 0.01147327,
      "auxiliary_loss_mlp": 0.01024502,
      "balance_loss_clip": 1.04465795,
      "balance_loss_mlp": 1.0177424,
      "epoch": 0.863223711898034,
      "flos": 24498547044480.0,
      "grad_norm": 1.7354827706606928,
      "language_loss": 0.71969587,
      "learning_rate": 1.9292791587406598e-07,
      "loss": 0.74141419,
      "num_input_tokens_seen": 155327945,
      "step": 7179,
      "time_per_iteration": 2.542449712753296
    },
    {
      "auxiliary_loss_clip": 0.01146913,
      "auxiliary_loss_mlp": 0.00760853,
      "balance_loss_clip": 1.04395747,
      "balance_loss_mlp": 1.00032067,
      "epoch": 0.8633439547886731,
      "flos": 17675376261120.0,
      "grad_norm": 2.174441748891654,
      "language_loss": 0.86911088,
      "learning_rate": 1.9259425417100661e-07,
      "loss": 0.8881886,
      "num_input_tokens_seen": 155344060,
      "step": 7180,
      "time_per_iteration": 2.465039014816284
    },
    {
      "auxiliary_loss_clip": 0.01087324,
      "auxiliary_loss_mlp": 0.01023671,
      "balance_loss_clip": 1.03266001,
      "balance_loss_mlp": 1.01615477,
      "epoch": 0.8634641976793123,
      "flos": 12895055677440.0,
      "grad_norm": 2.065646443381123,
      "language_loss": 0.74722522,
      "learning_rate": 1.9226086664996234e-07,
      "loss": 0.76833522,
      "num_input_tokens_seen": 155362305,
      "step": 7181,
      "time_per_iteration": 3.3828134536743164
    },
    {
      "auxiliary_loss_clip": 0.01135143,
      "auxiliary_loss_mlp": 0.01028439,
      "balance_loss_clip": 1.04637051,
      "balance_loss_mlp": 1.02157855,
      "epoch": 0.8635844405699513,
      "flos": 23878980328320.0,
      "grad_norm": 1.8530909394369006,
      "language_loss": 0.74166,
      "learning_rate": 1.9192775336150712e-07,
      "loss": 0.76329589,
      "num_input_tokens_seen": 155382605,
      "step": 7182,
      "time_per_iteration": 3.3846585750579834
    },
    {
      "auxiliary_loss_clip": 0.01044247,
      "auxiliary_loss_mlp": 0.01001704,
      "balance_loss_clip": 1.00890756,
      "balance_loss_mlp": 1.00060701,
      "epoch": 0.8637046834605904,
      "flos": 60453387521280.0,
      "grad_norm": 0.7625399855014642,
      "language_loss": 0.56324816,
      "learning_rate": 1.915949143561739e-07,
      "loss": 0.58370757,
      "num_input_tokens_seen": 155437280,
      "step": 7183,
      "time_per_iteration": 3.0729289054870605
    },
    {
      "auxiliary_loss_clip": 0.01150624,
      "auxiliary_loss_mlp": 0.01028061,
      "balance_loss_clip": 1.04870343,
      "balance_loss_mlp": 1.02099204,
      "epoch": 0.8638249263512295,
      "flos": 20558751690240.0,
      "grad_norm": 1.6264489840898542,
      "language_loss": 0.77925324,
      "learning_rate": 1.9126234968445498e-07,
      "loss": 0.80104005,
      "num_input_tokens_seen": 155456970,
      "step": 7184,
      "time_per_iteration": 3.2266340255737305
    },
    {
      "auxiliary_loss_clip": 0.01163852,
      "auxiliary_loss_mlp": 0.01026734,
      "balance_loss_clip": 1.0477159,
      "balance_loss_mlp": 1.01962566,
      "epoch": 0.8639451692418686,
      "flos": 26615768353920.0,
      "grad_norm": 1.4169971357044704,
      "language_loss": 0.67579746,
      "learning_rate": 1.9093005939679884e-07,
      "loss": 0.6977033,
      "num_input_tokens_seen": 155478925,
      "step": 7185,
      "time_per_iteration": 2.521273136138916
    },
    {
      "auxiliary_loss_clip": 0.01149342,
      "auxiliary_loss_mlp": 0.01029892,
      "balance_loss_clip": 1.04720998,
      "balance_loss_mlp": 1.0226264,
      "epoch": 0.8640654121325076,
      "flos": 15122450977920.0,
      "grad_norm": 1.8571380024509778,
      "language_loss": 0.76450729,
      "learning_rate": 1.9059804354361452e-07,
      "loss": 0.78629959,
      "num_input_tokens_seen": 155496700,
      "step": 7186,
      "time_per_iteration": 2.479379177093506
    },
    {
      "auxiliary_loss_clip": 0.01127155,
      "auxiliary_loss_mlp": 0.01024781,
      "balance_loss_clip": 1.04011893,
      "balance_loss_mlp": 1.01746726,
      "epoch": 0.8641856550231467,
      "flos": 31869068250240.0,
      "grad_norm": 1.4604738585374826,
      "language_loss": 0.69936788,
      "learning_rate": 1.902663021752684e-07,
      "loss": 0.72088724,
      "num_input_tokens_seen": 155518130,
      "step": 7187,
      "time_per_iteration": 2.6127097606658936
    },
    {
      "auxiliary_loss_clip": 0.01166611,
      "auxiliary_loss_mlp": 0.01029213,
      "balance_loss_clip": 1.04968369,
      "balance_loss_mlp": 1.02223361,
      "epoch": 0.8643058979137859,
      "flos": 14976545932800.0,
      "grad_norm": 2.4796348886020487,
      "language_loss": 0.8204428,
      "learning_rate": 1.8993483534208556e-07,
      "loss": 0.84240103,
      "num_input_tokens_seen": 155537040,
      "step": 7188,
      "time_per_iteration": 2.454071283340454
    },
    {
      "auxiliary_loss_clip": 0.01126795,
      "auxiliary_loss_mlp": 0.01026473,
      "balance_loss_clip": 1.04307425,
      "balance_loss_mlp": 1.01888561,
      "epoch": 0.8644261408044249,
      "flos": 13115726881920.0,
      "grad_norm": 2.390853335229563,
      "language_loss": 0.74767578,
      "learning_rate": 1.8960364309434884e-07,
      "loss": 0.76920849,
      "num_input_tokens_seen": 155554535,
      "step": 7189,
      "time_per_iteration": 2.4886999130249023
    },
    {
      "auxiliary_loss_clip": 0.01088778,
      "auxiliary_loss_mlp": 0.00760531,
      "balance_loss_clip": 1.03890038,
      "balance_loss_mlp": 1.00032711,
      "epoch": 0.864546383695064,
      "flos": 20850920916480.0,
      "grad_norm": 1.8257697527748729,
      "language_loss": 0.77975142,
      "learning_rate": 1.8927272548229967e-07,
      "loss": 0.79824448,
      "num_input_tokens_seen": 155574225,
      "step": 7190,
      "time_per_iteration": 2.6455118656158447
    },
    {
      "auxiliary_loss_clip": 0.01107029,
      "auxiliary_loss_mlp": 0.0102969,
      "balance_loss_clip": 1.04161167,
      "balance_loss_mlp": 1.02276349,
      "epoch": 0.8646666265857031,
      "flos": 21324582587520.0,
      "grad_norm": 1.5106593763284248,
      "language_loss": 0.82770705,
      "learning_rate": 1.8894208255613876e-07,
      "loss": 0.84907424,
      "num_input_tokens_seen": 155593540,
      "step": 7191,
      "time_per_iteration": 2.5900216102600098
    },
    {
      "auxiliary_loss_clip": 0.01162469,
      "auxiliary_loss_mlp": 0.01022367,
      "balance_loss_clip": 1.04790473,
      "balance_loss_mlp": 1.01545596,
      "epoch": 0.8647868694763422,
      "flos": 19750833031680.0,
      "grad_norm": 2.107367405147276,
      "language_loss": 0.77647632,
      "learning_rate": 1.8861171436602397e-07,
      "loss": 0.7983247,
      "num_input_tokens_seen": 155610655,
      "step": 7192,
      "time_per_iteration": 2.4750211238861084
    },
    {
      "auxiliary_loss_clip": 0.01153324,
      "auxiliary_loss_mlp": 0.01026343,
      "balance_loss_clip": 1.04839897,
      "balance_loss_mlp": 1.01892233,
      "epoch": 0.8649071123669813,
      "flos": 26176760328960.0,
      "grad_norm": 2.0752615694797343,
      "language_loss": 0.80177224,
      "learning_rate": 1.882816209620719e-07,
      "loss": 0.82356888,
      "num_input_tokens_seen": 155627365,
      "step": 7193,
      "time_per_iteration": 2.5357677936553955
    },
    {
      "auxiliary_loss_clip": 0.01136724,
      "auxiliary_loss_mlp": 0.01029723,
      "balance_loss_clip": 1.04676938,
      "balance_loss_mlp": 1.02255547,
      "epoch": 0.8650273552576204,
      "flos": 20302888135680.0,
      "grad_norm": 2.3821477184722033,
      "language_loss": 0.76528919,
      "learning_rate": 1.8795180239435738e-07,
      "loss": 0.78695363,
      "num_input_tokens_seen": 155646220,
      "step": 7194,
      "time_per_iteration": 2.5449724197387695
    },
    {
      "auxiliary_loss_clip": 0.01140298,
      "auxiliary_loss_mlp": 0.01026128,
      "balance_loss_clip": 1.04501629,
      "balance_loss_mlp": 1.01888013,
      "epoch": 0.8651475981482595,
      "flos": 23951088881280.0,
      "grad_norm": 2.676935995629338,
      "language_loss": 0.75432432,
      "learning_rate": 1.8762225871291348e-07,
      "loss": 0.77598858,
      "num_input_tokens_seen": 155662095,
      "step": 7195,
      "time_per_iteration": 2.545774221420288
    },
    {
      "auxiliary_loss_clip": 0.01162862,
      "auxiliary_loss_mlp": 0.00760194,
      "balance_loss_clip": 1.04724598,
      "balance_loss_mlp": 1.00032914,
      "epoch": 0.8652678410388985,
      "flos": 21684622561920.0,
      "grad_norm": 1.5858781981162235,
      "language_loss": 0.8098979,
      "learning_rate": 1.8729298996773201e-07,
      "loss": 0.82912844,
      "num_input_tokens_seen": 155680845,
      "step": 7196,
      "time_per_iteration": 2.482956647872925
    },
    {
      "auxiliary_loss_clip": 0.01041735,
      "auxiliary_loss_mlp": 0.01002018,
      "balance_loss_clip": 1.00716245,
      "balance_loss_mlp": 1.00096309,
      "epoch": 0.8653880839295377,
      "flos": 65224660855680.0,
      "grad_norm": 0.8299362979036086,
      "language_loss": 0.60961998,
      "learning_rate": 1.8696399620876301e-07,
      "loss": 0.63005751,
      "num_input_tokens_seen": 155737875,
      "step": 7197,
      "time_per_iteration": 3.767765998840332
    },
    {
      "auxiliary_loss_clip": 0.01114697,
      "auxiliary_loss_mlp": 0.01033465,
      "balance_loss_clip": 1.0368278,
      "balance_loss_mlp": 1.02621162,
      "epoch": 0.8655083268201768,
      "flos": 17749172753280.0,
      "grad_norm": 2.08357895748252,
      "language_loss": 0.79087007,
      "learning_rate": 1.866352774859141e-07,
      "loss": 0.8123517,
      "num_input_tokens_seen": 155753100,
      "step": 7198,
      "time_per_iteration": 2.595792055130005
    },
    {
      "auxiliary_loss_clip": 0.01123441,
      "auxiliary_loss_mlp": 0.01028232,
      "balance_loss_clip": 1.0400691,
      "balance_loss_mlp": 1.02196491,
      "epoch": 0.8656285697108158,
      "flos": 20703974376960.0,
      "grad_norm": 2.496611850782563,
      "language_loss": 0.68925655,
      "learning_rate": 1.8630683384905188e-07,
      "loss": 0.71077329,
      "num_input_tokens_seen": 155772430,
      "step": 7199,
      "time_per_iteration": 2.5724236965179443
    },
    {
      "auxiliary_loss_clip": 0.01164583,
      "auxiliary_loss_mlp": 0.00760869,
      "balance_loss_clip": 1.04868424,
      "balance_loss_mlp": 1.00029278,
      "epoch": 0.865748812601455,
      "flos": 18653833716480.0,
      "grad_norm": 1.8847250079531883,
      "language_loss": 0.88662255,
      "learning_rate": 1.8597866534800045e-07,
      "loss": 0.90587711,
      "num_input_tokens_seen": 155787545,
      "step": 7200,
      "time_per_iteration": 2.444260358810425
    },
    {
      "auxiliary_loss_clip": 0.01151253,
      "auxiliary_loss_mlp": 0.00760658,
      "balance_loss_clip": 1.04677773,
      "balance_loss_mlp": 1.00032926,
      "epoch": 0.865869055492094,
      "flos": 70652554807680.0,
      "grad_norm": 2.5001653280417075,
      "language_loss": 0.74532473,
      "learning_rate": 1.8565077203254398e-07,
      "loss": 0.76444381,
      "num_input_tokens_seen": 155813005,
      "step": 7201,
      "time_per_iteration": 2.9021856784820557
    },
    {
      "auxiliary_loss_clip": 0.01123032,
      "auxiliary_loss_mlp": 0.0102458,
      "balance_loss_clip": 1.04627419,
      "balance_loss_mlp": 1.01723373,
      "epoch": 0.8659892983827331,
      "flos": 17383961220480.0,
      "grad_norm": 2.4513084309942155,
      "language_loss": 0.72994065,
      "learning_rate": 1.8532315395242203e-07,
      "loss": 0.75141674,
      "num_input_tokens_seen": 155829455,
      "step": 7202,
      "time_per_iteration": 2.5436387062072754
    },
    {
      "auxiliary_loss_clip": 0.01124275,
      "auxiliary_loss_mlp": 0.01023636,
      "balance_loss_clip": 1.04457521,
      "balance_loss_mlp": 1.01704943,
      "epoch": 0.8661095412733723,
      "flos": 17895221452800.0,
      "grad_norm": 2.0089935093888944,
      "language_loss": 0.7217328,
      "learning_rate": 1.849958111573353e-07,
      "loss": 0.74321198,
      "num_input_tokens_seen": 155848060,
      "step": 7203,
      "time_per_iteration": 2.558933734893799
    },
    {
      "auxiliary_loss_clip": 0.01160438,
      "auxiliary_loss_mlp": 0.01022006,
      "balance_loss_clip": 1.04565239,
      "balance_loss_mlp": 1.01527071,
      "epoch": 0.8662297841640113,
      "flos": 18224163227520.0,
      "grad_norm": 1.7138758142657744,
      "language_loss": 0.63733268,
      "learning_rate": 1.8466874369694074e-07,
      "loss": 0.65915704,
      "num_input_tokens_seen": 155865755,
      "step": 7204,
      "time_per_iteration": 2.434722900390625
    },
    {
      "auxiliary_loss_clip": 0.01120993,
      "auxiliary_loss_mlp": 0.01024519,
      "balance_loss_clip": 1.03907883,
      "balance_loss_mlp": 1.01790857,
      "epoch": 0.8663500270546504,
      "flos": 16362159027840.0,
      "grad_norm": 3.38495134390697,
      "language_loss": 0.70397162,
      "learning_rate": 1.843419516208542e-07,
      "loss": 0.72542667,
      "num_input_tokens_seen": 155882680,
      "step": 7205,
      "time_per_iteration": 2.530596971511841
    },
    {
      "auxiliary_loss_clip": 0.01150419,
      "auxiliary_loss_mlp": 0.01026302,
      "balance_loss_clip": 1.04728866,
      "balance_loss_mlp": 1.01916122,
      "epoch": 0.8664702699452895,
      "flos": 17894431353600.0,
      "grad_norm": 2.1846291023592093,
      "language_loss": 0.79875207,
      "learning_rate": 1.8401543497865047e-07,
      "loss": 0.82051933,
      "num_input_tokens_seen": 155900680,
      "step": 7206,
      "time_per_iteration": 2.460141658782959
    },
    {
      "auxiliary_loss_clip": 0.01152635,
      "auxiliary_loss_mlp": 0.00760449,
      "balance_loss_clip": 1.04596853,
      "balance_loss_mlp": 1.00027347,
      "epoch": 0.8665905128359286,
      "flos": 30736373794560.0,
      "grad_norm": 2.6006692552746062,
      "language_loss": 0.64386648,
      "learning_rate": 1.836891938198608e-07,
      "loss": 0.66299731,
      "num_input_tokens_seen": 155921105,
      "step": 7207,
      "time_per_iteration": 3.4057466983795166
    },
    {
      "auxiliary_loss_clip": 0.01137899,
      "auxiliary_loss_mlp": 0.01024011,
      "balance_loss_clip": 1.0468868,
      "balance_loss_mlp": 1.01726997,
      "epoch": 0.8667107557265676,
      "flos": 18656419495680.0,
      "grad_norm": 2.075565037425898,
      "language_loss": 0.71056598,
      "learning_rate": 1.8336322819397677e-07,
      "loss": 0.73218513,
      "num_input_tokens_seen": 155938640,
      "step": 7208,
      "time_per_iteration": 3.3505895137786865
    },
    {
      "auxiliary_loss_clip": 0.01125565,
      "auxiliary_loss_mlp": 0.0102353,
      "balance_loss_clip": 1.04092216,
      "balance_loss_mlp": 1.0164696,
      "epoch": 0.8668309986172068,
      "flos": 20083725302400.0,
      "grad_norm": 1.8136044401186209,
      "language_loss": 0.62710959,
      "learning_rate": 1.8303753815044654e-07,
      "loss": 0.64860058,
      "num_input_tokens_seen": 155957945,
      "step": 7209,
      "time_per_iteration": 2.5447394847869873
    },
    {
      "auxiliary_loss_clip": 0.01140455,
      "auxiliary_loss_mlp": 0.01032326,
      "balance_loss_clip": 1.04178154,
      "balance_loss_mlp": 1.02441347,
      "epoch": 0.8669512415078459,
      "flos": 21615099788160.0,
      "grad_norm": 2.1561740829742746,
      "language_loss": 0.70651734,
      "learning_rate": 1.827121237386773e-07,
      "loss": 0.72824514,
      "num_input_tokens_seen": 155975390,
      "step": 7210,
      "time_per_iteration": 3.283843994140625
    },
    {
      "auxiliary_loss_clip": 0.01138802,
      "auxiliary_loss_mlp": 0.01027471,
      "balance_loss_clip": 1.04406834,
      "balance_loss_mlp": 1.02012753,
      "epoch": 0.8670714843984849,
      "flos": 17703601372800.0,
      "grad_norm": 3.4408673816974544,
      "language_loss": 0.75140381,
      "learning_rate": 1.8238698500803374e-07,
      "loss": 0.77306652,
      "num_input_tokens_seen": 155988155,
      "step": 7211,
      "time_per_iteration": 2.497735023498535
    },
    {
      "auxiliary_loss_clip": 0.01046422,
      "auxiliary_loss_mlp": 0.01001141,
      "balance_loss_clip": 1.00775504,
      "balance_loss_mlp": 1.00004983,
      "epoch": 0.8671917272891241,
      "flos": 60705483125760.0,
      "grad_norm": 0.7159631124734989,
      "language_loss": 0.56328845,
      "learning_rate": 1.820621220078391e-07,
      "loss": 0.58376408,
      "num_input_tokens_seen": 156052065,
      "step": 7212,
      "time_per_iteration": 3.159484386444092
    },
    {
      "auxiliary_loss_clip": 0.01163022,
      "auxiliary_loss_mlp": 0.01024133,
      "balance_loss_clip": 1.0477432,
      "balance_loss_mlp": 1.01754701,
      "epoch": 0.8673119701797631,
      "flos": 20451881750400.0,
      "grad_norm": 2.1140888384569267,
      "language_loss": 0.67794698,
      "learning_rate": 1.8173753478737553e-07,
      "loss": 0.69981849,
      "num_input_tokens_seen": 156072500,
      "step": 7213,
      "time_per_iteration": 2.4833598136901855
    },
    {
      "auxiliary_loss_clip": 0.01162316,
      "auxiliary_loss_mlp": 0.01026756,
      "balance_loss_clip": 1.0463382,
      "balance_loss_mlp": 1.02006269,
      "epoch": 0.8674322130704022,
      "flos": 19647410797440.0,
      "grad_norm": 1.9247443315881014,
      "language_loss": 0.79533732,
      "learning_rate": 1.8141322339588205e-07,
      "loss": 0.81722808,
      "num_input_tokens_seen": 156089840,
      "step": 7214,
      "time_per_iteration": 2.4648218154907227
    },
    {
      "auxiliary_loss_clip": 0.01162547,
      "auxiliary_loss_mlp": 0.01027554,
      "balance_loss_clip": 1.04711163,
      "balance_loss_mlp": 1.02078879,
      "epoch": 0.8675524559610414,
      "flos": 26025001367040.0,
      "grad_norm": 2.1880669250889673,
      "language_loss": 0.70154804,
      "learning_rate": 1.810891878825569e-07,
      "loss": 0.72344899,
      "num_input_tokens_seen": 156109815,
      "step": 7215,
      "time_per_iteration": 2.4922640323638916
    },
    {
      "auxiliary_loss_clip": 0.01133835,
      "auxiliary_loss_mlp": 0.01024303,
      "balance_loss_clip": 1.04262519,
      "balance_loss_mlp": 1.01716864,
      "epoch": 0.8676726988516804,
      "flos": 15049444584960.0,
      "grad_norm": 1.806592876116107,
      "language_loss": 0.719926,
      "learning_rate": 1.8076542829655561e-07,
      "loss": 0.74150741,
      "num_input_tokens_seen": 156128620,
      "step": 7216,
      "time_per_iteration": 2.547414541244507
    },
    {
      "auxiliary_loss_clip": 0.01132588,
      "auxiliary_loss_mlp": 0.01026213,
      "balance_loss_clip": 1.04416513,
      "balance_loss_mlp": 1.01905465,
      "epoch": 0.8677929417423195,
      "flos": 16288111140480.0,
      "grad_norm": 1.9667870584154197,
      "language_loss": 0.79285955,
      "learning_rate": 1.8044194468699203e-07,
      "loss": 0.81444758,
      "num_input_tokens_seen": 156145930,
      "step": 7217,
      "time_per_iteration": 2.4898715019226074
    },
    {
      "auxiliary_loss_clip": 0.0113438,
      "auxiliary_loss_mlp": 0.01024173,
      "balance_loss_clip": 1.04806554,
      "balance_loss_mlp": 1.01705623,
      "epoch": 0.8679131846329585,
      "flos": 18844160906880.0,
      "grad_norm": 2.4406140402838963,
      "language_loss": 0.760023,
      "learning_rate": 1.8011873710293912e-07,
      "loss": 0.78160852,
      "num_input_tokens_seen": 156164435,
      "step": 7218,
      "time_per_iteration": 2.5345187187194824
    },
    {
      "auxiliary_loss_clip": 0.01145642,
      "auxiliary_loss_mlp": 0.01026166,
      "balance_loss_clip": 1.04529572,
      "balance_loss_mlp": 1.01940346,
      "epoch": 0.8680334275235977,
      "flos": 33620718890880.0,
      "grad_norm": 1.8993937492551125,
      "language_loss": 0.70036644,
      "learning_rate": 1.7979580559342677e-07,
      "loss": 0.72208452,
      "num_input_tokens_seen": 156185165,
      "step": 7219,
      "time_per_iteration": 2.5918309688568115
    },
    {
      "auxiliary_loss_clip": 0.01136944,
      "auxiliary_loss_mlp": 0.01032871,
      "balance_loss_clip": 1.04674459,
      "balance_loss_mlp": 1.02596259,
      "epoch": 0.8681536704142367,
      "flos": 24681152810880.0,
      "grad_norm": 1.717289454406398,
      "language_loss": 0.67124176,
      "learning_rate": 1.7947315020744358e-07,
      "loss": 0.69293994,
      "num_input_tokens_seen": 156206260,
      "step": 7220,
      "time_per_iteration": 2.573223114013672
    },
    {
      "auxiliary_loss_clip": 0.01134693,
      "auxiliary_loss_mlp": 0.01028125,
      "balance_loss_clip": 1.04282212,
      "balance_loss_mlp": 1.02131534,
      "epoch": 0.8682739133048758,
      "flos": 20011042131840.0,
      "grad_norm": 2.550953516243246,
      "language_loss": 0.80306327,
      "learning_rate": 1.7915077099393594e-07,
      "loss": 0.82469141,
      "num_input_tokens_seen": 156222860,
      "step": 7221,
      "time_per_iteration": 2.5091428756713867
    },
    {
      "auxiliary_loss_clip": 0.01151914,
      "auxiliary_loss_mlp": 0.01022406,
      "balance_loss_clip": 1.0460459,
      "balance_loss_mlp": 1.01520848,
      "epoch": 0.868394156195515,
      "flos": 16654759217280.0,
      "grad_norm": 1.8681559982762532,
      "language_loss": 0.7291832,
      "learning_rate": 1.788286680018083e-07,
      "loss": 0.75092638,
      "num_input_tokens_seen": 156241570,
      "step": 7222,
      "time_per_iteration": 2.475740909576416
    },
    {
      "auxiliary_loss_clip": 0.01138452,
      "auxiliary_loss_mlp": 0.0102294,
      "balance_loss_clip": 1.04483795,
      "balance_loss_mlp": 1.01639581,
      "epoch": 0.868514399086154,
      "flos": 28001381448960.0,
      "grad_norm": 1.7817053768466888,
      "language_loss": 0.72261524,
      "learning_rate": 1.7850684127992443e-07,
      "loss": 0.74422914,
      "num_input_tokens_seen": 156261315,
      "step": 7223,
      "time_per_iteration": 3.424898624420166
    },
    {
      "auxiliary_loss_clip": 0.01125564,
      "auxiliary_loss_mlp": 0.01026392,
      "balance_loss_clip": 1.04567873,
      "balance_loss_mlp": 1.02003193,
      "epoch": 0.8686346419767931,
      "flos": 20084587228800.0,
      "grad_norm": 1.5172550678104952,
      "language_loss": 0.70296121,
      "learning_rate": 1.7818529087710378e-07,
      "loss": 0.72448081,
      "num_input_tokens_seen": 156281670,
      "step": 7224,
      "time_per_iteration": 2.558955192565918
    },
    {
      "auxiliary_loss_clip": 0.01147351,
      "auxiliary_loss_mlp": 0.00760315,
      "balance_loss_clip": 1.0452261,
      "balance_loss_mlp": 1.00029361,
      "epoch": 0.8687548848674322,
      "flos": 18223516782720.0,
      "grad_norm": 1.606313478254771,
      "language_loss": 0.84044576,
      "learning_rate": 1.7786401684212637e-07,
      "loss": 0.85952246,
      "num_input_tokens_seen": 156300500,
      "step": 7225,
      "time_per_iteration": 2.488352060317993
    },
    {
      "auxiliary_loss_clip": 0.01016361,
      "auxiliary_loss_mlp": 0.01001609,
      "balance_loss_clip": 1.00843859,
      "balance_loss_mlp": 1.00060809,
      "epoch": 0.8688751277580713,
      "flos": 70457885049600.0,
      "grad_norm": 0.7307650601745384,
      "language_loss": 0.55968791,
      "learning_rate": 1.7754301922372883e-07,
      "loss": 0.5798676,
      "num_input_tokens_seen": 156350145,
      "step": 7226,
      "time_per_iteration": 2.993025302886963
    },
    {
      "auxiliary_loss_clip": 0.01099023,
      "auxiliary_loss_mlp": 0.01028326,
      "balance_loss_clip": 1.04020715,
      "balance_loss_mlp": 1.02177811,
      "epoch": 0.8689953706487104,
      "flos": 26906788344960.0,
      "grad_norm": 1.9151947619700873,
      "language_loss": 0.80918247,
      "learning_rate": 1.7722229807060617e-07,
      "loss": 0.83045596,
      "num_input_tokens_seen": 156368725,
      "step": 7227,
      "time_per_iteration": 2.6684646606445312
    },
    {
      "auxiliary_loss_clip": 0.01110953,
      "auxiliary_loss_mlp": 0.01025777,
      "balance_loss_clip": 1.0392493,
      "balance_loss_mlp": 1.01916742,
      "epoch": 0.8691156135393495,
      "flos": 34637385438720.0,
      "grad_norm": 2.1039563268406907,
      "language_loss": 0.81963575,
      "learning_rate": 1.7690185343141172e-07,
      "loss": 0.84100306,
      "num_input_tokens_seen": 156388640,
      "step": 7228,
      "time_per_iteration": 2.6655049324035645
    },
    {
      "auxiliary_loss_clip": 0.01132452,
      "auxiliary_loss_mlp": 0.01020463,
      "balance_loss_clip": 1.04117429,
      "balance_loss_mlp": 1.01415396,
      "epoch": 0.8692358564299886,
      "flos": 18989814556800.0,
      "grad_norm": 1.8391238964722423,
      "language_loss": 0.69922686,
      "learning_rate": 1.7658168535475615e-07,
      "loss": 0.72075593,
      "num_input_tokens_seen": 156406425,
      "step": 7229,
      "time_per_iteration": 2.526275873184204
    },
    {
      "auxiliary_loss_clip": 0.01139718,
      "auxiliary_loss_mlp": 0.01023121,
      "balance_loss_clip": 1.04562187,
      "balance_loss_mlp": 1.01567912,
      "epoch": 0.8693560993206276,
      "flos": 30370839039360.0,
      "grad_norm": 1.5561703106822824,
      "language_loss": 0.64362925,
      "learning_rate": 1.7626179388920948e-07,
      "loss": 0.66525763,
      "num_input_tokens_seen": 156427705,
      "step": 7230,
      "time_per_iteration": 2.58770751953125
    },
    {
      "auxiliary_loss_clip": 0.0113446,
      "auxiliary_loss_mlp": 0.00760462,
      "balance_loss_clip": 1.04429674,
      "balance_loss_mlp": 1.00034571,
      "epoch": 0.8694763422112668,
      "flos": 27200430028800.0,
      "grad_norm": 1.82827266680265,
      "language_loss": 0.80581683,
      "learning_rate": 1.7594217908329866e-07,
      "loss": 0.82476604,
      "num_input_tokens_seen": 156449890,
      "step": 7231,
      "time_per_iteration": 2.6049158573150635
    },
    {
      "auxiliary_loss_clip": 0.01127777,
      "auxiliary_loss_mlp": 0.01025967,
      "balance_loss_clip": 1.04500175,
      "balance_loss_mlp": 1.01966715,
      "epoch": 0.8695965851019059,
      "flos": 26139161767680.0,
      "grad_norm": 1.9231715000523202,
      "language_loss": 0.73999178,
      "learning_rate": 1.7562284098550895e-07,
      "loss": 0.76152921,
      "num_input_tokens_seen": 156469600,
      "step": 7232,
      "time_per_iteration": 2.5568931102752686
    },
    {
      "auxiliary_loss_clip": 0.01033308,
      "auxiliary_loss_mlp": 0.01001143,
      "balance_loss_clip": 1.01137984,
      "balance_loss_mlp": 1.0001235,
      "epoch": 0.8697168279925449,
      "flos": 67332616456320.0,
      "grad_norm": 0.8342698516678518,
      "language_loss": 0.62279356,
      "learning_rate": 1.753037796442838e-07,
      "loss": 0.64313805,
      "num_input_tokens_seen": 156529040,
      "step": 7233,
      "time_per_iteration": 3.8936808109283447
    },
    {
      "auxiliary_loss_clip": 0.01162594,
      "auxiliary_loss_mlp": 0.01024842,
      "balance_loss_clip": 1.04615569,
      "balance_loss_mlp": 1.01771581,
      "epoch": 0.8698370708831841,
      "flos": 19718693337600.0,
      "grad_norm": 2.2405228660512444,
      "language_loss": 0.75602138,
      "learning_rate": 1.74984995108024e-07,
      "loss": 0.77789569,
      "num_input_tokens_seen": 156546970,
      "step": 7234,
      "time_per_iteration": 3.3659839630126953
    },
    {
      "auxiliary_loss_clip": 0.01151038,
      "auxiliary_loss_mlp": 0.01020654,
      "balance_loss_clip": 1.04601598,
      "balance_loss_mlp": 1.01401091,
      "epoch": 0.8699573137738231,
      "flos": 12859971068160.0,
      "grad_norm": 2.0295843381939247,
      "language_loss": 0.83074689,
      "learning_rate": 1.7466648742508981e-07,
      "loss": 0.85246378,
      "num_input_tokens_seen": 156563155,
      "step": 7235,
      "time_per_iteration": 3.2632036209106445
    },
    {
      "auxiliary_loss_clip": 0.01136602,
      "auxiliary_loss_mlp": 0.01022528,
      "balance_loss_clip": 1.04718876,
      "balance_loss_mlp": 1.01541114,
      "epoch": 0.8700775566644622,
      "flos": 17420733768960.0,
      "grad_norm": 1.930149562551747,
      "language_loss": 0.84745878,
      "learning_rate": 1.7434825664379837e-07,
      "loss": 0.86905009,
      "num_input_tokens_seen": 156581660,
      "step": 7236,
      "time_per_iteration": 2.5758790969848633
    },
    {
      "auxiliary_loss_clip": 0.01149112,
      "auxiliary_loss_mlp": 0.01024811,
      "balance_loss_clip": 1.04621935,
      "balance_loss_mlp": 1.0177542,
      "epoch": 0.8701977995551013,
      "flos": 13735221770880.0,
      "grad_norm": 5.281286732995591,
      "language_loss": 0.85937786,
      "learning_rate": 1.740303028124246e-07,
      "loss": 0.88111711,
      "num_input_tokens_seen": 156597720,
      "step": 7237,
      "time_per_iteration": 2.455282688140869
    },
    {
      "auxiliary_loss_clip": 0.01081065,
      "auxiliary_loss_mlp": 0.01022418,
      "balance_loss_clip": 1.03545618,
      "balance_loss_mlp": 1.01576614,
      "epoch": 0.8703180424457404,
      "flos": 30555707362560.0,
      "grad_norm": 2.121458099190664,
      "language_loss": 0.75855213,
      "learning_rate": 1.7371262597920212e-07,
      "loss": 0.77958697,
      "num_input_tokens_seen": 156619780,
      "step": 7238,
      "time_per_iteration": 2.751269578933716
    },
    {
      "auxiliary_loss_clip": 0.0110715,
      "auxiliary_loss_mlp": 0.01029007,
      "balance_loss_clip": 1.04329109,
      "balance_loss_mlp": 1.0223968,
      "epoch": 0.8704382853363795,
      "flos": 19608986223360.0,
      "grad_norm": 1.5337485672625784,
      "language_loss": 0.76407123,
      "learning_rate": 1.7339522619232195e-07,
      "loss": 0.78543282,
      "num_input_tokens_seen": 156638160,
      "step": 7239,
      "time_per_iteration": 2.5772736072540283
    },
    {
      "auxiliary_loss_clip": 0.01141675,
      "auxiliary_loss_mlp": 0.01023666,
      "balance_loss_clip": 1.04315519,
      "balance_loss_mlp": 1.01621556,
      "epoch": 0.8705585282270186,
      "flos": 26613900846720.0,
      "grad_norm": 1.8564530937630506,
      "language_loss": 0.75456315,
      "learning_rate": 1.730781034999338e-07,
      "loss": 0.77621663,
      "num_input_tokens_seen": 156659740,
      "step": 7240,
      "time_per_iteration": 2.614659309387207
    },
    {
      "auxiliary_loss_clip": 0.01159973,
      "auxiliary_loss_mlp": 0.01025625,
      "balance_loss_clip": 1.04741657,
      "balance_loss_mlp": 1.01957548,
      "epoch": 0.8706787711176577,
      "flos": 34090465979520.0,
      "grad_norm": 1.8920661939198817,
      "language_loss": 0.73265946,
      "learning_rate": 1.7276125795014497e-07,
      "loss": 0.75451541,
      "num_input_tokens_seen": 156678190,
      "step": 7241,
      "time_per_iteration": 2.5783803462982178
    },
    {
      "auxiliary_loss_clip": 0.01137328,
      "auxiliary_loss_mlp": 0.01025483,
      "balance_loss_clip": 1.04154158,
      "balance_loss_mlp": 1.01815808,
      "epoch": 0.8707990140082967,
      "flos": 14611513968000.0,
      "grad_norm": 1.8668120412543965,
      "language_loss": 0.67485589,
      "learning_rate": 1.7244468959102054e-07,
      "loss": 0.69648403,
      "num_input_tokens_seen": 156695245,
      "step": 7242,
      "time_per_iteration": 2.5282657146453857
    },
    {
      "auxiliary_loss_clip": 0.01150574,
      "auxiliary_loss_mlp": 0.01026613,
      "balance_loss_clip": 1.04816449,
      "balance_loss_mlp": 1.01949286,
      "epoch": 0.8709192568989359,
      "flos": 20084156265600.0,
      "grad_norm": 2.5127771107444556,
      "language_loss": 0.85430026,
      "learning_rate": 1.7212839847058348e-07,
      "loss": 0.87607217,
      "num_input_tokens_seen": 156710375,
      "step": 7243,
      "time_per_iteration": 2.4851412773132324
    },
    {
      "auxiliary_loss_clip": 0.01097427,
      "auxiliary_loss_mlp": 0.01023864,
      "balance_loss_clip": 1.03967404,
      "balance_loss_mlp": 1.01634228,
      "epoch": 0.871039499789575,
      "flos": 16727083251840.0,
      "grad_norm": 2.091154607841811,
      "language_loss": 0.73694909,
      "learning_rate": 1.718123846368147e-07,
      "loss": 0.75816202,
      "num_input_tokens_seen": 156729420,
      "step": 7244,
      "time_per_iteration": 2.654014825820923
    },
    {
      "auxiliary_loss_clip": 0.01136849,
      "auxiliary_loss_mlp": 0.00759787,
      "balance_loss_clip": 1.04660261,
      "balance_loss_mlp": 1.00028503,
      "epoch": 0.871159742680214,
      "flos": 21068790860160.0,
      "grad_norm": 1.5877359297874662,
      "language_loss": 0.71352363,
      "learning_rate": 1.714966481376543e-07,
      "loss": 0.73248994,
      "num_input_tokens_seen": 156746100,
      "step": 7245,
      "time_per_iteration": 2.5438969135284424
    },
    {
      "auxiliary_loss_clip": 0.01146786,
      "auxiliary_loss_mlp": 0.01028456,
      "balance_loss_clip": 1.04411983,
      "balance_loss_mlp": 1.0216347,
      "epoch": 0.8712799855708532,
      "flos": 28256526731520.0,
      "grad_norm": 2.305690620427168,
      "language_loss": 0.82821083,
      "learning_rate": 1.7118118902099797e-07,
      "loss": 0.84996331,
      "num_input_tokens_seen": 156764185,
      "step": 7246,
      "time_per_iteration": 2.5696041584014893
    },
    {
      "auxiliary_loss_clip": 0.011496,
      "auxiliary_loss_mlp": 0.01027389,
      "balance_loss_clip": 1.04541683,
      "balance_loss_mlp": 1.02093673,
      "epoch": 0.8714002284614922,
      "flos": 22236677665920.0,
      "grad_norm": 3.1693591445143356,
      "language_loss": 0.80885243,
      "learning_rate": 1.7086600733470146e-07,
      "loss": 0.83062226,
      "num_input_tokens_seen": 156784855,
      "step": 7247,
      "time_per_iteration": 2.5106236934661865
    },
    {
      "auxiliary_loss_clip": 0.01144534,
      "auxiliary_loss_mlp": 0.01022294,
      "balance_loss_clip": 1.04486835,
      "balance_loss_mlp": 1.01541829,
      "epoch": 0.8715204713521313,
      "flos": 21431919404160.0,
      "grad_norm": 1.7284481822148026,
      "language_loss": 0.7698341,
      "learning_rate": 1.7055110312657738e-07,
      "loss": 0.79150236,
      "num_input_tokens_seen": 156804350,
      "step": 7248,
      "time_per_iteration": 2.5389859676361084
    },
    {
      "auxiliary_loss_clip": 0.01130089,
      "auxiliary_loss_mlp": 0.01031512,
      "balance_loss_clip": 1.04533267,
      "balance_loss_mlp": 1.0242877,
      "epoch": 0.8716407142427703,
      "flos": 23440439180160.0,
      "grad_norm": 3.2894750535904516,
      "language_loss": 0.74189174,
      "learning_rate": 1.702364764443962e-07,
      "loss": 0.76350784,
      "num_input_tokens_seen": 156823425,
      "step": 7249,
      "time_per_iteration": 3.373906135559082
    },
    {
      "auxiliary_loss_clip": 0.01088272,
      "auxiliary_loss_mlp": 0.01032406,
      "balance_loss_clip": 1.03628683,
      "balance_loss_mlp": 1.0250001,
      "epoch": 0.8717609571334095,
      "flos": 27958683156480.0,
      "grad_norm": 2.1472587217199792,
      "language_loss": 0.72478497,
      "learning_rate": 1.6992212733588685e-07,
      "loss": 0.74599177,
      "num_input_tokens_seen": 156843090,
      "step": 7250,
      "time_per_iteration": 2.6924874782562256
    },
    {
      "auxiliary_loss_clip": 0.01129438,
      "auxiliary_loss_mlp": 0.01025759,
      "balance_loss_clip": 1.0417192,
      "balance_loss_mlp": 1.01889896,
      "epoch": 0.8718812000240486,
      "flos": 25479482538240.0,
      "grad_norm": 1.8335147948321373,
      "language_loss": 0.74926996,
      "learning_rate": 1.6960805584873538e-07,
      "loss": 0.77082193,
      "num_input_tokens_seen": 156861090,
      "step": 7251,
      "time_per_iteration": 2.5589346885681152
    },
    {
      "auxiliary_loss_clip": 0.01112045,
      "auxiliary_loss_mlp": 0.01024346,
      "balance_loss_clip": 1.04202628,
      "balance_loss_mlp": 1.01794767,
      "epoch": 0.8720014429146876,
      "flos": 23403056100480.0,
      "grad_norm": 1.5388293824582664,
      "language_loss": 0.78135675,
      "learning_rate": 1.6929426203058684e-07,
      "loss": 0.80272067,
      "num_input_tokens_seen": 156881515,
      "step": 7252,
      "time_per_iteration": 2.6125285625457764
    },
    {
      "auxiliary_loss_clip": 0.01167783,
      "auxiliary_loss_mlp": 0.00761851,
      "balance_loss_clip": 1.04758859,
      "balance_loss_mlp": 1.00033903,
      "epoch": 0.8721216858053268,
      "flos": 24352821567360.0,
      "grad_norm": 1.9986094983489777,
      "language_loss": 0.8006267,
      "learning_rate": 1.689807459290431e-07,
      "loss": 0.81992304,
      "num_input_tokens_seen": 156900170,
      "step": 7253,
      "time_per_iteration": 2.519680976867676
    },
    {
      "auxiliary_loss_clip": 0.01137177,
      "auxiliary_loss_mlp": 0.01022603,
      "balance_loss_clip": 1.04677093,
      "balance_loss_mlp": 1.01619864,
      "epoch": 0.8722419286959658,
      "flos": 33869687034240.0,
      "grad_norm": 2.102275616940193,
      "language_loss": 0.70788294,
      "learning_rate": 1.6866750759166437e-07,
      "loss": 0.72948074,
      "num_input_tokens_seen": 156920150,
      "step": 7254,
      "time_per_iteration": 2.637498140335083
    },
    {
      "auxiliary_loss_clip": 0.01115605,
      "auxiliary_loss_mlp": 0.01021894,
      "balance_loss_clip": 1.03885508,
      "balance_loss_mlp": 1.01514077,
      "epoch": 0.8723621715866049,
      "flos": 18369385914240.0,
      "grad_norm": 2.2487136727879977,
      "language_loss": 0.77408236,
      "learning_rate": 1.6835454706596865e-07,
      "loss": 0.79545736,
      "num_input_tokens_seen": 156937980,
      "step": 7255,
      "time_per_iteration": 2.539289951324463
    },
    {
      "auxiliary_loss_clip": 0.01163837,
      "auxiliary_loss_mlp": 0.01030522,
      "balance_loss_clip": 1.04870534,
      "balance_loss_mlp": 1.02321792,
      "epoch": 0.8724824144772441,
      "flos": 22013348855040.0,
      "grad_norm": 1.6274022178453251,
      "language_loss": 0.7375716,
      "learning_rate": 1.680418643994317e-07,
      "loss": 0.75951517,
      "num_input_tokens_seen": 156956550,
      "step": 7256,
      "time_per_iteration": 2.47017240524292
    },
    {
      "auxiliary_loss_clip": 0.01055422,
      "auxiliary_loss_mlp": 0.01004375,
      "balance_loss_clip": 1.00833106,
      "balance_loss_mlp": 1.00327194,
      "epoch": 0.8726026573678831,
      "flos": 66698720213760.0,
      "grad_norm": 0.8912306016425868,
      "language_loss": 0.64567095,
      "learning_rate": 1.6772945963948738e-07,
      "loss": 0.66626889,
      "num_input_tokens_seen": 157014715,
      "step": 7257,
      "time_per_iteration": 3.0807254314422607
    },
    {
      "auxiliary_loss_clip": 0.01133651,
      "auxiliary_loss_mlp": 0.01027314,
      "balance_loss_clip": 1.04584181,
      "balance_loss_mlp": 1.02052188,
      "epoch": 0.8727229002585222,
      "flos": 13370908078080.0,
      "grad_norm": 2.244922022086923,
      "language_loss": 0.77496481,
      "learning_rate": 1.6741733283352733e-07,
      "loss": 0.79657441,
      "num_input_tokens_seen": 157032320,
      "step": 7258,
      "time_per_iteration": 2.5111660957336426
    },
    {
      "auxiliary_loss_clip": 0.01106945,
      "auxiliary_loss_mlp": 0.01025759,
      "balance_loss_clip": 1.04205239,
      "balance_loss_mlp": 1.01835299,
      "epoch": 0.8728431431491613,
      "flos": 21796987282560.0,
      "grad_norm": 1.439710622335369,
      "language_loss": 0.83877206,
      "learning_rate": 1.6710548402890102e-07,
      "loss": 0.86009914,
      "num_input_tokens_seen": 157052845,
      "step": 7259,
      "time_per_iteration": 3.486658811569214
    },
    {
      "auxiliary_loss_clip": 0.01168007,
      "auxiliary_loss_mlp": 0.01029148,
      "balance_loss_clip": 1.04883504,
      "balance_loss_mlp": 1.02149796,
      "epoch": 0.8729633860398004,
      "flos": 36173823742080.0,
      "grad_norm": 2.003840232552996,
      "language_loss": 0.66733563,
      "learning_rate": 1.6679391327291527e-07,
      "loss": 0.68930721,
      "num_input_tokens_seen": 157074050,
      "step": 7260,
      "time_per_iteration": 2.5792763233184814
    },
    {
      "auxiliary_loss_clip": 0.01134071,
      "auxiliary_loss_mlp": 0.01024818,
      "balance_loss_clip": 1.04126096,
      "balance_loss_mlp": 1.01827645,
      "epoch": 0.8730836289304394,
      "flos": 16359680989440.0,
      "grad_norm": 4.476415297521812,
      "language_loss": 0.6755181,
      "learning_rate": 1.6648262061283492e-07,
      "loss": 0.69710696,
      "num_input_tokens_seen": 157089350,
      "step": 7261,
      "time_per_iteration": 3.1847312450408936
    },
    {
      "auxiliary_loss_clip": 0.01121706,
      "auxiliary_loss_mlp": 0.01020692,
      "balance_loss_clip": 1.04145992,
      "balance_loss_mlp": 1.01428413,
      "epoch": 0.8732038718210786,
      "flos": 21215126868480.0,
      "grad_norm": 1.8860919031324312,
      "language_loss": 0.73345923,
      "learning_rate": 1.6617160609588353e-07,
      "loss": 0.75488317,
      "num_input_tokens_seen": 157108525,
      "step": 7262,
      "time_per_iteration": 2.5628204345703125
    },
    {
      "auxiliary_loss_clip": 0.01136185,
      "auxiliary_loss_mlp": 0.01023252,
      "balance_loss_clip": 1.04461658,
      "balance_loss_mlp": 1.01649022,
      "epoch": 0.8733241147117177,
      "flos": 16610696208000.0,
      "grad_norm": 1.8933650243993077,
      "language_loss": 0.71737623,
      "learning_rate": 1.6586086976924163e-07,
      "loss": 0.73897058,
      "num_input_tokens_seen": 157124025,
      "step": 7263,
      "time_per_iteration": 2.475778341293335
    },
    {
      "auxiliary_loss_clip": 0.01149066,
      "auxiliary_loss_mlp": 0.01026004,
      "balance_loss_clip": 1.04471374,
      "balance_loss_mlp": 1.01928341,
      "epoch": 0.8734443576023567,
      "flos": 20193935207040.0,
      "grad_norm": 8.937802917663968,
      "language_loss": 0.78281009,
      "learning_rate": 1.6555041168004747e-07,
      "loss": 0.80456078,
      "num_input_tokens_seen": 157143345,
      "step": 7264,
      "time_per_iteration": 2.4973249435424805
    },
    {
      "auxiliary_loss_clip": 0.01131998,
      "auxiliary_loss_mlp": 0.0102517,
      "balance_loss_clip": 1.04360247,
      "balance_loss_mlp": 1.01860702,
      "epoch": 0.8735646004929959,
      "flos": 18041162411520.0,
      "grad_norm": 2.219088872395557,
      "language_loss": 0.68316293,
      "learning_rate": 1.6524023187539715e-07,
      "loss": 0.70473456,
      "num_input_tokens_seen": 157161630,
      "step": 7265,
      "time_per_iteration": 2.531047821044922
    },
    {
      "auxiliary_loss_clip": 0.01136004,
      "auxiliary_loss_mlp": 0.01028326,
      "balance_loss_clip": 1.04479539,
      "balance_loss_mlp": 1.02110755,
      "epoch": 0.873684843383635,
      "flos": 20262344659200.0,
      "grad_norm": 2.1141748876787134,
      "language_loss": 0.74912041,
      "learning_rate": 1.649303304023446e-07,
      "loss": 0.7707637,
      "num_input_tokens_seen": 157181385,
      "step": 7266,
      "time_per_iteration": 2.5353713035583496
    },
    {
      "auxiliary_loss_clip": 0.01121373,
      "auxiliary_loss_mlp": 0.01022918,
      "balance_loss_clip": 1.04833364,
      "balance_loss_mlp": 1.01620018,
      "epoch": 0.873805086274274,
      "flos": 16947287579520.0,
      "grad_norm": 1.7126221275916693,
      "language_loss": 0.78547233,
      "learning_rate": 1.6462070730790246e-07,
      "loss": 0.80691528,
      "num_input_tokens_seen": 157200545,
      "step": 7267,
      "time_per_iteration": 2.5447070598602295
    },
    {
      "auxiliary_loss_clip": 0.01129734,
      "auxiliary_loss_mlp": 0.0102724,
      "balance_loss_clip": 1.04041398,
      "balance_loss_mlp": 1.02030528,
      "epoch": 0.8739253291649132,
      "flos": 18041270152320.0,
      "grad_norm": 2.3813467403970052,
      "language_loss": 0.78627872,
      "learning_rate": 1.6431136263903912e-07,
      "loss": 0.80784851,
      "num_input_tokens_seen": 157219545,
      "step": 7268,
      "time_per_iteration": 2.502641439437866
    },
    {
      "auxiliary_loss_clip": 0.01152097,
      "auxiliary_loss_mlp": 0.00760964,
      "balance_loss_clip": 1.04388785,
      "balance_loss_mlp": 1.0003109,
      "epoch": 0.8740455720555522,
      "flos": 21325085377920.0,
      "grad_norm": 1.8710128239649635,
      "language_loss": 0.73075521,
      "learning_rate": 1.6400229644268282e-07,
      "loss": 0.7498858,
      "num_input_tokens_seen": 157237900,
      "step": 7269,
      "time_per_iteration": 2.4958252906799316
    },
    {
      "auxiliary_loss_clip": 0.01113736,
      "auxiliary_loss_mlp": 0.0103043,
      "balance_loss_clip": 1.04555845,
      "balance_loss_mlp": 1.02325654,
      "epoch": 0.8741658149461913,
      "flos": 15158684822400.0,
      "grad_norm": 2.2180817682497422,
      "language_loss": 0.80776381,
      "learning_rate": 1.6369350876571852e-07,
      "loss": 0.82920551,
      "num_input_tokens_seen": 157256055,
      "step": 7270,
      "time_per_iteration": 2.5387651920318604
    },
    {
      "auxiliary_loss_clip": 0.01103537,
      "auxiliary_loss_mlp": 0.01025569,
      "balance_loss_clip": 1.03918219,
      "balance_loss_mlp": 1.01859784,
      "epoch": 0.8742860578368304,
      "flos": 23039855729280.0,
      "grad_norm": 2.619662895972596,
      "language_loss": 0.81535876,
      "learning_rate": 1.6338499965498874e-07,
      "loss": 0.83664972,
      "num_input_tokens_seen": 157274785,
      "step": 7271,
      "time_per_iteration": 2.6033706665039062
    },
    {
      "auxiliary_loss_clip": 0.01113208,
      "auxiliary_loss_mlp": 0.01027621,
      "balance_loss_clip": 1.03969812,
      "balance_loss_mlp": 1.02022076,
      "epoch": 0.8744063007274695,
      "flos": 28145347159680.0,
      "grad_norm": 1.498136748090697,
      "language_loss": 0.77474028,
      "learning_rate": 1.630767691572943e-07,
      "loss": 0.7961486,
      "num_input_tokens_seen": 157294805,
      "step": 7272,
      "time_per_iteration": 2.630840539932251
    },
    {
      "auxiliary_loss_clip": 0.01037134,
      "auxiliary_loss_mlp": 0.01001892,
      "balance_loss_clip": 1.00822568,
      "balance_loss_mlp": 1.00072408,
      "epoch": 0.8745265436181086,
      "flos": 64034076654720.0,
      "grad_norm": 1.0097073822294282,
      "language_loss": 0.53512752,
      "learning_rate": 1.6276881731939306e-07,
      "loss": 0.55551779,
      "num_input_tokens_seen": 157356695,
      "step": 7273,
      "time_per_iteration": 3.1850881576538086
    },
    {
      "auxiliary_loss_clip": 0.01146517,
      "auxiliary_loss_mlp": 0.01020397,
      "balance_loss_clip": 1.04580176,
      "balance_loss_mlp": 1.0138495,
      "epoch": 0.8746467865087477,
      "flos": 28658618553600.0,
      "grad_norm": 1.9932146921174259,
      "language_loss": 0.75439781,
      "learning_rate": 1.6246114418800193e-07,
      "loss": 0.77606696,
      "num_input_tokens_seen": 157376975,
      "step": 7274,
      "time_per_iteration": 2.5598864555358887
    },
    {
      "auxiliary_loss_clip": 0.0114321,
      "auxiliary_loss_mlp": 0.01026073,
      "balance_loss_clip": 1.0448997,
      "balance_loss_mlp": 1.01889062,
      "epoch": 0.8747670293993868,
      "flos": 23985850268160.0,
      "grad_norm": 1.6832415177500095,
      "language_loss": 0.76897204,
      "learning_rate": 1.6215374980979423e-07,
      "loss": 0.79066479,
      "num_input_tokens_seen": 157397385,
      "step": 7275,
      "time_per_iteration": 3.311440944671631
    },
    {
      "auxiliary_loss_clip": 0.01144712,
      "auxiliary_loss_mlp": 0.01025893,
      "balance_loss_clip": 1.04730201,
      "balance_loss_mlp": 1.01895833,
      "epoch": 0.8748872722900258,
      "flos": 45221624478720.0,
      "grad_norm": 1.9649667744685888,
      "language_loss": 0.68826187,
      "learning_rate": 1.6184663423140133e-07,
      "loss": 0.70996791,
      "num_input_tokens_seen": 157417685,
      "step": 7276,
      "time_per_iteration": 2.6920692920684814
    },
    {
      "auxiliary_loss_clip": 0.01110997,
      "auxiliary_loss_mlp": 0.01031442,
      "balance_loss_clip": 1.04253948,
      "balance_loss_mlp": 1.02467084,
      "epoch": 0.875007515180665,
      "flos": 19754280737280.0,
      "grad_norm": 1.933814001514685,
      "language_loss": 0.64145964,
      "learning_rate": 1.615397974994126e-07,
      "loss": 0.662884,
      "num_input_tokens_seen": 157435490,
      "step": 7277,
      "time_per_iteration": 2.5908565521240234
    },
    {
      "auxiliary_loss_clip": 0.01160331,
      "auxiliary_loss_mlp": 0.01020217,
      "balance_loss_clip": 1.04577363,
      "balance_loss_mlp": 1.01364267,
      "epoch": 0.875127758071304,
      "flos": 22710734386560.0,
      "grad_norm": 2.2644843306513143,
      "language_loss": 0.80635691,
      "learning_rate": 1.6123323966037438e-07,
      "loss": 0.82816237,
      "num_input_tokens_seen": 157454010,
      "step": 7278,
      "time_per_iteration": 2.4663772583007812
    },
    {
      "auxiliary_loss_clip": 0.01162536,
      "auxiliary_loss_mlp": 0.0102438,
      "balance_loss_clip": 1.04779434,
      "balance_loss_mlp": 1.01743031,
      "epoch": 0.8752480009619431,
      "flos": 23403846199680.0,
      "grad_norm": 2.073633729710139,
      "language_loss": 0.78626877,
      "learning_rate": 1.6092696076079216e-07,
      "loss": 0.80813795,
      "num_input_tokens_seen": 157472385,
      "step": 7279,
      "time_per_iteration": 2.469346523284912
    },
    {
      "auxiliary_loss_clip": 0.01108096,
      "auxiliary_loss_mlp": 0.01021204,
      "balance_loss_clip": 1.03960109,
      "balance_loss_mlp": 1.01481736,
      "epoch": 0.8753682438525822,
      "flos": 26213101914240.0,
      "grad_norm": 1.6519055302722412,
      "language_loss": 0.73761839,
      "learning_rate": 1.6062096084712785e-07,
      "loss": 0.75891137,
      "num_input_tokens_seen": 157493735,
      "step": 7280,
      "time_per_iteration": 2.6007862091064453
    },
    {
      "auxiliary_loss_clip": 0.01125058,
      "auxiliary_loss_mlp": 0.0076093,
      "balance_loss_clip": 1.04037356,
      "balance_loss_mlp": 1.00029039,
      "epoch": 0.8754884867432213,
      "flos": 23326745656320.0,
      "grad_norm": 1.7095591901105083,
      "language_loss": 0.70159721,
      "learning_rate": 1.6031523996580098e-07,
      "loss": 0.72045708,
      "num_input_tokens_seen": 157511295,
      "step": 7281,
      "time_per_iteration": 2.5322630405426025
    },
    {
      "auxiliary_loss_clip": 0.01130045,
      "auxiliary_loss_mlp": 0.01028271,
      "balance_loss_clip": 1.04452038,
      "balance_loss_mlp": 1.02176499,
      "epoch": 0.8756087296338604,
      "flos": 12495226412160.0,
      "grad_norm": 1.9069893472463315,
      "language_loss": 0.66435218,
      "learning_rate": 1.6000979816318981e-07,
      "loss": 0.68593532,
      "num_input_tokens_seen": 157529760,
      "step": 7282,
      "time_per_iteration": 2.527578830718994
    },
    {
      "auxiliary_loss_clip": 0.01145394,
      "auxiliary_loss_mlp": 0.01028544,
      "balance_loss_clip": 1.04663348,
      "balance_loss_mlp": 1.02151632,
      "epoch": 0.8757289725244994,
      "flos": 18952898353920.0,
      "grad_norm": 2.2406820481520495,
      "language_loss": 0.74770451,
      "learning_rate": 1.5970463548562886e-07,
      "loss": 0.76944387,
      "num_input_tokens_seen": 157548915,
      "step": 7283,
      "time_per_iteration": 2.4771006107330322
    },
    {
      "auxiliary_loss_clip": 0.01133772,
      "auxiliary_loss_mlp": 0.01026426,
      "balance_loss_clip": 1.04506874,
      "balance_loss_mlp": 1.01986623,
      "epoch": 0.8758492154151386,
      "flos": 25265958140160.0,
      "grad_norm": 1.8291737100528047,
      "language_loss": 0.71282876,
      "learning_rate": 1.5939975197941192e-07,
      "loss": 0.73443067,
      "num_input_tokens_seen": 157570570,
      "step": 7284,
      "time_per_iteration": 2.5453524589538574
    },
    {
      "auxiliary_loss_clip": 0.01037372,
      "auxiliary_loss_mlp": 0.00999912,
      "balance_loss_clip": 1.00897384,
      "balance_loss_mlp": 0.99881554,
      "epoch": 0.8759694583057777,
      "flos": 65571664193280.0,
      "grad_norm": 0.8130133828426268,
      "language_loss": 0.53437507,
      "learning_rate": 1.5909514769078892e-07,
      "loss": 0.55474788,
      "num_input_tokens_seen": 157635675,
      "step": 7285,
      "time_per_iteration": 4.748593330383301
    },
    {
      "auxiliary_loss_clip": 0.01110934,
      "auxiliary_loss_mlp": 0.01026439,
      "balance_loss_clip": 1.04432464,
      "balance_loss_mlp": 1.01989472,
      "epoch": 0.8760897011964167,
      "flos": 25446193608960.0,
      "grad_norm": 1.5058453695565943,
      "language_loss": 0.77789986,
      "learning_rate": 1.5879082266596867e-07,
      "loss": 0.79927361,
      "num_input_tokens_seen": 157657015,
      "step": 7286,
      "time_per_iteration": 2.5773732662200928
    },
    {
      "auxiliary_loss_clip": 0.01126697,
      "auxiliary_loss_mlp": 0.01023593,
      "balance_loss_clip": 1.0377233,
      "balance_loss_mlp": 1.01657736,
      "epoch": 0.8762099440870559,
      "flos": 28984830894720.0,
      "grad_norm": 2.0585478556652306,
      "language_loss": 0.71847177,
      "learning_rate": 1.5848677695111645e-07,
      "loss": 0.73997462,
      "num_input_tokens_seen": 157678615,
      "step": 7287,
      "time_per_iteration": 3.2920982837677
    },
    {
      "auxiliary_loss_clip": 0.01130748,
      "auxiliary_loss_mlp": 0.01026654,
      "balance_loss_clip": 1.04547107,
      "balance_loss_mlp": 1.01922154,
      "epoch": 0.8763301869776949,
      "flos": 21609461352960.0,
      "grad_norm": 2.641360880554864,
      "language_loss": 0.70263755,
      "learning_rate": 1.5818301059235562e-07,
      "loss": 0.72421157,
      "num_input_tokens_seen": 157693790,
      "step": 7288,
      "time_per_iteration": 2.533047914505005
    },
    {
      "auxiliary_loss_clip": 0.01138436,
      "auxiliary_loss_mlp": 0.01028455,
      "balance_loss_clip": 1.04645872,
      "balance_loss_mlp": 1.02117157,
      "epoch": 0.876450429868334,
      "flos": 24644416176000.0,
      "grad_norm": 1.7257988638366073,
      "language_loss": 0.81458044,
      "learning_rate": 1.578795236357684e-07,
      "loss": 0.83624935,
      "num_input_tokens_seen": 157715255,
      "step": 7289,
      "time_per_iteration": 2.574993848800659
    },
    {
      "auxiliary_loss_clip": 0.01136731,
      "auxiliary_loss_mlp": 0.01031104,
      "balance_loss_clip": 1.04705048,
      "balance_loss_mlp": 1.02455091,
      "epoch": 0.8765706727589732,
      "flos": 20260046188800.0,
      "grad_norm": 1.9935633760287863,
      "language_loss": 0.85524219,
      "learning_rate": 1.5757631612739218e-07,
      "loss": 0.87692046,
      "num_input_tokens_seen": 157728800,
      "step": 7290,
      "time_per_iteration": 2.5234549045562744
    },
    {
      "auxiliary_loss_clip": 0.01054949,
      "auxiliary_loss_mlp": 0.01001834,
      "balance_loss_clip": 1.0077523,
      "balance_loss_mlp": 1.00073147,
      "epoch": 0.8766909156496122,
      "flos": 71371165276800.0,
      "grad_norm": 0.7818937620509236,
      "language_loss": 0.6145612,
      "learning_rate": 1.572733881132242e-07,
      "loss": 0.63512909,
      "num_input_tokens_seen": 157789445,
      "step": 7291,
      "time_per_iteration": 3.114102840423584
    },
    {
      "auxiliary_loss_clip": 0.01021873,
      "auxiliary_loss_mlp": 0.0100195,
      "balance_loss_clip": 1.00959802,
      "balance_loss_mlp": 1.00088918,
      "epoch": 0.8768111585402513,
      "flos": 69523490603520.0,
      "grad_norm": 0.785110947082407,
      "language_loss": 0.58556604,
      "learning_rate": 1.5697073963921814e-07,
      "loss": 0.60580426,
      "num_input_tokens_seen": 157848685,
      "step": 7292,
      "time_per_iteration": 3.0806939601898193
    },
    {
      "auxiliary_loss_clip": 0.01151391,
      "auxiliary_loss_mlp": 0.0102772,
      "balance_loss_clip": 1.04738998,
      "balance_loss_mlp": 1.02050185,
      "epoch": 0.8769314014308904,
      "flos": 18838558385280.0,
      "grad_norm": 2.196011523159466,
      "language_loss": 0.84755874,
      "learning_rate": 1.566683707512857e-07,
      "loss": 0.86934984,
      "num_input_tokens_seen": 157866360,
      "step": 7293,
      "time_per_iteration": 2.4787776470184326
    },
    {
      "auxiliary_loss_clip": 0.0113208,
      "auxiliary_loss_mlp": 0.01026241,
      "balance_loss_clip": 1.0433445,
      "balance_loss_mlp": 1.01863003,
      "epoch": 0.8770516443215295,
      "flos": 14976402278400.0,
      "grad_norm": 2.8567140703510048,
      "language_loss": 0.79494619,
      "learning_rate": 1.5636628149529553e-07,
      "loss": 0.81652939,
      "num_input_tokens_seen": 157884150,
      "step": 7294,
      "time_per_iteration": 2.5008068084716797
    },
    {
      "auxiliary_loss_clip": 0.01131987,
      "auxiliary_loss_mlp": 0.01023761,
      "balance_loss_clip": 1.04275203,
      "balance_loss_mlp": 1.01685858,
      "epoch": 0.8771718872121685,
      "flos": 31649654021760.0,
      "grad_norm": 2.122648365650335,
      "language_loss": 0.79613316,
      "learning_rate": 1.560644719170743e-07,
      "loss": 0.81769061,
      "num_input_tokens_seen": 157905020,
      "step": 7295,
      "time_per_iteration": 2.6019372940063477
    },
    {
      "auxiliary_loss_clip": 0.0111842,
      "auxiliary_loss_mlp": 0.01023824,
      "balance_loss_clip": 1.04065895,
      "balance_loss_mlp": 1.0167253,
      "epoch": 0.8772921301028077,
      "flos": 36095466222720.0,
      "grad_norm": 3.701098387421143,
      "language_loss": 0.72053802,
      "learning_rate": 1.5576294206240692e-07,
      "loss": 0.74196041,
      "num_input_tokens_seen": 157924545,
      "step": 7296,
      "time_per_iteration": 2.671232223510742
    },
    {
      "auxiliary_loss_clip": 0.01130549,
      "auxiliary_loss_mlp": 0.01026943,
      "balance_loss_clip": 1.04387164,
      "balance_loss_mlp": 1.01963854,
      "epoch": 0.8774123729934468,
      "flos": 57116961849600.0,
      "grad_norm": 1.705394105452408,
      "language_loss": 0.67963529,
      "learning_rate": 1.5546169197703507e-07,
      "loss": 0.7012102,
      "num_input_tokens_seen": 157950820,
      "step": 7297,
      "time_per_iteration": 2.8512649536132812
    },
    {
      "auxiliary_loss_clip": 0.01136494,
      "auxiliary_loss_mlp": 0.0102368,
      "balance_loss_clip": 1.04106116,
      "balance_loss_mlp": 1.01674235,
      "epoch": 0.8775326158840858,
      "flos": 23914495900800.0,
      "grad_norm": 2.439962239826139,
      "language_loss": 0.77431023,
      "learning_rate": 1.5516072170665774e-07,
      "loss": 0.79591197,
      "num_input_tokens_seen": 157968790,
      "step": 7298,
      "time_per_iteration": 2.540191650390625
    },
    {
      "auxiliary_loss_clip": 0.01151287,
      "auxiliary_loss_mlp": 0.01028742,
      "balance_loss_clip": 1.04690433,
      "balance_loss_mlp": 1.02229011,
      "epoch": 0.877652858774725,
      "flos": 17123285243520.0,
      "grad_norm": 1.7875150525199515,
      "language_loss": 0.86916173,
      "learning_rate": 1.5486003129693214e-07,
      "loss": 0.890962,
      "num_input_tokens_seen": 157986155,
      "step": 7299,
      "time_per_iteration": 2.475304365158081
    },
    {
      "auxiliary_loss_clip": 0.01152174,
      "auxiliary_loss_mlp": 0.01028693,
      "balance_loss_clip": 1.04714346,
      "balance_loss_mlp": 1.02168941,
      "epoch": 0.877773101665364,
      "flos": 16508961912960.0,
      "grad_norm": 2.6655416377275216,
      "language_loss": 0.7793721,
      "learning_rate": 1.545596207934725e-07,
      "loss": 0.80118072,
      "num_input_tokens_seen": 158004640,
      "step": 7300,
      "time_per_iteration": 3.2360286712646484
    },
    {
      "auxiliary_loss_clip": 0.01128124,
      "auxiliary_loss_mlp": 0.0102841,
      "balance_loss_clip": 1.04224193,
      "balance_loss_mlp": 1.02180922,
      "epoch": 0.8778933445560031,
      "flos": 22053209973120.0,
      "grad_norm": 1.6754033726673363,
      "language_loss": 0.77737141,
      "learning_rate": 1.5425949024185147e-07,
      "loss": 0.79893672,
      "num_input_tokens_seen": 158024665,
      "step": 7301,
      "time_per_iteration": 2.535475254058838
    },
    {
      "auxiliary_loss_clip": 0.01134381,
      "auxiliary_loss_mlp": 0.01026536,
      "balance_loss_clip": 1.04126191,
      "balance_loss_mlp": 1.02007151,
      "epoch": 0.8780135874466423,
      "flos": 22564757514240.0,
      "grad_norm": 2.5140450667222,
      "language_loss": 0.67463326,
      "learning_rate": 1.5395963968759818e-07,
      "loss": 0.69624245,
      "num_input_tokens_seen": 158044940,
      "step": 7302,
      "time_per_iteration": 2.529075860977173
    },
    {
      "auxiliary_loss_clip": 0.01132239,
      "auxiliary_loss_mlp": 0.01022995,
      "balance_loss_clip": 1.04041219,
      "balance_loss_mlp": 1.01619411,
      "epoch": 0.8781338303372813,
      "flos": 61531999073280.0,
      "grad_norm": 1.6287607465800513,
      "language_loss": 0.64327806,
      "learning_rate": 1.536600691761998e-07,
      "loss": 0.66483045,
      "num_input_tokens_seen": 158070770,
      "step": 7303,
      "time_per_iteration": 2.885249614715576
    },
    {
      "auxiliary_loss_clip": 0.01123448,
      "auxiliary_loss_mlp": 0.01031019,
      "balance_loss_clip": 1.04446232,
      "balance_loss_mlp": 1.02451575,
      "epoch": 0.8782540732279204,
      "flos": 22674751937280.0,
      "grad_norm": 3.9522549549295474,
      "language_loss": 0.71360546,
      "learning_rate": 1.5336077875310084e-07,
      "loss": 0.7351501,
      "num_input_tokens_seen": 158089995,
      "step": 7304,
      "time_per_iteration": 2.5707271099090576
    },
    {
      "auxiliary_loss_clip": 0.01111218,
      "auxiliary_loss_mlp": 0.01024384,
      "balance_loss_clip": 1.04138386,
      "balance_loss_mlp": 1.01783919,
      "epoch": 0.8783743161185595,
      "flos": 16070348937600.0,
      "grad_norm": 2.0386429423878707,
      "language_loss": 0.73834884,
      "learning_rate": 1.5306176846370321e-07,
      "loss": 0.75970483,
      "num_input_tokens_seen": 158108140,
      "step": 7305,
      "time_per_iteration": 2.557757616043091
    },
    {
      "auxiliary_loss_clip": 0.011393,
      "auxiliary_loss_mlp": 0.01023961,
      "balance_loss_clip": 1.04238963,
      "balance_loss_mlp": 1.01698732,
      "epoch": 0.8784945590091986,
      "flos": 26067879227520.0,
      "grad_norm": 2.2803253980253846,
      "language_loss": 0.74224722,
      "learning_rate": 1.5276303835336712e-07,
      "loss": 0.7638799,
      "num_input_tokens_seen": 158128680,
      "step": 7306,
      "time_per_iteration": 2.5732719898223877
    },
    {
      "auxiliary_loss_clip": 0.01046401,
      "auxiliary_loss_mlp": 0.01001765,
      "balance_loss_clip": 1.00811589,
      "balance_loss_mlp": 1.00065076,
      "epoch": 0.8786148018998376,
      "flos": 62720643939840.0,
      "grad_norm": 0.7588595532118178,
      "language_loss": 0.53583342,
      "learning_rate": 1.524645884674094e-07,
      "loss": 0.55631512,
      "num_input_tokens_seen": 158185610,
      "step": 7307,
      "time_per_iteration": 3.087132453918457
    },
    {
      "auxiliary_loss_clip": 0.01166762,
      "auxiliary_loss_mlp": 0.00761619,
      "balance_loss_clip": 1.04889488,
      "balance_loss_mlp": 1.00029969,
      "epoch": 0.8787350447904768,
      "flos": 21652734263040.0,
      "grad_norm": 2.228070257230499,
      "language_loss": 0.78813982,
      "learning_rate": 1.521664188511047e-07,
      "loss": 0.80742365,
      "num_input_tokens_seen": 158205635,
      "step": 7308,
      "time_per_iteration": 2.4924256801605225
    },
    {
      "auxiliary_loss_clip": 0.01136425,
      "auxiliary_loss_mlp": 0.00760642,
      "balance_loss_clip": 1.04763997,
      "balance_loss_mlp": 1.00030541,
      "epoch": 0.8788552876811159,
      "flos": 25478476957440.0,
      "grad_norm": 2.644741572598983,
      "language_loss": 0.80026078,
      "learning_rate": 1.518685295496851e-07,
      "loss": 0.81923151,
      "num_input_tokens_seen": 158223495,
      "step": 7309,
      "time_per_iteration": 2.5555264949798584
    },
    {
      "auxiliary_loss_clip": 0.0114717,
      "auxiliary_loss_mlp": 0.01021056,
      "balance_loss_clip": 1.04284692,
      "balance_loss_mlp": 1.01485062,
      "epoch": 0.8789755305717549,
      "flos": 22310222762880.0,
      "grad_norm": 1.6320634304775283,
      "language_loss": 0.8519001,
      "learning_rate": 1.5157092060833975e-07,
      "loss": 0.87358236,
      "num_input_tokens_seen": 158243145,
      "step": 7310,
      "time_per_iteration": 2.5036461353302
    },
    {
      "auxiliary_loss_clip": 0.01130123,
      "auxiliary_loss_mlp": 0.01026148,
      "balance_loss_clip": 1.04393566,
      "balance_loss_mlp": 1.01952553,
      "epoch": 0.879095773462394,
      "flos": 29310971408640.0,
      "grad_norm": 1.5942504863525593,
      "language_loss": 0.65886426,
      "learning_rate": 1.5127359207221658e-07,
      "loss": 0.68042696,
      "num_input_tokens_seen": 158262625,
      "step": 7311,
      "time_per_iteration": 4.176614046096802
    },
    {
      "auxiliary_loss_clip": 0.01079937,
      "auxiliary_loss_mlp": 0.01028004,
      "balance_loss_clip": 1.03321862,
      "balance_loss_mlp": 1.02026165,
      "epoch": 0.8792160163530331,
      "flos": 16690023394560.0,
      "grad_norm": 1.875180700518652,
      "language_loss": 0.7339679,
      "learning_rate": 1.5097654398641923e-07,
      "loss": 0.75504732,
      "num_input_tokens_seen": 158280530,
      "step": 7312,
      "time_per_iteration": 2.603555679321289
    },
    {
      "auxiliary_loss_clip": 0.01151224,
      "auxiliary_loss_mlp": 0.01028652,
      "balance_loss_clip": 1.04686475,
      "balance_loss_mlp": 1.02225065,
      "epoch": 0.8793362592436722,
      "flos": 24499301230080.0,
      "grad_norm": 1.3967297493337751,
      "language_loss": 0.73043603,
      "learning_rate": 1.5067977639601014e-07,
      "loss": 0.75223482,
      "num_input_tokens_seen": 158303290,
      "step": 7313,
      "time_per_iteration": 3.279355764389038
    },
    {
      "auxiliary_loss_clip": 0.01130936,
      "auxiliary_loss_mlp": 0.01020292,
      "balance_loss_clip": 1.04373348,
      "balance_loss_mlp": 1.01410484,
      "epoch": 0.8794565021343113,
      "flos": 14538399834240.0,
      "grad_norm": 2.0413218217624527,
      "language_loss": 0.70648813,
      "learning_rate": 1.5038328934600864e-07,
      "loss": 0.7280004,
      "num_input_tokens_seen": 158319925,
      "step": 7314,
      "time_per_iteration": 2.4867377281188965
    },
    {
      "auxiliary_loss_clip": 0.01134447,
      "auxiliary_loss_mlp": 0.01025711,
      "balance_loss_clip": 1.0452311,
      "balance_loss_mlp": 1.01901722,
      "epoch": 0.8795767450249504,
      "flos": 39530286224640.0,
      "grad_norm": 1.9543542302356374,
      "language_loss": 0.69627976,
      "learning_rate": 1.5008708288139161e-07,
      "loss": 0.71788132,
      "num_input_tokens_seen": 158342285,
      "step": 7315,
      "time_per_iteration": 2.677274703979492
    },
    {
      "auxiliary_loss_clip": 0.01146673,
      "auxiliary_loss_mlp": 0.01024457,
      "balance_loss_clip": 1.04635084,
      "balance_loss_mlp": 1.01726866,
      "epoch": 0.8796969879155895,
      "flos": 22960672197120.0,
      "grad_norm": 2.5716493356786745,
      "language_loss": 0.73158562,
      "learning_rate": 1.497911570470931e-07,
      "loss": 0.75329691,
      "num_input_tokens_seen": 158362290,
      "step": 7316,
      "time_per_iteration": 2.5011093616485596
    },
    {
      "auxiliary_loss_clip": 0.01110968,
      "auxiliary_loss_mlp": 0.01023288,
      "balance_loss_clip": 1.04251266,
      "balance_loss_mlp": 1.01626062,
      "epoch": 0.8798172308062285,
      "flos": 28362427004160.0,
      "grad_norm": 1.6101021079444675,
      "language_loss": 0.85820913,
      "learning_rate": 1.494955118880048e-07,
      "loss": 0.87955177,
      "num_input_tokens_seen": 158383275,
      "step": 7317,
      "time_per_iteration": 2.621006488800049
    },
    {
      "auxiliary_loss_clip": 0.01148474,
      "auxiliary_loss_mlp": 0.01025003,
      "balance_loss_clip": 1.04508233,
      "balance_loss_mlp": 1.01808858,
      "epoch": 0.8799374736968677,
      "flos": 23988974751360.0,
      "grad_norm": 1.7523627014992615,
      "language_loss": 0.72563553,
      "learning_rate": 1.4920014744897634e-07,
      "loss": 0.74737024,
      "num_input_tokens_seen": 158402690,
      "step": 7318,
      "time_per_iteration": 2.509491205215454
    },
    {
      "auxiliary_loss_clip": 0.01126111,
      "auxiliary_loss_mlp": 0.01031995,
      "balance_loss_clip": 1.0434339,
      "balance_loss_mlp": 1.02503633,
      "epoch": 0.8800577165875068,
      "flos": 25630271832960.0,
      "grad_norm": 1.6560313201568533,
      "language_loss": 0.86032712,
      "learning_rate": 1.4890506377481392e-07,
      "loss": 0.88190824,
      "num_input_tokens_seen": 158421780,
      "step": 7319,
      "time_per_iteration": 2.562244176864624
    },
    {
      "auxiliary_loss_clip": 0.01086751,
      "auxiliary_loss_mlp": 0.01023624,
      "balance_loss_clip": 1.03943181,
      "balance_loss_mlp": 1.01763976,
      "epoch": 0.8801779594781458,
      "flos": 23440331439360.0,
      "grad_norm": 1.4042037413526858,
      "language_loss": 0.63861346,
      "learning_rate": 1.486102609102815e-07,
      "loss": 0.6597172,
      "num_input_tokens_seen": 158442330,
      "step": 7320,
      "time_per_iteration": 2.628953218460083
    },
    {
      "auxiliary_loss_clip": 0.01130155,
      "auxiliary_loss_mlp": 0.01026829,
      "balance_loss_clip": 1.04420698,
      "balance_loss_mlp": 1.02042103,
      "epoch": 0.880298202368785,
      "flos": 11508580656000.0,
      "grad_norm": 2.608853108470578,
      "language_loss": 0.8571862,
      "learning_rate": 1.483157389001004e-07,
      "loss": 0.87875605,
      "num_input_tokens_seen": 158459890,
      "step": 7321,
      "time_per_iteration": 2.49715256690979
    },
    {
      "auxiliary_loss_clip": 0.01133187,
      "auxiliary_loss_mlp": 0.01028905,
      "balance_loss_clip": 1.04163742,
      "balance_loss_mlp": 1.02169561,
      "epoch": 0.880418445259424,
      "flos": 22671447886080.0,
      "grad_norm": 4.746455685420052,
      "language_loss": 0.78898871,
      "learning_rate": 1.4802149778894933e-07,
      "loss": 0.81060958,
      "num_input_tokens_seen": 158478680,
      "step": 7322,
      "time_per_iteration": 2.5328519344329834
    },
    {
      "auxiliary_loss_clip": 0.01139603,
      "auxiliary_loss_mlp": 0.01024107,
      "balance_loss_clip": 1.0412612,
      "balance_loss_mlp": 1.01769984,
      "epoch": 0.8805386881500631,
      "flos": 20522158709760.0,
      "grad_norm": 1.8254619345139635,
      "language_loss": 0.87679756,
      "learning_rate": 1.4772753762146484e-07,
      "loss": 0.89843464,
      "num_input_tokens_seen": 158497935,
      "step": 7323,
      "time_per_iteration": 2.4870591163635254
    },
    {
      "auxiliary_loss_clip": 0.01141018,
      "auxiliary_loss_mlp": 0.01027423,
      "balance_loss_clip": 1.04275358,
      "balance_loss_mlp": 1.02038705,
      "epoch": 0.8806589310407023,
      "flos": 36538891620480.0,
      "grad_norm": 1.7526352216504806,
      "language_loss": 0.70505375,
      "learning_rate": 1.474338584422401e-07,
      "loss": 0.72673815,
      "num_input_tokens_seen": 158523145,
      "step": 7324,
      "time_per_iteration": 2.6090919971466064
    },
    {
      "auxiliary_loss_clip": 0.0114664,
      "auxiliary_loss_mlp": 0.01023081,
      "balance_loss_clip": 1.04683387,
      "balance_loss_mlp": 1.01624727,
      "epoch": 0.8807791739313413,
      "flos": 23440187784960.0,
      "grad_norm": 1.626497069638088,
      "language_loss": 0.75914007,
      "learning_rate": 1.4714046029582595e-07,
      "loss": 0.7808373,
      "num_input_tokens_seen": 158542210,
      "step": 7325,
      "time_per_iteration": 2.5231151580810547
    },
    {
      "auxiliary_loss_clip": 0.01123448,
      "auxiliary_loss_mlp": 0.0102056,
      "balance_loss_clip": 1.04172552,
      "balance_loss_mlp": 1.01374125,
      "epoch": 0.8808994168219804,
      "flos": 25956843310080.0,
      "grad_norm": 1.6735718278256764,
      "language_loss": 0.75846261,
      "learning_rate": 1.46847343226731e-07,
      "loss": 0.7799027,
      "num_input_tokens_seen": 158563250,
      "step": 7326,
      "time_per_iteration": 3.441106081008911
    },
    {
      "auxiliary_loss_clip": 0.01153632,
      "auxiliary_loss_mlp": 0.01031115,
      "balance_loss_clip": 1.04739606,
      "balance_loss_mlp": 1.02417111,
      "epoch": 0.8810196597126195,
      "flos": 17092079303040.0,
      "grad_norm": 1.8671867086991916,
      "language_loss": 0.69585323,
      "learning_rate": 1.465545072794203e-07,
      "loss": 0.71770072,
      "num_input_tokens_seen": 158581125,
      "step": 7327,
      "time_per_iteration": 2.4602608680725098
    },
    {
      "auxiliary_loss_clip": 0.01098591,
      "auxiliary_loss_mlp": 0.01023071,
      "balance_loss_clip": 1.04326129,
      "balance_loss_mlp": 1.01655972,
      "epoch": 0.8811399026032586,
      "flos": 23002831785600.0,
      "grad_norm": 1.5449566819734843,
      "language_loss": 0.75728607,
      "learning_rate": 1.4626195249831774e-07,
      "loss": 0.77850276,
      "num_input_tokens_seen": 158602025,
      "step": 7328,
      "time_per_iteration": 2.6025633811950684
    },
    {
      "auxiliary_loss_clip": 0.0114608,
      "auxiliary_loss_mlp": 0.01026724,
      "balance_loss_clip": 1.04378355,
      "balance_loss_mlp": 1.01986611,
      "epoch": 0.8812601454938976,
      "flos": 14463813242880.0,
      "grad_norm": 1.7666090464176314,
      "language_loss": 0.71985507,
      "learning_rate": 1.4596967892780244e-07,
      "loss": 0.74158311,
      "num_input_tokens_seen": 158618355,
      "step": 7329,
      "time_per_iteration": 2.4452786445617676
    },
    {
      "auxiliary_loss_clip": 0.01164007,
      "auxiliary_loss_mlp": 0.01028628,
      "balance_loss_clip": 1.0489583,
      "balance_loss_mlp": 1.02167773,
      "epoch": 0.8813803883845368,
      "flos": 22493223578880.0,
      "grad_norm": 1.656981427546696,
      "language_loss": 0.74887705,
      "learning_rate": 1.4567768661221314e-07,
      "loss": 0.77080339,
      "num_input_tokens_seen": 158638925,
      "step": 7330,
      "time_per_iteration": 2.4883689880371094
    },
    {
      "auxiliary_loss_clip": 0.01154075,
      "auxiliary_loss_mlp": 0.00760974,
      "balance_loss_clip": 1.04746878,
      "balance_loss_mlp": 1.00027955,
      "epoch": 0.8815006312751759,
      "flos": 21506901045120.0,
      "grad_norm": 1.8672902227794987,
      "language_loss": 0.74125326,
      "learning_rate": 1.4538597559584442e-07,
      "loss": 0.76040375,
      "num_input_tokens_seen": 158656715,
      "step": 7331,
      "time_per_iteration": 2.5114455223083496
    },
    {
      "auxiliary_loss_clip": 0.01131464,
      "auxiliary_loss_mlp": 0.01030845,
      "balance_loss_clip": 1.04435158,
      "balance_loss_mlp": 1.02398419,
      "epoch": 0.8816208741658149,
      "flos": 22784566792320.0,
      "grad_norm": 1.9192033410166822,
      "language_loss": 0.78914511,
      "learning_rate": 1.4509454592294823e-07,
      "loss": 0.81076819,
      "num_input_tokens_seen": 158677200,
      "step": 7332,
      "time_per_iteration": 2.535228729248047
    },
    {
      "auxiliary_loss_clip": 0.01121223,
      "auxiliary_loss_mlp": 0.00760688,
      "balance_loss_clip": 1.04388905,
      "balance_loss_mlp": 1.00031328,
      "epoch": 0.8817411170564541,
      "flos": 17779409026560.0,
      "grad_norm": 1.8987620234170015,
      "language_loss": 0.79005396,
      "learning_rate": 1.448033976377354e-07,
      "loss": 0.808873,
      "num_input_tokens_seen": 158692185,
      "step": 7333,
      "time_per_iteration": 2.526254653930664
    },
    {
      "auxiliary_loss_clip": 0.0114954,
      "auxiliary_loss_mlp": 0.01026404,
      "balance_loss_clip": 1.04388452,
      "balance_loss_mlp": 1.01962638,
      "epoch": 0.8818613599470931,
      "flos": 18551812112640.0,
      "grad_norm": 1.9585159626882462,
      "language_loss": 0.74066234,
      "learning_rate": 1.445125307843713e-07,
      "loss": 0.76242179,
      "num_input_tokens_seen": 158710410,
      "step": 7334,
      "time_per_iteration": 2.486567497253418
    },
    {
      "auxiliary_loss_clip": 0.01145798,
      "auxiliary_loss_mlp": 0.01026963,
      "balance_loss_clip": 1.04640961,
      "balance_loss_mlp": 1.02029884,
      "epoch": 0.8819816028377322,
      "flos": 27599792417280.0,
      "grad_norm": 1.6689588857647721,
      "language_loss": 0.75491405,
      "learning_rate": 1.442219454069813e-07,
      "loss": 0.77664173,
      "num_input_tokens_seen": 158731435,
      "step": 7335,
      "time_per_iteration": 2.539588451385498
    },
    {
      "auxiliary_loss_clip": 0.01110073,
      "auxiliary_loss_mlp": 0.01023698,
      "balance_loss_clip": 1.0416193,
      "balance_loss_mlp": 1.01728153,
      "epoch": 0.8821018457283714,
      "flos": 23404600385280.0,
      "grad_norm": 1.7777051803667316,
      "language_loss": 0.66295779,
      "learning_rate": 1.4393164154964676e-07,
      "loss": 0.68429554,
      "num_input_tokens_seen": 158750965,
      "step": 7336,
      "time_per_iteration": 2.5931131839752197
    },
    {
      "auxiliary_loss_clip": 0.01144744,
      "auxiliary_loss_mlp": 0.01024014,
      "balance_loss_clip": 1.04592383,
      "balance_loss_mlp": 1.01799345,
      "epoch": 0.8822220886190104,
      "flos": 29132459792640.0,
      "grad_norm": 1.6999958714610666,
      "language_loss": 0.93961334,
      "learning_rate": 1.4364161925640649e-07,
      "loss": 0.96130091,
      "num_input_tokens_seen": 158772365,
      "step": 7337,
      "time_per_iteration": 4.095608472824097
    },
    {
      "auxiliary_loss_clip": 0.01163246,
      "auxiliary_loss_mlp": 0.01020944,
      "balance_loss_clip": 1.0478276,
      "balance_loss_mlp": 1.01453948,
      "epoch": 0.8823423315096495,
      "flos": 20485422074880.0,
      "grad_norm": 1.8294444673109223,
      "language_loss": 0.85090828,
      "learning_rate": 1.4335187857125663e-07,
      "loss": 0.87275016,
      "num_input_tokens_seen": 158791065,
      "step": 7338,
      "time_per_iteration": 2.448507070541382
    },
    {
      "auxiliary_loss_clip": 0.01152192,
      "auxiliary_loss_mlp": 0.01024869,
      "balance_loss_clip": 1.04668427,
      "balance_loss_mlp": 1.01825547,
      "epoch": 0.8824625744002886,
      "flos": 24206377818240.0,
      "grad_norm": 1.7269263415583849,
      "language_loss": 0.75605804,
      "learning_rate": 1.4306241953815023e-07,
      "loss": 0.77782863,
      "num_input_tokens_seen": 158812125,
      "step": 7339,
      "time_per_iteration": 3.3029303550720215
    },
    {
      "auxiliary_loss_clip": 0.01148819,
      "auxiliary_loss_mlp": 0.01029668,
      "balance_loss_clip": 1.04492688,
      "balance_loss_mlp": 1.02243829,
      "epoch": 0.8825828172909277,
      "flos": 24679500785280.0,
      "grad_norm": 1.8557441176959373,
      "language_loss": 0.70574415,
      "learning_rate": 1.4277324220099862e-07,
      "loss": 0.72752905,
      "num_input_tokens_seen": 158834035,
      "step": 7340,
      "time_per_iteration": 2.541755199432373
    },
    {
      "auxiliary_loss_clip": 0.01114627,
      "auxiliary_loss_mlp": 0.0102882,
      "balance_loss_clip": 1.03919482,
      "balance_loss_mlp": 1.02201319,
      "epoch": 0.8827030601815667,
      "flos": 22456163721600.0,
      "grad_norm": 1.747398887515792,
      "language_loss": 0.73767161,
      "learning_rate": 1.4248434660366938e-07,
      "loss": 0.7591061,
      "num_input_tokens_seen": 158853510,
      "step": 7341,
      "time_per_iteration": 2.552534818649292
    },
    {
      "auxiliary_loss_clip": 0.01136443,
      "auxiliary_loss_mlp": 0.01025697,
      "balance_loss_clip": 1.04609585,
      "balance_loss_mlp": 1.01908731,
      "epoch": 0.8828233030722058,
      "flos": 19865639877120.0,
      "grad_norm": 1.7630697276135998,
      "language_loss": 0.70652723,
      "learning_rate": 1.4219573278998808e-07,
      "loss": 0.72814858,
      "num_input_tokens_seen": 158871970,
      "step": 7342,
      "time_per_iteration": 2.518187999725342
    },
    {
      "auxiliary_loss_clip": 0.01131631,
      "auxiliary_loss_mlp": 0.01028234,
      "balance_loss_clip": 1.04168737,
      "balance_loss_mlp": 1.02079821,
      "epoch": 0.882943545962845,
      "flos": 39347213581440.0,
      "grad_norm": 2.21319245036463,
      "language_loss": 0.6453948,
      "learning_rate": 1.4190740080373685e-07,
      "loss": 0.6669935,
      "num_input_tokens_seen": 158892250,
      "step": 7343,
      "time_per_iteration": 2.662992477416992
    },
    {
      "auxiliary_loss_clip": 0.01107075,
      "auxiliary_loss_mlp": 0.01027694,
      "balance_loss_clip": 1.04293728,
      "balance_loss_mlp": 1.02068162,
      "epoch": 0.883063788853484,
      "flos": 19054524908160.0,
      "grad_norm": 1.9128470448761485,
      "language_loss": 0.8428821,
      "learning_rate": 1.4161935068865538e-07,
      "loss": 0.8642298,
      "num_input_tokens_seen": 158907395,
      "step": 7344,
      "time_per_iteration": 2.5572757720947266
    },
    {
      "auxiliary_loss_clip": 0.01163098,
      "auxiliary_loss_mlp": 0.01023069,
      "balance_loss_clip": 1.04748154,
      "balance_loss_mlp": 1.01597643,
      "epoch": 0.8831840317441231,
      "flos": 18733196816640.0,
      "grad_norm": 1.9884806716925736,
      "language_loss": 0.75598997,
      "learning_rate": 1.4133158248844113e-07,
      "loss": 0.77785164,
      "num_input_tokens_seen": 158926300,
      "step": 7345,
      "time_per_iteration": 2.451802968978882
    },
    {
      "auxiliary_loss_clip": 0.01121531,
      "auxiliary_loss_mlp": 0.01026553,
      "balance_loss_clip": 1.04423833,
      "balance_loss_mlp": 1.0192039,
      "epoch": 0.8833042746347622,
      "flos": 26827712553600.0,
      "grad_norm": 1.9815866041175771,
      "language_loss": 0.7358411,
      "learning_rate": 1.4104409624674785e-07,
      "loss": 0.75732195,
      "num_input_tokens_seen": 158946085,
      "step": 7346,
      "time_per_iteration": 2.5765669345855713
    },
    {
      "auxiliary_loss_clip": 0.01150279,
      "auxiliary_loss_mlp": 0.01025849,
      "balance_loss_clip": 1.04803395,
      "balance_loss_mlp": 1.01897049,
      "epoch": 0.8834245175254013,
      "flos": 26104077158400.0,
      "grad_norm": 1.6847689682890183,
      "language_loss": 0.78214753,
      "learning_rate": 1.407568920071873e-07,
      "loss": 0.80390882,
      "num_input_tokens_seen": 158964950,
      "step": 7347,
      "time_per_iteration": 2.5381319522857666
    },
    {
      "auxiliary_loss_clip": 0.01167442,
      "auxiliary_loss_mlp": 0.01025769,
      "balance_loss_clip": 1.04882574,
      "balance_loss_mlp": 1.01834261,
      "epoch": 0.8835447604160404,
      "flos": 30629036977920.0,
      "grad_norm": 2.321426649769578,
      "language_loss": 0.68081951,
      "learning_rate": 1.4046996981332782e-07,
      "loss": 0.7027517,
      "num_input_tokens_seen": 158984835,
      "step": 7348,
      "time_per_iteration": 2.5215439796447754
    },
    {
      "auxiliary_loss_clip": 0.01121036,
      "auxiliary_loss_mlp": 0.01027281,
      "balance_loss_clip": 1.04358435,
      "balance_loss_mlp": 1.02011085,
      "epoch": 0.8836650033066795,
      "flos": 24718356322560.0,
      "grad_norm": 1.7675047709782992,
      "language_loss": 0.78017128,
      "learning_rate": 1.4018332970869516e-07,
      "loss": 0.80165446,
      "num_input_tokens_seen": 159002775,
      "step": 7349,
      "time_per_iteration": 2.583707571029663
    },
    {
      "auxiliary_loss_clip": 0.011261,
      "auxiliary_loss_mlp": 0.01025604,
      "balance_loss_clip": 1.04228497,
      "balance_loss_mlp": 1.01844525,
      "epoch": 0.8837852461973186,
      "flos": 25413371556480.0,
      "grad_norm": 1.9032862358436824,
      "language_loss": 0.84903371,
      "learning_rate": 1.398969717367733e-07,
      "loss": 0.87055075,
      "num_input_tokens_seen": 159024100,
      "step": 7350,
      "time_per_iteration": 2.5595128536224365
    },
    {
      "auxiliary_loss_clip": 0.01105732,
      "auxiliary_loss_mlp": 0.01026767,
      "balance_loss_clip": 1.04614639,
      "balance_loss_mlp": 1.02020407,
      "epoch": 0.8839054890879576,
      "flos": 17822574195840.0,
      "grad_norm": 1.6291694021849228,
      "language_loss": 0.7616117,
      "learning_rate": 1.396108959410014e-07,
      "loss": 0.78293669,
      "num_input_tokens_seen": 159043315,
      "step": 7351,
      "time_per_iteration": 2.562119960784912
    },
    {
      "auxiliary_loss_clip": 0.01147806,
      "auxiliary_loss_mlp": 0.00761121,
      "balance_loss_clip": 1.04664159,
      "balance_loss_mlp": 1.00028276,
      "epoch": 0.8840257319785968,
      "flos": 23769021818880.0,
      "grad_norm": 1.5707513263604735,
      "language_loss": 0.81177318,
      "learning_rate": 1.3932510236477745e-07,
      "loss": 0.83086246,
      "num_input_tokens_seen": 159063985,
      "step": 7352,
      "time_per_iteration": 3.251771926879883
    },
    {
      "auxiliary_loss_clip": 0.01147163,
      "auxiliary_loss_mlp": 0.0102789,
      "balance_loss_clip": 1.04332614,
      "balance_loss_mlp": 1.02079654,
      "epoch": 0.8841459748692359,
      "flos": 29059776622080.0,
      "grad_norm": 1.6565967636463592,
      "language_loss": 0.55732012,
      "learning_rate": 1.3903959105145636e-07,
      "loss": 0.57907069,
      "num_input_tokens_seen": 159084475,
      "step": 7353,
      "time_per_iteration": 2.550734281539917
    },
    {
      "auxiliary_loss_clip": 0.01163297,
      "auxiliary_loss_mlp": 0.01029623,
      "balance_loss_clip": 1.0483191,
      "balance_loss_mlp": 1.0230248,
      "epoch": 0.8842662177598749,
      "flos": 24311523905280.0,
      "grad_norm": 1.8964161696598811,
      "language_loss": 0.82971114,
      "learning_rate": 1.387543620443492e-07,
      "loss": 0.85164034,
      "num_input_tokens_seen": 159101320,
      "step": 7354,
      "time_per_iteration": 2.4645819664001465
    },
    {
      "auxiliary_loss_clip": 0.01161973,
      "auxiliary_loss_mlp": 0.01021499,
      "balance_loss_clip": 1.04799008,
      "balance_loss_mlp": 1.01476407,
      "epoch": 0.8843864606505141,
      "flos": 25007867942400.0,
      "grad_norm": 1.7537293080966447,
      "language_loss": 0.84263122,
      "learning_rate": 1.3846941538672606e-07,
      "loss": 0.86446595,
      "num_input_tokens_seen": 159120025,
      "step": 7355,
      "time_per_iteration": 2.4901413917541504
    },
    {
      "auxiliary_loss_clip": 0.01112876,
      "auxiliary_loss_mlp": 0.01029878,
      "balance_loss_clip": 1.04501057,
      "balance_loss_mlp": 1.0227139,
      "epoch": 0.8845067035411531,
      "flos": 28183915388160.0,
      "grad_norm": 2.353654000184152,
      "language_loss": 0.81194133,
      "learning_rate": 1.3818475112181193e-07,
      "loss": 0.83336884,
      "num_input_tokens_seen": 159138820,
      "step": 7356,
      "time_per_iteration": 2.646470308303833
    },
    {
      "auxiliary_loss_clip": 0.01131635,
      "auxiliary_loss_mlp": 0.01021357,
      "balance_loss_clip": 1.04525256,
      "balance_loss_mlp": 1.01523304,
      "epoch": 0.8846269464317922,
      "flos": 12853219311360.0,
      "grad_norm": 1.9190170150527706,
      "language_loss": 0.79512346,
      "learning_rate": 1.3790036929279091e-07,
      "loss": 0.81665337,
      "num_input_tokens_seen": 159155975,
      "step": 7357,
      "time_per_iteration": 2.5092339515686035
    },
    {
      "auxiliary_loss_clip": 0.01153483,
      "auxiliary_loss_mlp": 0.0076108,
      "balance_loss_clip": 1.04823232,
      "balance_loss_mlp": 1.00034904,
      "epoch": 0.8847471893224313,
      "flos": 18624351628800.0,
      "grad_norm": 2.103822450582153,
      "language_loss": 0.58460033,
      "learning_rate": 1.3761626994280363e-07,
      "loss": 0.60374594,
      "num_input_tokens_seen": 159173445,
      "step": 7358,
      "time_per_iteration": 2.480623722076416
    },
    {
      "auxiliary_loss_clip": 0.01123239,
      "auxiliary_loss_mlp": 0.01023023,
      "balance_loss_clip": 1.04128671,
      "balance_loss_mlp": 1.01642513,
      "epoch": 0.8848674322130704,
      "flos": 35769433449600.0,
      "grad_norm": 1.6916460702151148,
      "language_loss": 0.73470736,
      "learning_rate": 1.3733245311494735e-07,
      "loss": 0.75617003,
      "num_input_tokens_seen": 159196100,
      "step": 7359,
      "time_per_iteration": 2.6984171867370605
    },
    {
      "auxiliary_loss_clip": 0.01148963,
      "auxiliary_loss_mlp": 0.01025351,
      "balance_loss_clip": 1.04602134,
      "balance_loss_mlp": 1.01820421,
      "epoch": 0.8849876751037095,
      "flos": 24243760897920.0,
      "grad_norm": 2.4818108571349256,
      "language_loss": 0.7062704,
      "learning_rate": 1.3704891885227676e-07,
      "loss": 0.72801352,
      "num_input_tokens_seen": 159216145,
      "step": 7360,
      "time_per_iteration": 2.5198798179626465
    },
    {
      "auxiliary_loss_clip": 0.01118148,
      "auxiliary_loss_mlp": 0.01027542,
      "balance_loss_clip": 1.03866065,
      "balance_loss_mlp": 1.02041876,
      "epoch": 0.8851079179943486,
      "flos": 21500580251520.0,
      "grad_norm": 2.0336068332147015,
      "language_loss": 0.77887738,
      "learning_rate": 1.367656671978037e-07,
      "loss": 0.80033433,
      "num_input_tokens_seen": 159233610,
      "step": 7361,
      "time_per_iteration": 2.5557496547698975
    },
    {
      "auxiliary_loss_clip": 0.01138313,
      "auxiliary_loss_mlp": 0.01025114,
      "balance_loss_clip": 1.04252517,
      "balance_loss_mlp": 1.01867104,
      "epoch": 0.8852281608849877,
      "flos": 15300711198720.0,
      "grad_norm": 2.496801702282319,
      "language_loss": 0.73762441,
      "learning_rate": 1.36482698194498e-07,
      "loss": 0.75925869,
      "num_input_tokens_seen": 159250155,
      "step": 7362,
      "time_per_iteration": 3.2355852127075195
    },
    {
      "auxiliary_loss_clip": 0.01136212,
      "auxiliary_loss_mlp": 0.01028742,
      "balance_loss_clip": 1.04363954,
      "balance_loss_mlp": 1.02126729,
      "epoch": 0.8853484037756267,
      "flos": 23295719283840.0,
      "grad_norm": 1.7319929563162109,
      "language_loss": 0.71714276,
      "learning_rate": 1.3620001188528506e-07,
      "loss": 0.7387923,
      "num_input_tokens_seen": 159270875,
      "step": 7363,
      "time_per_iteration": 2.5511672496795654
    },
    {
      "auxiliary_loss_clip": 0.01150313,
      "auxiliary_loss_mlp": 0.01028741,
      "balance_loss_clip": 1.04491985,
      "balance_loss_mlp": 1.0216217,
      "epoch": 0.8854686466662659,
      "flos": 25114773795840.0,
      "grad_norm": 6.542907712405155,
      "language_loss": 0.7352742,
      "learning_rate": 1.3591760831304865e-07,
      "loss": 0.7570647,
      "num_input_tokens_seen": 159288565,
      "step": 7364,
      "time_per_iteration": 3.3268024921417236
    },
    {
      "auxiliary_loss_clip": 0.011618,
      "auxiliary_loss_mlp": 0.0103097,
      "balance_loss_clip": 1.04669142,
      "balance_loss_mlp": 1.02354312,
      "epoch": 0.885588889556905,
      "flos": 21390873137280.0,
      "grad_norm": 1.673553517308425,
      "language_loss": 0.79280424,
      "learning_rate": 1.356354875206287e-07,
      "loss": 0.8147319,
      "num_input_tokens_seen": 159306400,
      "step": 7365,
      "time_per_iteration": 3.187958240509033
    },
    {
      "auxiliary_loss_clip": 0.011205,
      "auxiliary_loss_mlp": 0.0102693,
      "balance_loss_clip": 1.04521346,
      "balance_loss_mlp": 1.01975632,
      "epoch": 0.885709132447544,
      "flos": 26906752431360.0,
      "grad_norm": 1.8966104663288512,
      "language_loss": 0.69828367,
      "learning_rate": 1.3535364955082296e-07,
      "loss": 0.71975797,
      "num_input_tokens_seen": 159326250,
      "step": 7366,
      "time_per_iteration": 2.598649740219116
    },
    {
      "auxiliary_loss_clip": 0.01160573,
      "auxiliary_loss_mlp": 0.01023878,
      "balance_loss_clip": 1.04672301,
      "balance_loss_mlp": 1.01757455,
      "epoch": 0.8858293753381832,
      "flos": 26103394800000.0,
      "grad_norm": 1.8056448074686748,
      "language_loss": 0.64303648,
      "learning_rate": 1.3507209444638613e-07,
      "loss": 0.66488099,
      "num_input_tokens_seen": 159348250,
      "step": 7367,
      "time_per_iteration": 2.504650354385376
    },
    {
      "auxiliary_loss_clip": 0.0114933,
      "auxiliary_loss_mlp": 0.010276,
      "balance_loss_clip": 1.04592288,
      "balance_loss_mlp": 1.02076888,
      "epoch": 0.8859496182288222,
      "flos": 23292810282240.0,
      "grad_norm": 1.8402854358864091,
      "language_loss": 0.73796588,
      "learning_rate": 1.347908222500298e-07,
      "loss": 0.75973523,
      "num_input_tokens_seen": 159368325,
      "step": 7368,
      "time_per_iteration": 2.4949305057525635
    },
    {
      "auxiliary_loss_clip": 0.01107113,
      "auxiliary_loss_mlp": 0.01021209,
      "balance_loss_clip": 1.04102755,
      "balance_loss_mlp": 1.01442909,
      "epoch": 0.8860698611194613,
      "flos": 16872916469760.0,
      "grad_norm": 1.8947453828312526,
      "language_loss": 0.69706559,
      "learning_rate": 1.3450983300442276e-07,
      "loss": 0.71834886,
      "num_input_tokens_seen": 159387555,
      "step": 7369,
      "time_per_iteration": 2.5385656356811523
    },
    {
      "auxiliary_loss_clip": 0.01153072,
      "auxiliary_loss_mlp": 0.01027591,
      "balance_loss_clip": 1.04765272,
      "balance_loss_mlp": 1.02086115,
      "epoch": 0.8861901040101005,
      "flos": 24681404206080.0,
      "grad_norm": 1.8269480298829808,
      "language_loss": 0.73396206,
      "learning_rate": 1.3422912675219068e-07,
      "loss": 0.75576872,
      "num_input_tokens_seen": 159407310,
      "step": 7370,
      "time_per_iteration": 2.53542160987854
    },
    {
      "auxiliary_loss_clip": 0.01159309,
      "auxiliary_loss_mlp": 0.0102542,
      "balance_loss_clip": 1.04656386,
      "balance_loss_mlp": 1.01877093,
      "epoch": 0.8863103469007395,
      "flos": 24423026699520.0,
      "grad_norm": 1.5070411842244116,
      "language_loss": 0.78914177,
      "learning_rate": 1.339487035359166e-07,
      "loss": 0.81098902,
      "num_input_tokens_seen": 159427680,
      "step": 7371,
      "time_per_iteration": 2.501157522201538
    },
    {
      "auxiliary_loss_clip": 0.01138981,
      "auxiliary_loss_mlp": 0.00759813,
      "balance_loss_clip": 1.04781425,
      "balance_loss_mlp": 1.00035965,
      "epoch": 0.8864305897913786,
      "flos": 22053964158720.0,
      "grad_norm": 1.5255304552770708,
      "language_loss": 0.84975827,
      "learning_rate": 1.336685633981409e-07,
      "loss": 0.86874622,
      "num_input_tokens_seen": 159448765,
      "step": 7372,
      "time_per_iteration": 2.552380323410034
    },
    {
      "auxiliary_loss_clip": 0.01148912,
      "auxiliary_loss_mlp": 0.01025314,
      "balance_loss_clip": 1.0439707,
      "balance_loss_mlp": 1.01869464,
      "epoch": 0.8865508326820177,
      "flos": 19099449843840.0,
      "grad_norm": 1.7667924984640728,
      "language_loss": 0.74797893,
      "learning_rate": 1.333887063813597e-07,
      "loss": 0.76972121,
      "num_input_tokens_seen": 159466870,
      "step": 7373,
      "time_per_iteration": 2.492511749267578
    },
    {
      "auxiliary_loss_clip": 0.01136233,
      "auxiliary_loss_mlp": 0.01025417,
      "balance_loss_clip": 1.04292011,
      "balance_loss_mlp": 1.01895869,
      "epoch": 0.8866710755726568,
      "flos": 15414189240960.0,
      "grad_norm": 1.8519643705507336,
      "language_loss": 0.66675442,
      "learning_rate": 1.331091325280278e-07,
      "loss": 0.68837094,
      "num_input_tokens_seen": 159485840,
      "step": 7374,
      "time_per_iteration": 2.510791301727295
    },
    {
      "auxiliary_loss_clip": 0.01100482,
      "auxiliary_loss_mlp": 0.01026336,
      "balance_loss_clip": 1.03991127,
      "balance_loss_mlp": 1.01876664,
      "epoch": 0.8867913184632958,
      "flos": 20083689388800.0,
      "grad_norm": 3.947824893526396,
      "language_loss": 0.78547907,
      "learning_rate": 1.3282984188055625e-07,
      "loss": 0.80674732,
      "num_input_tokens_seen": 159505630,
      "step": 7375,
      "time_per_iteration": 2.6000115871429443
    },
    {
      "auxiliary_loss_clip": 0.0116162,
      "auxiliary_loss_mlp": 0.01023174,
      "balance_loss_clip": 1.04643416,
      "balance_loss_mlp": 1.01603353,
      "epoch": 0.8869115613539349,
      "flos": 23365852588800.0,
      "grad_norm": 2.056981995635614,
      "language_loss": 0.79639435,
      "learning_rate": 1.3255083448131288e-07,
      "loss": 0.81824231,
      "num_input_tokens_seen": 159524675,
      "step": 7376,
      "time_per_iteration": 2.468514919281006
    },
    {
      "auxiliary_loss_clip": 0.01153826,
      "auxiliary_loss_mlp": 0.01023867,
      "balance_loss_clip": 1.04635096,
      "balance_loss_mlp": 1.0168097,
      "epoch": 0.8870318042445741,
      "flos": 21286840371840.0,
      "grad_norm": 2.051706950523332,
      "language_loss": 0.79144704,
      "learning_rate": 1.3227211037262365e-07,
      "loss": 0.81322396,
      "num_input_tokens_seen": 159541915,
      "step": 7377,
      "time_per_iteration": 2.4925196170806885
    },
    {
      "auxiliary_loss_clip": 0.01106324,
      "auxiliary_loss_mlp": 0.01026293,
      "balance_loss_clip": 1.0411036,
      "balance_loss_mlp": 1.0193013,
      "epoch": 0.8871520471352131,
      "flos": 20010862563840.0,
      "grad_norm": 2.2896742697620596,
      "language_loss": 0.85037845,
      "learning_rate": 1.319936695967696e-07,
      "loss": 0.87170464,
      "num_input_tokens_seen": 159559740,
      "step": 7378,
      "time_per_iteration": 3.3878891468048096
    },
    {
      "auxiliary_loss_clip": 0.01168705,
      "auxiliary_loss_mlp": 0.01025512,
      "balance_loss_clip": 1.04749191,
      "balance_loss_mlp": 1.01831126,
      "epoch": 0.8872722900258522,
      "flos": 22601422321920.0,
      "grad_norm": 2.28804989655411,
      "language_loss": 0.82376409,
      "learning_rate": 1.3171551219599097e-07,
      "loss": 0.84570622,
      "num_input_tokens_seen": 159578265,
      "step": 7379,
      "time_per_iteration": 2.465461254119873
    },
    {
      "auxiliary_loss_clip": 0.01163571,
      "auxiliary_loss_mlp": 0.01029357,
      "balance_loss_clip": 1.04819334,
      "balance_loss_mlp": 1.02224648,
      "epoch": 0.8873925329164913,
      "flos": 22163276223360.0,
      "grad_norm": 2.1278482347027547,
      "language_loss": 0.78018987,
      "learning_rate": 1.3143763821248377e-07,
      "loss": 0.80211914,
      "num_input_tokens_seen": 159595350,
      "step": 7380,
      "time_per_iteration": 2.46467661857605
    },
    {
      "auxiliary_loss_clip": 0.01161501,
      "auxiliary_loss_mlp": 0.01026119,
      "balance_loss_clip": 1.04731095,
      "balance_loss_mlp": 1.01956558,
      "epoch": 0.8875127758071304,
      "flos": 19208223204480.0,
      "grad_norm": 1.757366773108,
      "language_loss": 0.72181118,
      "learning_rate": 1.3116004768840118e-07,
      "loss": 0.74368739,
      "num_input_tokens_seen": 159613725,
      "step": 7381,
      "time_per_iteration": 2.4442265033721924
    },
    {
      "auxiliary_loss_clip": 0.01164293,
      "auxiliary_loss_mlp": 0.01030167,
      "balance_loss_clip": 1.04834008,
      "balance_loss_mlp": 1.02225423,
      "epoch": 0.8876330186977694,
      "flos": 18110900666880.0,
      "grad_norm": 1.5771448108851112,
      "language_loss": 0.74344814,
      "learning_rate": 1.3088274066585348e-07,
      "loss": 0.76539278,
      "num_input_tokens_seen": 159631335,
      "step": 7382,
      "time_per_iteration": 2.449061155319214
    },
    {
      "auxiliary_loss_clip": 0.01125708,
      "auxiliary_loss_mlp": 0.01024447,
      "balance_loss_clip": 1.04239285,
      "balance_loss_mlp": 1.01778913,
      "epoch": 0.8877532615884086,
      "flos": 22009434272640.0,
      "grad_norm": 2.0557404573500357,
      "language_loss": 0.90528738,
      "learning_rate": 1.3060571718690749e-07,
      "loss": 0.92678893,
      "num_input_tokens_seen": 159648830,
      "step": 7383,
      "time_per_iteration": 2.580018997192383
    },
    {
      "auxiliary_loss_clip": 0.0102685,
      "auxiliary_loss_mlp": 0.00751233,
      "balance_loss_clip": 1.0080682,
      "balance_loss_mlp": 1.00024748,
      "epoch": 0.8878735044790477,
      "flos": 72136924346880.0,
      "grad_norm": 0.746934924187684,
      "language_loss": 0.56938475,
      "learning_rate": 1.3032897729358805e-07,
      "loss": 0.58716553,
      "num_input_tokens_seen": 159709785,
      "step": 7384,
      "time_per_iteration": 3.155881643295288
    },
    {
      "auxiliary_loss_clip": 0.0108379,
      "auxiliary_loss_mlp": 0.00761577,
      "balance_loss_clip": 1.03728628,
      "balance_loss_mlp": 1.00031304,
      "epoch": 0.8879937473696867,
      "flos": 27526355061120.0,
      "grad_norm": 2.2727654404173516,
      "language_loss": 0.79572964,
      "learning_rate": 1.3005252102787645e-07,
      "loss": 0.81418329,
      "num_input_tokens_seen": 159728725,
      "step": 7385,
      "time_per_iteration": 2.6646227836608887
    },
    {
      "auxiliary_loss_clip": 0.01153129,
      "auxiliary_loss_mlp": 0.01027769,
      "balance_loss_clip": 1.04690695,
      "balance_loss_mlp": 1.02093256,
      "epoch": 0.8881139902603259,
      "flos": 22234091886720.0,
      "grad_norm": 1.5645379472624619,
      "language_loss": 0.73570222,
      "learning_rate": 1.297763484317105e-07,
      "loss": 0.7575112,
      "num_input_tokens_seen": 159747020,
      "step": 7386,
      "time_per_iteration": 2.49581241607666
    },
    {
      "auxiliary_loss_clip": 0.01099854,
      "auxiliary_loss_mlp": 0.00761581,
      "balance_loss_clip": 1.03918254,
      "balance_loss_mlp": 1.00029051,
      "epoch": 0.888234233150965,
      "flos": 20299548170880.0,
      "grad_norm": 2.0828391090877973,
      "language_loss": 0.70298672,
      "learning_rate": 1.2950045954698551e-07,
      "loss": 0.72160101,
      "num_input_tokens_seen": 159764855,
      "step": 7387,
      "time_per_iteration": 2.578559160232544
    },
    {
      "auxiliary_loss_clip": 0.01113617,
      "auxiliary_loss_mlp": 0.01024403,
      "balance_loss_clip": 1.04286075,
      "balance_loss_mlp": 1.01792097,
      "epoch": 0.888354476041604,
      "flos": 18147996437760.0,
      "grad_norm": 1.5841152538447005,
      "language_loss": 0.75465393,
      "learning_rate": 1.2922485441555343e-07,
      "loss": 0.77603412,
      "num_input_tokens_seen": 159783935,
      "step": 7388,
      "time_per_iteration": 3.3947603702545166
    },
    {
      "auxiliary_loss_clip": 0.0116153,
      "auxiliary_loss_mlp": 0.01024315,
      "balance_loss_clip": 1.04593158,
      "balance_loss_mlp": 1.01813388,
      "epoch": 0.8884747189322432,
      "flos": 22014282608640.0,
      "grad_norm": 1.722132608100089,
      "language_loss": 0.81813335,
      "learning_rate": 1.2894953307922363e-07,
      "loss": 0.83999181,
      "num_input_tokens_seen": 159802895,
      "step": 7389,
      "time_per_iteration": 2.4670357704162598
    },
    {
      "auxiliary_loss_clip": 0.01116183,
      "auxiliary_loss_mlp": 0.01026269,
      "balance_loss_clip": 1.04351723,
      "balance_loss_mlp": 1.01891077,
      "epoch": 0.8885949618228822,
      "flos": 19786779567360.0,
      "grad_norm": 1.904915795204902,
      "language_loss": 0.84140193,
      "learning_rate": 1.2867449557976208e-07,
      "loss": 0.86282647,
      "num_input_tokens_seen": 159820995,
      "step": 7390,
      "time_per_iteration": 3.308199882507324
    },
    {
      "auxiliary_loss_clip": 0.01148953,
      "auxiliary_loss_mlp": 0.01028342,
      "balance_loss_clip": 1.04743493,
      "balance_loss_mlp": 1.02181578,
      "epoch": 0.8887152047135213,
      "flos": 20047599198720.0,
      "grad_norm": 2.475454977518529,
      "language_loss": 0.75669563,
      "learning_rate": 1.283997419588916e-07,
      "loss": 0.77846861,
      "num_input_tokens_seen": 159840465,
      "step": 7391,
      "time_per_iteration": 3.155033588409424
    },
    {
      "auxiliary_loss_clip": 0.01152029,
      "auxiliary_loss_mlp": 0.01025831,
      "balance_loss_clip": 1.04579449,
      "balance_loss_mlp": 1.01880932,
      "epoch": 0.8888354476041604,
      "flos": 18588117784320.0,
      "grad_norm": 2.1358218219947385,
      "language_loss": 0.61874312,
      "learning_rate": 1.2812527225829216e-07,
      "loss": 0.64052176,
      "num_input_tokens_seen": 159858690,
      "step": 7392,
      "time_per_iteration": 2.4807190895080566
    },
    {
      "auxiliary_loss_clip": 0.01157475,
      "auxiliary_loss_mlp": 0.01027721,
      "balance_loss_clip": 1.04877675,
      "balance_loss_mlp": 1.02017844,
      "epoch": 0.8889556904947995,
      "flos": 21689794120320.0,
      "grad_norm": 1.8629029828097068,
      "language_loss": 0.76669598,
      "learning_rate": 1.2785108651960052e-07,
      "loss": 0.78854793,
      "num_input_tokens_seen": 159880325,
      "step": 7393,
      "time_per_iteration": 2.527873992919922
    },
    {
      "auxiliary_loss_clip": 0.01150541,
      "auxiliary_loss_mlp": 0.01025819,
      "balance_loss_clip": 1.04360723,
      "balance_loss_mlp": 1.01839566,
      "epoch": 0.8890759333854386,
      "flos": 27381204201600.0,
      "grad_norm": 1.9930803251801332,
      "language_loss": 0.80737621,
      "learning_rate": 1.2757718478441094e-07,
      "loss": 0.82913983,
      "num_input_tokens_seen": 159901070,
      "step": 7394,
      "time_per_iteration": 2.5457708835601807
    },
    {
      "auxiliary_loss_clip": 0.01132686,
      "auxiliary_loss_mlp": 0.01025926,
      "balance_loss_clip": 1.04221559,
      "balance_loss_mlp": 1.0190264,
      "epoch": 0.8891961762760777,
      "flos": 24498834353280.0,
      "grad_norm": 1.9109857340534795,
      "language_loss": 0.77172261,
      "learning_rate": 1.2730356709427302e-07,
      "loss": 0.79330873,
      "num_input_tokens_seen": 159919750,
      "step": 7395,
      "time_per_iteration": 2.554243803024292
    },
    {
      "auxiliary_loss_clip": 0.01146875,
      "auxiliary_loss_mlp": 0.01024404,
      "balance_loss_clip": 1.04758501,
      "balance_loss_mlp": 1.0178709,
      "epoch": 0.8893164191667168,
      "flos": 41499770895360.0,
      "grad_norm": 1.6241878572554849,
      "language_loss": 0.59897184,
      "learning_rate": 1.2703023349069542e-07,
      "loss": 0.62068462,
      "num_input_tokens_seen": 159944600,
      "step": 7396,
      "time_per_iteration": 2.6725103855133057
    },
    {
      "auxiliary_loss_clip": 0.01144284,
      "auxiliary_loss_mlp": 0.01024081,
      "balance_loss_clip": 1.04452491,
      "balance_loss_mlp": 1.01715159,
      "epoch": 0.8894366620573558,
      "flos": 33583623120000.0,
      "grad_norm": 2.0022390176584115,
      "language_loss": 0.61713123,
      "learning_rate": 1.2675718401514223e-07,
      "loss": 0.63881493,
      "num_input_tokens_seen": 159968780,
      "step": 7397,
      "time_per_iteration": 2.6006500720977783
    },
    {
      "auxiliary_loss_clip": 0.01133954,
      "auxiliary_loss_mlp": 0.01025974,
      "balance_loss_clip": 1.04388475,
      "balance_loss_mlp": 1.01896715,
      "epoch": 0.889556904947995,
      "flos": 16909832672640.0,
      "grad_norm": 3.7461030506979465,
      "language_loss": 0.74372351,
      "learning_rate": 1.264844187090346e-07,
      "loss": 0.7653228,
      "num_input_tokens_seen": 159985905,
      "step": 7398,
      "time_per_iteration": 2.512767791748047
    },
    {
      "auxiliary_loss_clip": 0.01129699,
      "auxiliary_loss_mlp": 0.01022794,
      "balance_loss_clip": 1.04250598,
      "balance_loss_mlp": 1.01593304,
      "epoch": 0.889677147838634,
      "flos": 26030855283840.0,
      "grad_norm": 1.7541093110588333,
      "language_loss": 0.75031447,
      "learning_rate": 1.262119376137516e-07,
      "loss": 0.77183938,
      "num_input_tokens_seen": 160006965,
      "step": 7399,
      "time_per_iteration": 2.558415412902832
    },
    {
      "auxiliary_loss_clip": 0.011371,
      "auxiliary_loss_mlp": 0.01028719,
      "balance_loss_clip": 1.0417968,
      "balance_loss_mlp": 1.02180469,
      "epoch": 0.8897973907292731,
      "flos": 26468283110400.0,
      "grad_norm": 1.4715787350143474,
      "language_loss": 0.85056758,
      "learning_rate": 1.2593974077062707e-07,
      "loss": 0.87222576,
      "num_input_tokens_seen": 160028585,
      "step": 7400,
      "time_per_iteration": 2.538681983947754
    },
    {
      "auxiliary_loss_clip": 0.0111146,
      "auxiliary_loss_mlp": 0.01030206,
      "balance_loss_clip": 1.04126358,
      "balance_loss_mlp": 1.02356339,
      "epoch": 0.8899176336199123,
      "flos": 26249694894720.0,
      "grad_norm": 1.9622732756288723,
      "language_loss": 0.63728619,
      "learning_rate": 1.2566782822095423e-07,
      "loss": 0.65870285,
      "num_input_tokens_seen": 160048840,
      "step": 7401,
      "time_per_iteration": 2.613985538482666
    },
    {
      "auxiliary_loss_clip": 0.01126955,
      "auxiliary_loss_mlp": 0.01028797,
      "balance_loss_clip": 1.04613304,
      "balance_loss_mlp": 1.02249694,
      "epoch": 0.8900378765105513,
      "flos": 20811742156800.0,
      "grad_norm": 2.1168286665464535,
      "language_loss": 0.71203077,
      "learning_rate": 1.2539620000598162e-07,
      "loss": 0.73358834,
      "num_input_tokens_seen": 160068175,
      "step": 7402,
      "time_per_iteration": 2.576333522796631
    },
    {
      "auxiliary_loss_clip": 0.01160417,
      "auxiliary_loss_mlp": 0.0102638,
      "balance_loss_clip": 1.04513204,
      "balance_loss_mlp": 1.01913214,
      "epoch": 0.8901581194011904,
      "flos": 16472333018880.0,
      "grad_norm": 1.8496561117810337,
      "language_loss": 0.79724431,
      "learning_rate": 1.2512485616691492e-07,
      "loss": 0.8191123,
      "num_input_tokens_seen": 160085230,
      "step": 7403,
      "time_per_iteration": 3.1410610675811768
    },
    {
      "auxiliary_loss_clip": 0.01123477,
      "auxiliary_loss_mlp": 0.01030844,
      "balance_loss_clip": 1.04315186,
      "balance_loss_mlp": 1.02355146,
      "epoch": 0.8902783622918296,
      "flos": 35155253773440.0,
      "grad_norm": 1.3794865600093456,
      "language_loss": 0.80994177,
      "learning_rate": 1.2485379674491681e-07,
      "loss": 0.83148497,
      "num_input_tokens_seen": 160111425,
      "step": 7404,
      "time_per_iteration": 2.705625534057617
    },
    {
      "auxiliary_loss_clip": 0.01136268,
      "auxiliary_loss_mlp": 0.01025623,
      "balance_loss_clip": 1.04631126,
      "balance_loss_mlp": 1.01862001,
      "epoch": 0.8903986051824686,
      "flos": 17201068145280.0,
      "grad_norm": 2.128251317112393,
      "language_loss": 0.7882039,
      "learning_rate": 1.2458302178110657e-07,
      "loss": 0.8098228,
      "num_input_tokens_seen": 160129790,
      "step": 7405,
      "time_per_iteration": 2.4934699535369873
    },
    {
      "auxiliary_loss_clip": 0.0111287,
      "auxiliary_loss_mlp": 0.01024317,
      "balance_loss_clip": 1.04147315,
      "balance_loss_mlp": 1.01801646,
      "epoch": 0.8905188480731077,
      "flos": 25483863997440.0,
      "grad_norm": 1.8294227201459328,
      "language_loss": 0.82171118,
      "learning_rate": 1.2431253131656118e-07,
      "loss": 0.84308302,
      "num_input_tokens_seen": 160149265,
      "step": 7406,
      "time_per_iteration": 2.585108995437622
    },
    {
      "auxiliary_loss_clip": 0.01129008,
      "auxiliary_loss_mlp": 0.01028282,
      "balance_loss_clip": 1.04535151,
      "balance_loss_mlp": 1.02130497,
      "epoch": 0.8906390909637467,
      "flos": 23365888502400.0,
      "grad_norm": 1.7925640152392257,
      "language_loss": 0.76818728,
      "learning_rate": 1.240423253923133e-07,
      "loss": 0.78976017,
      "num_input_tokens_seen": 160168870,
      "step": 7407,
      "time_per_iteration": 2.536597728729248
    },
    {
      "auxiliary_loss_clip": 0.01150407,
      "auxiliary_loss_mlp": 0.01025501,
      "balance_loss_clip": 1.04546547,
      "balance_loss_mlp": 1.01847994,
      "epoch": 0.8907593338543859,
      "flos": 21068790860160.0,
      "grad_norm": 2.20307130548792,
      "language_loss": 0.69325078,
      "learning_rate": 1.237724040493533e-07,
      "loss": 0.71500993,
      "num_input_tokens_seen": 160187495,
      "step": 7408,
      "time_per_iteration": 2.4833898544311523
    },
    {
      "auxiliary_loss_clip": 0.01167063,
      "auxiliary_loss_mlp": 0.01028702,
      "balance_loss_clip": 1.04895508,
      "balance_loss_mlp": 1.02105141,
      "epoch": 0.8908795767450249,
      "flos": 21869562712320.0,
      "grad_norm": 2.865595984824438,
      "language_loss": 0.72897494,
      "learning_rate": 1.2350276732862773e-07,
      "loss": 0.75093257,
      "num_input_tokens_seen": 160208520,
      "step": 7409,
      "time_per_iteration": 2.477720260620117
    },
    {
      "auxiliary_loss_clip": 0.01046172,
      "auxiliary_loss_mlp": 0.01003259,
      "balance_loss_clip": 1.00811172,
      "balance_loss_mlp": 1.00220954,
      "epoch": 0.890999819635664,
      "flos": 66307869348480.0,
      "grad_norm": 0.8373480587881046,
      "language_loss": 0.5672555,
      "learning_rate": 1.2323341527103993e-07,
      "loss": 0.58774984,
      "num_input_tokens_seen": 160263720,
      "step": 7410,
      "time_per_iteration": 3.007127285003662
    },
    {
      "auxiliary_loss_clip": 0.01163099,
      "auxiliary_loss_mlp": 0.01024288,
      "balance_loss_clip": 1.04729903,
      "balance_loss_mlp": 1.01757705,
      "epoch": 0.8911200625263032,
      "flos": 26869908055680.0,
      "grad_norm": 1.9842470986994716,
      "language_loss": 0.8505584,
      "learning_rate": 1.2296434791745135e-07,
      "loss": 0.87243229,
      "num_input_tokens_seen": 160282170,
      "step": 7411,
      "time_per_iteration": 2.496389865875244
    },
    {
      "auxiliary_loss_clip": 0.01150365,
      "auxiliary_loss_mlp": 0.01028823,
      "balance_loss_clip": 1.04617453,
      "balance_loss_mlp": 1.02178645,
      "epoch": 0.8912403054169422,
      "flos": 20885825957760.0,
      "grad_norm": 1.6041375959149036,
      "language_loss": 0.76385522,
      "learning_rate": 1.2269556530867875e-07,
      "loss": 0.78564703,
      "num_input_tokens_seen": 160300725,
      "step": 7412,
      "time_per_iteration": 2.478358030319214
    },
    {
      "auxiliary_loss_clip": 0.01168715,
      "auxiliary_loss_mlp": 0.01031622,
      "balance_loss_clip": 1.0491817,
      "balance_loss_mlp": 1.02406073,
      "epoch": 0.8913605483075813,
      "flos": 27016567286400.0,
      "grad_norm": 2.0281353569641163,
      "language_loss": 0.81472683,
      "learning_rate": 1.2242706748549614e-07,
      "loss": 0.83673024,
      "num_input_tokens_seen": 160318720,
      "step": 7413,
      "time_per_iteration": 2.4962806701660156
    },
    {
      "auxiliary_loss_clip": 0.01134037,
      "auxiliary_loss_mlp": 0.01023889,
      "balance_loss_clip": 1.04022205,
      "balance_loss_mlp": 1.01700449,
      "epoch": 0.8914807911982204,
      "flos": 23621500661760.0,
      "grad_norm": 2.6399933719858146,
      "language_loss": 0.82265896,
      "learning_rate": 1.2215885448863473e-07,
      "loss": 0.84423822,
      "num_input_tokens_seen": 160339595,
      "step": 7414,
      "time_per_iteration": 3.3771467208862305
    },
    {
      "auxiliary_loss_clip": 0.01134231,
      "auxiliary_loss_mlp": 0.01026934,
      "balance_loss_clip": 1.04533386,
      "balance_loss_mlp": 1.02024031,
      "epoch": 0.8916010340888595,
      "flos": 24462277286400.0,
      "grad_norm": 1.6638265038590265,
      "language_loss": 0.79944813,
      "learning_rate": 1.2189092635878152e-07,
      "loss": 0.8210597,
      "num_input_tokens_seen": 160361045,
      "step": 7415,
      "time_per_iteration": 3.3647654056549072
    },
    {
      "auxiliary_loss_clip": 0.01108925,
      "auxiliary_loss_mlp": 0.01024053,
      "balance_loss_clip": 1.04085612,
      "balance_loss_mlp": 1.01708865,
      "epoch": 0.8917212769794985,
      "flos": 21215773313280.0,
      "grad_norm": 1.7055034256665786,
      "language_loss": 0.77369088,
      "learning_rate": 1.216232831365822e-07,
      "loss": 0.7950207,
      "num_input_tokens_seen": 160379990,
      "step": 7416,
      "time_per_iteration": 2.5688788890838623
    },
    {
      "auxiliary_loss_clip": 0.0114272,
      "auxiliary_loss_mlp": 0.01032085,
      "balance_loss_clip": 1.04704022,
      "balance_loss_mlp": 1.02515602,
      "epoch": 0.8918415198701377,
      "flos": 25513992529920.0,
      "grad_norm": 2.008331427591492,
      "language_loss": 0.80988884,
      "learning_rate": 1.2135592486263678e-07,
      "loss": 0.83163685,
      "num_input_tokens_seen": 160399240,
      "step": 7417,
      "time_per_iteration": 2.5471713542938232
    },
    {
      "auxiliary_loss_clip": 0.01134281,
      "auxiliary_loss_mlp": 0.01025518,
      "balance_loss_clip": 1.04373598,
      "balance_loss_mlp": 1.01860356,
      "epoch": 0.8919617627607768,
      "flos": 37853006693760.0,
      "grad_norm": 1.5306463921455495,
      "language_loss": 0.6103965,
      "learning_rate": 1.2108885157750415e-07,
      "loss": 0.63199449,
      "num_input_tokens_seen": 160421600,
      "step": 7418,
      "time_per_iteration": 3.342547655105591
    },
    {
      "auxiliary_loss_clip": 0.01118443,
      "auxiliary_loss_mlp": 0.00760693,
      "balance_loss_clip": 1.04547417,
      "balance_loss_mlp": 1.00031996,
      "epoch": 0.8920820056514158,
      "flos": 26213676531840.0,
      "grad_norm": 1.6823357497303655,
      "language_loss": 0.80433202,
      "learning_rate": 1.2082206332169897e-07,
      "loss": 0.8231234,
      "num_input_tokens_seen": 160441695,
      "step": 7419,
      "time_per_iteration": 2.6026644706726074
    },
    {
      "auxiliary_loss_clip": 0.01132414,
      "auxiliary_loss_mlp": 0.01024381,
      "balance_loss_clip": 1.04629779,
      "balance_loss_mlp": 1.01738334,
      "epoch": 0.892202248542055,
      "flos": 17383135207680.0,
      "grad_norm": 2.528356360430867,
      "language_loss": 0.73176026,
      "learning_rate": 1.2055556013569225e-07,
      "loss": 0.7533282,
      "num_input_tokens_seen": 160457205,
      "step": 7420,
      "time_per_iteration": 2.526818037033081
    },
    {
      "auxiliary_loss_clip": 0.01135826,
      "auxiliary_loss_mlp": 0.0102825,
      "balance_loss_clip": 1.04561555,
      "balance_loss_mlp": 1.02167892,
      "epoch": 0.892322491432694,
      "flos": 21324223451520.0,
      "grad_norm": 1.5949630715450362,
      "language_loss": 0.82172787,
      "learning_rate": 1.2028934205991315e-07,
      "loss": 0.84336865,
      "num_input_tokens_seen": 160476525,
      "step": 7421,
      "time_per_iteration": 2.523819923400879
    },
    {
      "auxiliary_loss_clip": 0.01147816,
      "auxiliary_loss_mlp": 0.01026366,
      "balance_loss_clip": 1.04353619,
      "balance_loss_mlp": 1.0192436,
      "epoch": 0.8924427343233331,
      "flos": 24029374573440.0,
      "grad_norm": 1.3436758099633799,
      "language_loss": 0.76613557,
      "learning_rate": 1.2002340913474607e-07,
      "loss": 0.78787744,
      "num_input_tokens_seen": 160500160,
      "step": 7422,
      "time_per_iteration": 2.5745363235473633
    },
    {
      "auxiliary_loss_clip": 0.01166622,
      "auxiliary_loss_mlp": 0.01025032,
      "balance_loss_clip": 1.04903519,
      "balance_loss_mlp": 1.01762319,
      "epoch": 0.8925629772139723,
      "flos": 30008069631360.0,
      "grad_norm": 4.816366655085421,
      "language_loss": 0.74097252,
      "learning_rate": 1.1975776140053317e-07,
      "loss": 0.76288909,
      "num_input_tokens_seen": 160520130,
      "step": 7423,
      "time_per_iteration": 2.5222017765045166
    },
    {
      "auxiliary_loss_clip": 0.01109818,
      "auxiliary_loss_mlp": 0.01027586,
      "balance_loss_clip": 1.04213154,
      "balance_loss_mlp": 1.02021313,
      "epoch": 0.8926832201046113,
      "flos": 22601709630720.0,
      "grad_norm": 2.0255575629179003,
      "language_loss": 0.73527914,
      "learning_rate": 1.194923988975729e-07,
      "loss": 0.75665319,
      "num_input_tokens_seen": 160539730,
      "step": 7424,
      "time_per_iteration": 2.597662925720215
    },
    {
      "auxiliary_loss_clip": 0.0111811,
      "auxiliary_loss_mlp": 0.0102545,
      "balance_loss_clip": 1.04261017,
      "balance_loss_mlp": 1.01762354,
      "epoch": 0.8928034629952504,
      "flos": 13297722117120.0,
      "grad_norm": 2.1316676612759204,
      "language_loss": 0.73224986,
      "learning_rate": 1.192273216661206e-07,
      "loss": 0.75368547,
      "num_input_tokens_seen": 160557820,
      "step": 7425,
      "time_per_iteration": 2.512511730194092
    },
    {
      "auxiliary_loss_clip": 0.01003574,
      "auxiliary_loss_mlp": 0.0100238,
      "balance_loss_clip": 1.00683701,
      "balance_loss_mlp": 1.00131309,
      "epoch": 0.8929237058858895,
      "flos": 54854556744960.0,
      "grad_norm": 0.7688657710720881,
      "language_loss": 0.57465923,
      "learning_rate": 1.189625297463881e-07,
      "loss": 0.59471875,
      "num_input_tokens_seen": 160619510,
      "step": 7426,
      "time_per_iteration": 3.160731315612793
    },
    {
      "auxiliary_loss_clip": 0.01084909,
      "auxiliary_loss_mlp": 0.01020986,
      "balance_loss_clip": 1.03409195,
      "balance_loss_mlp": 1.01408708,
      "epoch": 0.8930439487765286,
      "flos": 28883850785280.0,
      "grad_norm": 1.5367520762357738,
      "language_loss": 0.79482758,
      "learning_rate": 1.1869802317854394e-07,
      "loss": 0.81588656,
      "num_input_tokens_seen": 160643295,
      "step": 7427,
      "time_per_iteration": 2.7322096824645996
    },
    {
      "auxiliary_loss_clip": 0.01113607,
      "auxiliary_loss_mlp": 0.01023409,
      "balance_loss_clip": 1.04364181,
      "balance_loss_mlp": 1.01615262,
      "epoch": 0.8931641916671677,
      "flos": 22419283432320.0,
      "grad_norm": 1.7343995445743285,
      "language_loss": 0.72074425,
      "learning_rate": 1.1843380200271425e-07,
      "loss": 0.74211448,
      "num_input_tokens_seen": 160662495,
      "step": 7428,
      "time_per_iteration": 2.5945115089416504
    },
    {
      "auxiliary_loss_clip": 0.01113214,
      "auxiliary_loss_mlp": 0.01026338,
      "balance_loss_clip": 1.04279768,
      "balance_loss_mlp": 1.01928401,
      "epoch": 0.8932844345578068,
      "flos": 25843149786240.0,
      "grad_norm": 1.6787148635646005,
      "language_loss": 0.80171049,
      "learning_rate": 1.181698662589805e-07,
      "loss": 0.82310599,
      "num_input_tokens_seen": 160682080,
      "step": 7429,
      "time_per_iteration": 3.3840930461883545
    },
    {
      "auxiliary_loss_clip": 0.01147868,
      "auxiliary_loss_mlp": 0.01022089,
      "balance_loss_clip": 1.04457545,
      "balance_loss_mlp": 1.01528823,
      "epoch": 0.8934046774484459,
      "flos": 22925803069440.0,
      "grad_norm": 1.7884042904628068,
      "language_loss": 0.76093256,
      "learning_rate": 1.1790621598738249e-07,
      "loss": 0.78263211,
      "num_input_tokens_seen": 160700395,
      "step": 7430,
      "time_per_iteration": 2.497098684310913
    },
    {
      "auxiliary_loss_clip": 0.0116352,
      "auxiliary_loss_mlp": 0.01019771,
      "balance_loss_clip": 1.05053544,
      "balance_loss_mlp": 1.01339912,
      "epoch": 0.8935249203390849,
      "flos": 24462097718400.0,
      "grad_norm": 1.825975352902499,
      "language_loss": 0.74477571,
      "learning_rate": 1.1764285122791461e-07,
      "loss": 0.76660866,
      "num_input_tokens_seen": 160721115,
      "step": 7431,
      "time_per_iteration": 2.509669065475464
    },
    {
      "auxiliary_loss_clip": 0.01148967,
      "auxiliary_loss_mlp": 0.01024102,
      "balance_loss_clip": 1.04490519,
      "balance_loss_mlp": 1.01721764,
      "epoch": 0.8936451632297241,
      "flos": 15742735966080.0,
      "grad_norm": 1.7320821003056215,
      "language_loss": 0.76656246,
      "learning_rate": 1.173797720205294e-07,
      "loss": 0.78829312,
      "num_input_tokens_seen": 160739150,
      "step": 7432,
      "time_per_iteration": 2.466038465499878
    },
    {
      "auxiliary_loss_clip": 0.01149589,
      "auxiliary_loss_mlp": 0.01024683,
      "balance_loss_clip": 1.04573584,
      "balance_loss_mlp": 1.01737273,
      "epoch": 0.8937654061203631,
      "flos": 35115500396160.0,
      "grad_norm": 2.102042160492293,
      "language_loss": 0.7177195,
      "learning_rate": 1.1711697840513602e-07,
      "loss": 0.7394622,
      "num_input_tokens_seen": 160758585,
      "step": 7433,
      "time_per_iteration": 2.6021134853363037
    },
    {
      "auxiliary_loss_clip": 0.01141241,
      "auxiliary_loss_mlp": 0.01029486,
      "balance_loss_clip": 1.04322314,
      "balance_loss_mlp": 1.02330542,
      "epoch": 0.8938856490110022,
      "flos": 16107444708480.0,
      "grad_norm": 1.9450508951035508,
      "language_loss": 0.70586014,
      "learning_rate": 1.1685447042160012e-07,
      "loss": 0.72756743,
      "num_input_tokens_seen": 160776620,
      "step": 7434,
      "time_per_iteration": 2.4583957195281982
    },
    {
      "auxiliary_loss_clip": 0.01165408,
      "auxiliary_loss_mlp": 0.0102299,
      "balance_loss_clip": 1.04764819,
      "balance_loss_mlp": 1.01574779,
      "epoch": 0.8940058919016414,
      "flos": 20704189858560.0,
      "grad_norm": 1.588797043846843,
      "language_loss": 0.71360916,
      "learning_rate": 1.1659224810974367e-07,
      "loss": 0.73549312,
      "num_input_tokens_seen": 160796580,
      "step": 7435,
      "time_per_iteration": 2.4584853649139404
    },
    {
      "auxiliary_loss_clip": 0.01136,
      "auxiliary_loss_mlp": 0.01031563,
      "balance_loss_clip": 1.04699516,
      "balance_loss_mlp": 1.02447927,
      "epoch": 0.8941261347922804,
      "flos": 25229041937280.0,
      "grad_norm": 2.909459196440415,
      "language_loss": 0.68370497,
      "learning_rate": 1.1633031150934591e-07,
      "loss": 0.70538062,
      "num_input_tokens_seen": 160819610,
      "step": 7436,
      "time_per_iteration": 2.605761766433716
    },
    {
      "auxiliary_loss_clip": 0.01153984,
      "auxiliary_loss_mlp": 0.01038199,
      "balance_loss_clip": 1.04989719,
      "balance_loss_mlp": 1.03088593,
      "epoch": 0.8942463776829195,
      "flos": 19537236806400.0,
      "grad_norm": 1.8993624182873627,
      "language_loss": 0.79636002,
      "learning_rate": 1.1606866066014176e-07,
      "loss": 0.81828189,
      "num_input_tokens_seen": 160838660,
      "step": 7437,
      "time_per_iteration": 2.4946765899658203
    },
    {
      "auxiliary_loss_clip": 0.01122198,
      "auxiliary_loss_mlp": 0.01028045,
      "balance_loss_clip": 1.04480362,
      "balance_loss_mlp": 1.02105618,
      "epoch": 0.8943666205735585,
      "flos": 22301567585280.0,
      "grad_norm": 2.2158719709381343,
      "language_loss": 0.75341177,
      "learning_rate": 1.1580729560182434e-07,
      "loss": 0.77491421,
      "num_input_tokens_seen": 160854515,
      "step": 7438,
      "time_per_iteration": 2.52472186088562
    },
    {
      "auxiliary_loss_clip": 0.01162116,
      "auxiliary_loss_mlp": 0.00760191,
      "balance_loss_clip": 1.0468812,
      "balance_loss_mlp": 1.00031042,
      "epoch": 0.8944868634641977,
      "flos": 18912893581440.0,
      "grad_norm": 1.6757554611644978,
      "language_loss": 0.70859236,
      "learning_rate": 1.1554621637404171e-07,
      "loss": 0.72781545,
      "num_input_tokens_seen": 160872605,
      "step": 7439,
      "time_per_iteration": 2.481964588165283
    },
    {
      "auxiliary_loss_clip": 0.01150441,
      "auxiliary_loss_mlp": 0.01023019,
      "balance_loss_clip": 1.04643595,
      "balance_loss_mlp": 1.01635528,
      "epoch": 0.8946071063548368,
      "flos": 14460904241280.0,
      "grad_norm": 2.112028418107696,
      "language_loss": 0.6068247,
      "learning_rate": 1.1528542301639999e-07,
      "loss": 0.62855923,
      "num_input_tokens_seen": 160889395,
      "step": 7440,
      "time_per_iteration": 3.166775941848755
    },
    {
      "auxiliary_loss_clip": 0.01121998,
      "auxiliary_loss_mlp": 0.01026113,
      "balance_loss_clip": 1.0417397,
      "balance_loss_mlp": 1.01856959,
      "epoch": 0.8947273492454758,
      "flos": 20084084438400.0,
      "grad_norm": 2.1974336730111164,
      "language_loss": 0.82231319,
      "learning_rate": 1.1502491556846105e-07,
      "loss": 0.84379423,
      "num_input_tokens_seen": 160907890,
      "step": 7441,
      "time_per_iteration": 2.5549275875091553
    },
    {
      "auxiliary_loss_clip": 0.01136374,
      "auxiliary_loss_mlp": 0.01023903,
      "balance_loss_clip": 1.04626226,
      "balance_loss_mlp": 1.01727486,
      "epoch": 0.894847592136115,
      "flos": 18550555136640.0,
      "grad_norm": 2.1440468063828164,
      "language_loss": 0.81231463,
      "learning_rate": 1.1476469406974331e-07,
      "loss": 0.83391738,
      "num_input_tokens_seen": 160923490,
      "step": 7442,
      "time_per_iteration": 3.264305591583252
    },
    {
      "auxiliary_loss_clip": 0.01160518,
      "auxiliary_loss_mlp": 0.0102365,
      "balance_loss_clip": 1.04757857,
      "balance_loss_mlp": 1.01681352,
      "epoch": 0.894967835026754,
      "flos": 23478468704640.0,
      "grad_norm": 1.5551516823389329,
      "language_loss": 0.76947987,
      "learning_rate": 1.1450475855972341e-07,
      "loss": 0.79132152,
      "num_input_tokens_seen": 160944280,
      "step": 7443,
      "time_per_iteration": 3.2135822772979736
    },
    {
      "auxiliary_loss_clip": 0.01131981,
      "auxiliary_loss_mlp": 0.00760756,
      "balance_loss_clip": 1.04137886,
      "balance_loss_mlp": 1.00028944,
      "epoch": 0.8950880779173931,
      "flos": 15188310564480.0,
      "grad_norm": 1.8687978739229985,
      "language_loss": 0.7036041,
      "learning_rate": 1.1424510907783158e-07,
      "loss": 0.72253144,
      "num_input_tokens_seen": 160961560,
      "step": 7444,
      "time_per_iteration": 2.5024240016937256
    },
    {
      "auxiliary_loss_clip": 0.01139113,
      "auxiliary_loss_mlp": 0.01022613,
      "balance_loss_clip": 1.04281092,
      "balance_loss_mlp": 1.0168221,
      "epoch": 0.8952083208080323,
      "flos": 22091957769600.0,
      "grad_norm": 1.6666500281271945,
      "language_loss": 0.82345879,
      "learning_rate": 1.1398574566345787e-07,
      "loss": 0.84507602,
      "num_input_tokens_seen": 160982195,
      "step": 7445,
      "time_per_iteration": 2.539337158203125
    },
    {
      "auxiliary_loss_clip": 0.01140588,
      "auxiliary_loss_mlp": 0.01026101,
      "balance_loss_clip": 1.04338813,
      "balance_loss_mlp": 1.01895475,
      "epoch": 0.8953285636986713,
      "flos": 23254026572160.0,
      "grad_norm": 2.078158186959809,
      "language_loss": 0.8251037,
      "learning_rate": 1.1372666835594702e-07,
      "loss": 0.84677064,
      "num_input_tokens_seen": 161000520,
      "step": 7446,
      "time_per_iteration": 2.5285720825195312
    },
    {
      "auxiliary_loss_clip": 0.01133592,
      "auxiliary_loss_mlp": 0.01020942,
      "balance_loss_clip": 1.0455569,
      "balance_loss_mlp": 1.01389694,
      "epoch": 0.8954488065893104,
      "flos": 16362661818240.0,
      "grad_norm": 1.8030804772716174,
      "language_loss": 0.72040135,
      "learning_rate": 1.1346787719460071e-07,
      "loss": 0.7419467,
      "num_input_tokens_seen": 161019405,
      "step": 7447,
      "time_per_iteration": 2.4996230602264404
    },
    {
      "auxiliary_loss_clip": 0.01133577,
      "auxiliary_loss_mlp": 0.01020934,
      "balance_loss_clip": 1.04350352,
      "balance_loss_mlp": 1.01446056,
      "epoch": 0.8955690494799495,
      "flos": 18257883120000.0,
      "grad_norm": 1.72502686710768,
      "language_loss": 0.72023404,
      "learning_rate": 1.1320937221867732e-07,
      "loss": 0.74177915,
      "num_input_tokens_seen": 161036985,
      "step": 7448,
      "time_per_iteration": 2.5138773918151855
    },
    {
      "auxiliary_loss_clip": 0.01131618,
      "auxiliary_loss_mlp": 0.01023663,
      "balance_loss_clip": 1.04153264,
      "balance_loss_mlp": 1.01722622,
      "epoch": 0.8956892923705886,
      "flos": 25447486498560.0,
      "grad_norm": 1.7951921376168438,
      "language_loss": 0.79578722,
      "learning_rate": 1.1295115346739192e-07,
      "loss": 0.81734002,
      "num_input_tokens_seen": 161056985,
      "step": 7449,
      "time_per_iteration": 2.5525810718536377
    },
    {
      "auxiliary_loss_clip": 0.01138899,
      "auxiliary_loss_mlp": 0.01026463,
      "balance_loss_clip": 1.04548371,
      "balance_loss_mlp": 1.01937556,
      "epoch": 0.8958095352612276,
      "flos": 52661883939840.0,
      "grad_norm": 2.4013256355457884,
      "language_loss": 0.7294091,
      "learning_rate": 1.1269322097991629e-07,
      "loss": 0.75106269,
      "num_input_tokens_seen": 161080270,
      "step": 7450,
      "time_per_iteration": 2.8006279468536377
    },
    {
      "auxiliary_loss_clip": 0.0115476,
      "auxiliary_loss_mlp": 0.01028667,
      "balance_loss_clip": 1.04804885,
      "balance_loss_mlp": 1.02125812,
      "epoch": 0.8959297781518668,
      "flos": 23186335392000.0,
      "grad_norm": 2.644634662661846,
      "language_loss": 0.67502606,
      "learning_rate": 1.1243557479537846e-07,
      "loss": 0.69686037,
      "num_input_tokens_seen": 161100160,
      "step": 7451,
      "time_per_iteration": 2.4947285652160645
    },
    {
      "auxiliary_loss_clip": 0.01163043,
      "auxiliary_loss_mlp": 0.01027613,
      "balance_loss_clip": 1.04613662,
      "balance_loss_mlp": 1.02032375,
      "epoch": 0.8960500210425059,
      "flos": 20334309557760.0,
      "grad_norm": 2.2007025409116805,
      "language_loss": 0.68958431,
      "learning_rate": 1.121782149528634e-07,
      "loss": 0.71149093,
      "num_input_tokens_seen": 161117260,
      "step": 7452,
      "time_per_iteration": 2.453657865524292
    },
    {
      "auxiliary_loss_clip": 0.01135394,
      "auxiliary_loss_mlp": 0.01030421,
      "balance_loss_clip": 1.04743433,
      "balance_loss_mlp": 1.02423954,
      "epoch": 0.8961702639331449,
      "flos": 19901694153600.0,
      "grad_norm": 1.8170416858581582,
      "language_loss": 0.78708577,
      "learning_rate": 1.1192114149141208e-07,
      "loss": 0.80874395,
      "num_input_tokens_seen": 161136895,
      "step": 7453,
      "time_per_iteration": 2.523460865020752
    },
    {
      "auxiliary_loss_clip": 0.01138004,
      "auxiliary_loss_mlp": 0.01027343,
      "balance_loss_clip": 1.04257989,
      "balance_loss_mlp": 1.01942205,
      "epoch": 0.8962905068237841,
      "flos": 12896348567040.0,
      "grad_norm": 2.1331954849948733,
      "language_loss": 0.6549015,
      "learning_rate": 1.1166435445002197e-07,
      "loss": 0.67655498,
      "num_input_tokens_seen": 161154565,
      "step": 7454,
      "time_per_iteration": 2.5036587715148926
    },
    {
      "auxiliary_loss_clip": 0.0115244,
      "auxiliary_loss_mlp": 0.01026927,
      "balance_loss_clip": 1.04877079,
      "balance_loss_mlp": 1.01988792,
      "epoch": 0.8964107497144231,
      "flos": 23440331439360.0,
      "grad_norm": 1.9806021595522296,
      "language_loss": 0.68371624,
      "learning_rate": 1.1140785386764818e-07,
      "loss": 0.7055099,
      "num_input_tokens_seen": 161173265,
      "step": 7455,
      "time_per_iteration": 3.2580881118774414
    },
    {
      "auxiliary_loss_clip": 0.01142439,
      "auxiliary_loss_mlp": 0.01026517,
      "balance_loss_clip": 1.04297376,
      "balance_loss_mlp": 1.01957917,
      "epoch": 0.8965309926050622,
      "flos": 19500176949120.0,
      "grad_norm": 1.8482093808832363,
      "language_loss": 0.69741678,
      "learning_rate": 1.1115163978320153e-07,
      "loss": 0.71910632,
      "num_input_tokens_seen": 161191995,
      "step": 7456,
      "time_per_iteration": 2.4956843852996826
    },
    {
      "auxiliary_loss_clip": 0.01154136,
      "auxiliary_loss_mlp": 0.00760955,
      "balance_loss_clip": 1.04702401,
      "balance_loss_mlp": 1.00029397,
      "epoch": 0.8966512354957014,
      "flos": 28658008022400.0,
      "grad_norm": 2.1507491962526046,
      "language_loss": 0.82433355,
      "learning_rate": 1.1089571223554917e-07,
      "loss": 0.84348446,
      "num_input_tokens_seen": 161212880,
      "step": 7457,
      "time_per_iteration": 2.5518147945404053
    },
    {
      "auxiliary_loss_clip": 0.011488,
      "auxiliary_loss_mlp": 0.01025729,
      "balance_loss_clip": 1.04337263,
      "balance_loss_mlp": 1.0188266,
      "epoch": 0.8967714783863404,
      "flos": 23370916406400.0,
      "grad_norm": 1.6205393697015267,
      "language_loss": 0.85538733,
      "learning_rate": 1.1064007126351537e-07,
      "loss": 0.87713265,
      "num_input_tokens_seen": 161233595,
      "step": 7458,
      "time_per_iteration": 2.551846981048584
    },
    {
      "auxiliary_loss_clip": 0.01131153,
      "auxiliary_loss_mlp": 0.01025815,
      "balance_loss_clip": 1.04457664,
      "balance_loss_mlp": 1.01857305,
      "epoch": 0.8968917212769795,
      "flos": 24535175938560.0,
      "grad_norm": 2.2252109249181147,
      "language_loss": 0.76584154,
      "learning_rate": 1.1038471690588003e-07,
      "loss": 0.78741121,
      "num_input_tokens_seen": 161252740,
      "step": 7459,
      "time_per_iteration": 2.5679402351379395
    },
    {
      "auxiliary_loss_clip": 0.01106162,
      "auxiliary_loss_mlp": 0.01023798,
      "balance_loss_clip": 1.04195023,
      "balance_loss_mlp": 1.0170958,
      "epoch": 0.8970119641676186,
      "flos": 23475416048640.0,
      "grad_norm": 1.8712703579686236,
      "language_loss": 0.79938948,
      "learning_rate": 1.1012964920138145e-07,
      "loss": 0.82068908,
      "num_input_tokens_seen": 161272325,
      "step": 7460,
      "time_per_iteration": 2.611543893814087
    },
    {
      "auxiliary_loss_clip": 0.01129379,
      "auxiliary_loss_mlp": 0.01025774,
      "balance_loss_clip": 1.04149199,
      "balance_loss_mlp": 1.01869869,
      "epoch": 0.8971322070582577,
      "flos": 24538192680960.0,
      "grad_norm": 1.6115542468821047,
      "language_loss": 0.75844646,
      "learning_rate": 1.0987486818871205e-07,
      "loss": 0.779998,
      "num_input_tokens_seen": 161295915,
      "step": 7461,
      "time_per_iteration": 2.5886144638061523
    },
    {
      "auxiliary_loss_clip": 0.01148992,
      "auxiliary_loss_mlp": 0.00761069,
      "balance_loss_clip": 1.04524672,
      "balance_loss_mlp": 1.00036502,
      "epoch": 0.8972524499488967,
      "flos": 21797454159360.0,
      "grad_norm": 2.1668010811060694,
      "language_loss": 0.73092866,
      "learning_rate": 1.0962037390652245e-07,
      "loss": 0.75002927,
      "num_input_tokens_seen": 161314935,
      "step": 7462,
      "time_per_iteration": 2.4966514110565186
    },
    {
      "auxiliary_loss_clip": 0.01132905,
      "auxiliary_loss_mlp": 0.01028569,
      "balance_loss_clip": 1.04324603,
      "balance_loss_mlp": 1.0218513,
      "epoch": 0.8973726928395359,
      "flos": 21726243446400.0,
      "grad_norm": 1.6758241947683257,
      "language_loss": 0.71576154,
      "learning_rate": 1.0936616639341911e-07,
      "loss": 0.73737621,
      "num_input_tokens_seen": 161335225,
      "step": 7463,
      "time_per_iteration": 2.5411312580108643
    },
    {
      "auxiliary_loss_clip": 0.0103955,
      "auxiliary_loss_mlp": 0.01002419,
      "balance_loss_clip": 1.00897431,
      "balance_loss_mlp": 1.00139928,
      "epoch": 0.897492935730175,
      "flos": 53837100097920.0,
      "grad_norm": 0.8129002436099186,
      "language_loss": 0.54798645,
      "learning_rate": 1.0911224568796473e-07,
      "loss": 0.56840611,
      "num_input_tokens_seen": 161393420,
      "step": 7464,
      "time_per_iteration": 3.110926866531372
    },
    {
      "auxiliary_loss_clip": 0.01146853,
      "auxiliary_loss_mlp": 0.01023963,
      "balance_loss_clip": 1.0460999,
      "balance_loss_mlp": 1.01699495,
      "epoch": 0.897613178620814,
      "flos": 18290346036480.0,
      "grad_norm": 2.119088007614086,
      "language_loss": 0.70793706,
      "learning_rate": 1.0885861182867984e-07,
      "loss": 0.72964525,
      "num_input_tokens_seen": 161411525,
      "step": 7465,
      "time_per_iteration": 2.475480079650879
    },
    {
      "auxiliary_loss_clip": 0.0113641,
      "auxiliary_loss_mlp": 0.01027551,
      "balance_loss_clip": 1.04358995,
      "balance_loss_mlp": 1.02103615,
      "epoch": 0.8977334215114532,
      "flos": 32993718059520.0,
      "grad_norm": 13.499940689087065,
      "language_loss": 0.70925289,
      "learning_rate": 1.0860526485403942e-07,
      "loss": 0.73089254,
      "num_input_tokens_seen": 161432800,
      "step": 7466,
      "time_per_iteration": 3.430245876312256
    },
    {
      "auxiliary_loss_clip": 0.01162232,
      "auxiliary_loss_mlp": 0.01023247,
      "balance_loss_clip": 1.04652929,
      "balance_loss_mlp": 1.01653266,
      "epoch": 0.8978536644020922,
      "flos": 15195636938880.0,
      "grad_norm": 1.5223614169245245,
      "language_loss": 0.76937968,
      "learning_rate": 1.0835220480247675e-07,
      "loss": 0.79123449,
      "num_input_tokens_seen": 161451295,
      "step": 7467,
      "time_per_iteration": 2.4269371032714844
    },
    {
      "auxiliary_loss_clip": 0.01132855,
      "auxiliary_loss_mlp": 0.01028946,
      "balance_loss_clip": 1.04478419,
      "balance_loss_mlp": 1.02231479,
      "epoch": 0.8979739072927313,
      "flos": 18004389863040.0,
      "grad_norm": 2.0641071025158992,
      "language_loss": 0.83205509,
      "learning_rate": 1.0809943171238067e-07,
      "loss": 0.8536731,
      "num_input_tokens_seen": 161469220,
      "step": 7468,
      "time_per_iteration": 3.280802011489868
    },
    {
      "auxiliary_loss_clip": 0.01142962,
      "auxiliary_loss_mlp": 0.01030336,
      "balance_loss_clip": 1.04590416,
      "balance_loss_mlp": 1.02255774,
      "epoch": 0.8980941501833704,
      "flos": 22271546793600.0,
      "grad_norm": 3.0133644366609564,
      "language_loss": 0.63119388,
      "learning_rate": 1.078469456220965e-07,
      "loss": 0.65292692,
      "num_input_tokens_seen": 161489375,
      "step": 7469,
      "time_per_iteration": 3.2443292140960693
    },
    {
      "auxiliary_loss_clip": 0.01146077,
      "auxiliary_loss_mlp": 0.01023525,
      "balance_loss_clip": 1.04392767,
      "balance_loss_mlp": 1.01671243,
      "epoch": 0.8982143930740095,
      "flos": 37560729726720.0,
      "grad_norm": 1.7322224856893929,
      "language_loss": 0.69429421,
      "learning_rate": 1.0759474656992606e-07,
      "loss": 0.71599025,
      "num_input_tokens_seen": 161512145,
      "step": 7470,
      "time_per_iteration": 2.6585495471954346
    },
    {
      "auxiliary_loss_clip": 0.0113698,
      "auxiliary_loss_mlp": 0.01024885,
      "balance_loss_clip": 1.04101002,
      "balance_loss_mlp": 1.01792657,
      "epoch": 0.8983346359646486,
      "flos": 18076893465600.0,
      "grad_norm": 3.114738911094555,
      "language_loss": 0.78126603,
      "learning_rate": 1.0734283459412785e-07,
      "loss": 0.8028847,
      "num_input_tokens_seen": 161528995,
      "step": 7471,
      "time_per_iteration": 2.497157096862793
    },
    {
      "auxiliary_loss_clip": 0.01110702,
      "auxiliary_loss_mlp": 0.01028791,
      "balance_loss_clip": 1.04030585,
      "balance_loss_mlp": 1.0207206,
      "epoch": 0.8984548788552876,
      "flos": 20558895344640.0,
      "grad_norm": 1.6545965743637256,
      "language_loss": 0.80343813,
      "learning_rate": 1.0709120973291707e-07,
      "loss": 0.82483304,
      "num_input_tokens_seen": 161548775,
      "step": 7472,
      "time_per_iteration": 2.5817360877990723
    },
    {
      "auxiliary_loss_clip": 0.01164421,
      "auxiliary_loss_mlp": 0.01025965,
      "balance_loss_clip": 1.04827023,
      "balance_loss_mlp": 1.01858616,
      "epoch": 0.8985751217459268,
      "flos": 17785442511360.0,
      "grad_norm": 3.0913500382141312,
      "language_loss": 0.77688742,
      "learning_rate": 1.0683987202446475e-07,
      "loss": 0.79879129,
      "num_input_tokens_seen": 161566960,
      "step": 7473,
      "time_per_iteration": 2.4521634578704834
    },
    {
      "auxiliary_loss_clip": 0.0115223,
      "auxiliary_loss_mlp": 0.01026046,
      "balance_loss_clip": 1.04579496,
      "balance_loss_mlp": 1.01929832,
      "epoch": 0.8986953646365659,
      "flos": 21617003208960.0,
      "grad_norm": 1.795356077941309,
      "language_loss": 0.6985867,
      "learning_rate": 1.0658882150689862e-07,
      "loss": 0.72036946,
      "num_input_tokens_seen": 161585820,
      "step": 7474,
      "time_per_iteration": 2.4952924251556396
    },
    {
      "auxiliary_loss_clip": 0.01124242,
      "auxiliary_loss_mlp": 0.01029055,
      "balance_loss_clip": 1.04340482,
      "balance_loss_mlp": 1.02149093,
      "epoch": 0.8988156075272049,
      "flos": 14027355083520.0,
      "grad_norm": 2.2232479736372768,
      "language_loss": 0.78403056,
      "learning_rate": 1.0633805821830288e-07,
      "loss": 0.80556351,
      "num_input_tokens_seen": 161602505,
      "step": 7475,
      "time_per_iteration": 2.5213968753814697
    },
    {
      "auxiliary_loss_clip": 0.01136803,
      "auxiliary_loss_mlp": 0.01022219,
      "balance_loss_clip": 1.04563403,
      "balance_loss_mlp": 1.01549566,
      "epoch": 0.8989358504178441,
      "flos": 29059202004480.0,
      "grad_norm": 2.3011930441441995,
      "language_loss": 0.826065,
      "learning_rate": 1.0608758219671753e-07,
      "loss": 0.8476553,
      "num_input_tokens_seen": 161621545,
      "step": 7476,
      "time_per_iteration": 2.578660726547241
    },
    {
      "auxiliary_loss_clip": 0.01139468,
      "auxiliary_loss_mlp": 0.01029267,
      "balance_loss_clip": 1.04446089,
      "balance_loss_mlp": 1.02254677,
      "epoch": 0.8990560933084831,
      "flos": 20230420446720.0,
      "grad_norm": 1.5018307382456926,
      "language_loss": 0.70488739,
      "learning_rate": 1.0583739348014065e-07,
      "loss": 0.72657478,
      "num_input_tokens_seen": 161642630,
      "step": 7477,
      "time_per_iteration": 2.5316243171691895
    },
    {
      "auxiliary_loss_clip": 0.01163763,
      "auxiliary_loss_mlp": 0.01020837,
      "balance_loss_clip": 1.04833055,
      "balance_loss_mlp": 1.01399732,
      "epoch": 0.8991763361991222,
      "flos": 25520672459520.0,
      "grad_norm": 2.208276624875259,
      "language_loss": 0.84706175,
      "learning_rate": 1.0558749210652518e-07,
      "loss": 0.86890775,
      "num_input_tokens_seen": 161662560,
      "step": 7478,
      "time_per_iteration": 2.4788291454315186
    },
    {
      "auxiliary_loss_clip": 0.01126581,
      "auxiliary_loss_mlp": 0.01028604,
      "balance_loss_clip": 1.0435276,
      "balance_loss_mlp": 1.02159202,
      "epoch": 0.8992965790897613,
      "flos": 25119191168640.0,
      "grad_norm": 1.5644262317940125,
      "language_loss": 0.85515124,
      "learning_rate": 1.053378781137808e-07,
      "loss": 0.87670302,
      "num_input_tokens_seen": 161683480,
      "step": 7479,
      "time_per_iteration": 2.5802230834960938
    },
    {
      "auxiliary_loss_clip": 0.01134825,
      "auxiliary_loss_mlp": 0.01026872,
      "balance_loss_clip": 1.04352546,
      "balance_loss_mlp": 1.01906931,
      "epoch": 0.8994168219804004,
      "flos": 16070815814400.0,
      "grad_norm": 1.7254861305304443,
      "language_loss": 0.77797115,
      "learning_rate": 1.0508855153977392e-07,
      "loss": 0.79958814,
      "num_input_tokens_seen": 161699945,
      "step": 7480,
      "time_per_iteration": 2.487196683883667
    },
    {
      "auxiliary_loss_clip": 0.01148596,
      "auxiliary_loss_mlp": 0.01029211,
      "balance_loss_clip": 1.04328418,
      "balance_loss_mlp": 1.02265167,
      "epoch": 0.8995370648710395,
      "flos": 24825764966400.0,
      "grad_norm": 2.293590329276333,
      "language_loss": 0.67037427,
      "learning_rate": 1.0483951242232669e-07,
      "loss": 0.69215232,
      "num_input_tokens_seen": 161720420,
      "step": 7481,
      "time_per_iteration": 3.312591314315796
    },
    {
      "auxiliary_loss_clip": 0.0105507,
      "auxiliary_loss_mlp": 0.01000815,
      "balance_loss_clip": 1.00791121,
      "balance_loss_mlp": 0.99978334,
      "epoch": 0.8996573077616786,
      "flos": 63116238378240.0,
      "grad_norm": 0.9760248461513609,
      "language_loss": 0.57760805,
      "learning_rate": 1.0459076079921936e-07,
      "loss": 0.59816688,
      "num_input_tokens_seen": 161773080,
      "step": 7482,
      "time_per_iteration": 3.0893590450286865
    },
    {
      "auxiliary_loss_clip": 0.01131394,
      "auxiliary_loss_mlp": 0.01031582,
      "balance_loss_clip": 1.04457569,
      "balance_loss_mlp": 1.02456951,
      "epoch": 0.8997775506523177,
      "flos": 18219674027520.0,
      "grad_norm": 2.169149424833463,
      "language_loss": 0.84844077,
      "learning_rate": 1.0434229670818618e-07,
      "loss": 0.87007052,
      "num_input_tokens_seen": 161789755,
      "step": 7483,
      "time_per_iteration": 2.4851512908935547
    },
    {
      "auxiliary_loss_clip": 0.01129021,
      "auxiliary_loss_mlp": 0.01025345,
      "balance_loss_clip": 1.0440799,
      "balance_loss_mlp": 1.01846087,
      "epoch": 0.8998977935429567,
      "flos": 24166768095360.0,
      "grad_norm": 1.4448591315838006,
      "language_loss": 0.79861963,
      "learning_rate": 1.0409412018691944e-07,
      "loss": 0.82016325,
      "num_input_tokens_seen": 161810220,
      "step": 7484,
      "time_per_iteration": 2.5464789867401123
    },
    {
      "auxiliary_loss_clip": 0.01133274,
      "auxiliary_loss_mlp": 0.01030074,
      "balance_loss_clip": 1.04570222,
      "balance_loss_mlp": 1.02296865,
      "epoch": 0.9000180364335959,
      "flos": 20773030273920.0,
      "grad_norm": 2.0062908454965496,
      "language_loss": 0.75151974,
      "learning_rate": 1.0384623127306724e-07,
      "loss": 0.77315319,
      "num_input_tokens_seen": 161827565,
      "step": 7485,
      "time_per_iteration": 2.604987382888794
    },
    {
      "auxiliary_loss_clip": 0.01118831,
      "auxiliary_loss_mlp": 0.01025808,
      "balance_loss_clip": 1.04153621,
      "balance_loss_mlp": 1.01928115,
      "epoch": 0.900138279324235,
      "flos": 19205745166080.0,
      "grad_norm": 1.720456731000409,
      "language_loss": 0.79304266,
      "learning_rate": 1.0359863000423397e-07,
      "loss": 0.81448907,
      "num_input_tokens_seen": 161845700,
      "step": 7486,
      "time_per_iteration": 2.5270373821258545
    },
    {
      "auxiliary_loss_clip": 0.01162524,
      "auxiliary_loss_mlp": 0.01028202,
      "balance_loss_clip": 1.0469166,
      "balance_loss_mlp": 1.02180052,
      "epoch": 0.900258522214874,
      "flos": 28731158069760.0,
      "grad_norm": 1.7493665357723611,
      "language_loss": 0.71955711,
      "learning_rate": 1.0335131641798112e-07,
      "loss": 0.74146438,
      "num_input_tokens_seen": 161867660,
      "step": 7487,
      "time_per_iteration": 2.525736093521118
    },
    {
      "auxiliary_loss_clip": 0.01033556,
      "auxiliary_loss_mlp": 0.01000271,
      "balance_loss_clip": 1.00719714,
      "balance_loss_mlp": 0.99920416,
      "epoch": 0.9003787651055132,
      "flos": 58280685655680.0,
      "grad_norm": 0.8028463903110241,
      "language_loss": 0.55687737,
      "learning_rate": 1.0310429055182512e-07,
      "loss": 0.57721567,
      "num_input_tokens_seen": 161921980,
      "step": 7488,
      "time_per_iteration": 2.952817678451538
    },
    {
      "auxiliary_loss_clip": 0.01126393,
      "auxiliary_loss_mlp": 0.01026319,
      "balance_loss_clip": 1.04525256,
      "balance_loss_mlp": 1.01867735,
      "epoch": 0.9004990079961522,
      "flos": 25556475340800.0,
      "grad_norm": 1.5818077682595881,
      "language_loss": 0.73708564,
      "learning_rate": 1.0285755244324024e-07,
      "loss": 0.75861275,
      "num_input_tokens_seen": 161942725,
      "step": 7489,
      "time_per_iteration": 2.6101534366607666
    },
    {
      "auxiliary_loss_clip": 0.0113784,
      "auxiliary_loss_mlp": 0.00760625,
      "balance_loss_clip": 1.04205406,
      "balance_loss_mlp": 1.00032806,
      "epoch": 0.9006192508867913,
      "flos": 23335185352320.0,
      "grad_norm": 1.517151244312872,
      "language_loss": 0.68486702,
      "learning_rate": 1.0261110212965629e-07,
      "loss": 0.7038517,
      "num_input_tokens_seen": 161964520,
      "step": 7490,
      "time_per_iteration": 2.5561819076538086
    },
    {
      "auxiliary_loss_clip": 0.01133608,
      "auxiliary_loss_mlp": 0.01023006,
      "balance_loss_clip": 1.042647,
      "balance_loss_mlp": 1.01647377,
      "epoch": 0.9007394937774305,
      "flos": 18040300485120.0,
      "grad_norm": 1.9181939712576286,
      "language_loss": 0.79236823,
      "learning_rate": 1.023649396484596e-07,
      "loss": 0.81393439,
      "num_input_tokens_seen": 161983575,
      "step": 7491,
      "time_per_iteration": 2.5147461891174316
    },
    {
      "auxiliary_loss_clip": 0.01163252,
      "auxiliary_loss_mlp": 0.0102622,
      "balance_loss_clip": 1.04727352,
      "balance_loss_mlp": 1.01913345,
      "epoch": 0.9008597366680695,
      "flos": 43068456633600.0,
      "grad_norm": 1.7760916080221485,
      "language_loss": 0.67558622,
      "learning_rate": 1.0211906503699275e-07,
      "loss": 0.69748092,
      "num_input_tokens_seen": 162006550,
      "step": 7492,
      "time_per_iteration": 3.387376308441162
    },
    {
      "auxiliary_loss_clip": 0.01154197,
      "auxiliary_loss_mlp": 0.01024406,
      "balance_loss_clip": 1.04905343,
      "balance_loss_mlp": 1.01745939,
      "epoch": 0.9009799795587086,
      "flos": 14939055112320.0,
      "grad_norm": 2.7793425758749186,
      "language_loss": 0.8214485,
      "learning_rate": 1.0187347833255455e-07,
      "loss": 0.84323454,
      "num_input_tokens_seen": 162022455,
      "step": 7493,
      "time_per_iteration": 2.4649908542633057
    },
    {
      "auxiliary_loss_clip": 0.01161527,
      "auxiliary_loss_mlp": 0.01031461,
      "balance_loss_clip": 1.04779911,
      "balance_loss_mlp": 1.02460372,
      "epoch": 0.9011002224493477,
      "flos": 21579584215680.0,
      "grad_norm": 1.6534673814803902,
      "language_loss": 0.79717535,
      "learning_rate": 1.0162817957240056e-07,
      "loss": 0.81910527,
      "num_input_tokens_seen": 162042350,
      "step": 7494,
      "time_per_iteration": 3.2698261737823486
    },
    {
      "auxiliary_loss_clip": 0.01046369,
      "auxiliary_loss_mlp": 0.01001797,
      "balance_loss_clip": 1.00892019,
      "balance_loss_mlp": 1.00072384,
      "epoch": 0.9012204653399868,
      "flos": 71166367883520.0,
      "grad_norm": 0.9189728465620935,
      "language_loss": 0.63017696,
      "learning_rate": 1.0138316879374253e-07,
      "loss": 0.65065861,
      "num_input_tokens_seen": 162111640,
      "step": 7495,
      "time_per_iteration": 3.9736175537109375
    },
    {
      "auxiliary_loss_clip": 0.01141582,
      "auxiliary_loss_mlp": 0.01031062,
      "balance_loss_clip": 1.04859424,
      "balance_loss_mlp": 1.02373695,
      "epoch": 0.9013407082306258,
      "flos": 15594963413760.0,
      "grad_norm": 2.363472194692641,
      "language_loss": 0.74235451,
      "learning_rate": 1.0113844603374833e-07,
      "loss": 0.764081,
      "num_input_tokens_seen": 162128165,
      "step": 7496,
      "time_per_iteration": 2.5218915939331055
    },
    {
      "auxiliary_loss_clip": 0.01134547,
      "auxiliary_loss_mlp": 0.01029906,
      "balance_loss_clip": 1.04266858,
      "balance_loss_mlp": 1.0227567,
      "epoch": 0.901460951121265,
      "flos": 15049157276160.0,
      "grad_norm": 2.1393418234394574,
      "language_loss": 0.71383053,
      "learning_rate": 1.0089401132954178e-07,
      "loss": 0.73547506,
      "num_input_tokens_seen": 162146145,
      "step": 7497,
      "time_per_iteration": 2.499688148498535
    },
    {
      "auxiliary_loss_clip": 0.01133996,
      "auxiliary_loss_mlp": 0.01022582,
      "balance_loss_clip": 1.04505587,
      "balance_loss_mlp": 1.01576638,
      "epoch": 0.9015811940119041,
      "flos": 22236857233920.0,
      "grad_norm": 1.6619970583703152,
      "language_loss": 0.72475213,
      "learning_rate": 1.006498647182037e-07,
      "loss": 0.74631792,
      "num_input_tokens_seen": 162164800,
      "step": 7498,
      "time_per_iteration": 2.534601926803589
    },
    {
      "auxiliary_loss_clip": 0.01092062,
      "auxiliary_loss_mlp": 0.01038246,
      "balance_loss_clip": 1.03939044,
      "balance_loss_mlp": 1.03099477,
      "epoch": 0.9017014369025431,
      "flos": 24973824827520.0,
      "grad_norm": 1.9075477075488705,
      "language_loss": 0.71420908,
      "learning_rate": 1.004060062367713e-07,
      "loss": 0.73551214,
      "num_input_tokens_seen": 162185895,
      "step": 7499,
      "time_per_iteration": 2.6510870456695557
    },
    {
      "auxiliary_loss_clip": 0.01151838,
      "auxiliary_loss_mlp": 0.01030875,
      "balance_loss_clip": 1.04721427,
      "balance_loss_mlp": 1.02427077,
      "epoch": 0.9018216797931822,
      "flos": 18114168804480.0,
      "grad_norm": 1.6451170771158339,
      "language_loss": 0.69276577,
      "learning_rate": 1.0016243592223728e-07,
      "loss": 0.71459293,
      "num_input_tokens_seen": 162206295,
      "step": 7500,
      "time_per_iteration": 2.5159363746643066
    },
    {
      "auxiliary_loss_clip": 0.01090219,
      "auxiliary_loss_mlp": 0.01024392,
      "balance_loss_clip": 1.04088032,
      "balance_loss_mlp": 1.01734066,
      "epoch": 0.9019419226838213,
      "flos": 37268452759680.0,
      "grad_norm": 1.9391163216616054,
      "language_loss": 0.65897995,
      "learning_rate": 9.991915381155114e-08,
      "loss": 0.68012601,
      "num_input_tokens_seen": 162229275,
      "step": 7501,
      "time_per_iteration": 2.724933385848999
    },
    {
      "auxiliary_loss_clip": 0.01152926,
      "auxiliary_loss_mlp": 0.01028176,
      "balance_loss_clip": 1.04628479,
      "balance_loss_mlp": 1.02101159,
      "epoch": 0.9020621655744604,
      "flos": 23441121538560.0,
      "grad_norm": 8.101052024098221,
      "language_loss": 0.74599332,
      "learning_rate": 9.967615994161871e-08,
      "loss": 0.76780427,
      "num_input_tokens_seen": 162248935,
      "step": 7502,
      "time_per_iteration": 2.529355764389038
    },
    {
      "auxiliary_loss_clip": 0.01161818,
      "auxiliary_loss_mlp": 0.01026779,
      "balance_loss_clip": 1.04751754,
      "balance_loss_mlp": 1.02033257,
      "epoch": 0.9021824084650995,
      "flos": 22857465444480.0,
      "grad_norm": 1.9214071505250454,
      "language_loss": 0.78274548,
      "learning_rate": 9.943345434930161e-08,
      "loss": 0.80463147,
      "num_input_tokens_seen": 162269185,
      "step": 7503,
      "time_per_iteration": 2.4573049545288086
    },
    {
      "auxiliary_loss_clip": 0.01120519,
      "auxiliary_loss_mlp": 0.01021157,
      "balance_loss_clip": 1.04462183,
      "balance_loss_mlp": 1.01421595,
      "epoch": 0.9023026513557386,
      "flos": 22127581082880.0,
      "grad_norm": 5.489373961705314,
      "language_loss": 0.6887958,
      "learning_rate": 9.919103707141885e-08,
      "loss": 0.71021259,
      "num_input_tokens_seen": 162288065,
      "step": 7504,
      "time_per_iteration": 2.543691635131836
    },
    {
      "auxiliary_loss_clip": 0.01149148,
      "auxiliary_loss_mlp": 0.01031081,
      "balance_loss_clip": 1.04666746,
      "balance_loss_mlp": 1.0233351,
      "epoch": 0.9024228942463777,
      "flos": 24199087357440.0,
      "grad_norm": 3.2708931545217967,
      "language_loss": 0.76516986,
      "learning_rate": 9.89489081447441e-08,
      "loss": 0.78697217,
      "num_input_tokens_seen": 162305265,
      "step": 7505,
      "time_per_iteration": 2.494748592376709
    },
    {
      "auxiliary_loss_clip": 0.011336,
      "auxiliary_loss_mlp": 0.01028322,
      "balance_loss_clip": 1.04212785,
      "balance_loss_mlp": 1.02092481,
      "epoch": 0.9025431371370167,
      "flos": 25008262992000.0,
      "grad_norm": 1.9243922285930786,
      "language_loss": 0.83041209,
      "learning_rate": 9.870706760600844e-08,
      "loss": 0.85203135,
      "num_input_tokens_seen": 162325215,
      "step": 7506,
      "time_per_iteration": 2.54723858833313
    },
    {
      "auxiliary_loss_clip": 0.01111143,
      "auxiliary_loss_mlp": 0.01034209,
      "balance_loss_clip": 1.0463407,
      "balance_loss_mlp": 1.0268209,
      "epoch": 0.9026633800276559,
      "flos": 18952862440320.0,
      "grad_norm": 1.912244342603789,
      "language_loss": 0.72878385,
      "learning_rate": 9.846551549189918e-08,
      "loss": 0.75023735,
      "num_input_tokens_seen": 162344820,
      "step": 7507,
      "time_per_iteration": 3.391357183456421
    },
    {
      "auxiliary_loss_clip": 0.01135264,
      "auxiliary_loss_mlp": 0.0103047,
      "balance_loss_clip": 1.04461706,
      "balance_loss_mlp": 1.02298379,
      "epoch": 0.902783622918295,
      "flos": 32416059536640.0,
      "grad_norm": 1.9272336757262407,
      "language_loss": 0.68760484,
      "learning_rate": 9.822425183905902e-08,
      "loss": 0.70926219,
      "num_input_tokens_seen": 162365345,
      "step": 7508,
      "time_per_iteration": 2.626051187515259
    },
    {
      "auxiliary_loss_clip": 0.01024719,
      "auxiliary_loss_mlp": 0.01000882,
      "balance_loss_clip": 1.00777125,
      "balance_loss_mlp": 0.99979752,
      "epoch": 0.902903865808934,
      "flos": 63717453244800.0,
      "grad_norm": 0.9233634456334515,
      "language_loss": 0.75238323,
      "learning_rate": 9.798327668408823e-08,
      "loss": 0.77263916,
      "num_input_tokens_seen": 162426980,
      "step": 7509,
      "time_per_iteration": 3.2298030853271484
    },
    {
      "auxiliary_loss_clip": 0.01166386,
      "auxiliary_loss_mlp": 0.01025971,
      "balance_loss_clip": 1.04777789,
      "balance_loss_mlp": 1.01833606,
      "epoch": 0.9030241086995732,
      "flos": 23804034600960.0,
      "grad_norm": 1.8709971239884235,
      "language_loss": 0.68870342,
      "learning_rate": 9.774259006354158e-08,
      "loss": 0.71062696,
      "num_input_tokens_seen": 162447050,
      "step": 7510,
      "time_per_iteration": 2.4815921783447266
    },
    {
      "auxiliary_loss_clip": 0.01140261,
      "auxiliary_loss_mlp": 0.01026776,
      "balance_loss_clip": 1.04457378,
      "balance_loss_mlp": 1.01932228,
      "epoch": 0.9031443515902122,
      "flos": 26395887248640.0,
      "grad_norm": 1.8341046496718703,
      "language_loss": 0.76209921,
      "learning_rate": 9.750219201393184e-08,
      "loss": 0.78376955,
      "num_input_tokens_seen": 162467015,
      "step": 7511,
      "time_per_iteration": 2.564682722091675
    },
    {
      "auxiliary_loss_clip": 0.01148868,
      "auxiliary_loss_mlp": 0.0102787,
      "balance_loss_clip": 1.04609859,
      "balance_loss_mlp": 1.02119088,
      "epoch": 0.9032645944808513,
      "flos": 24939350749440.0,
      "grad_norm": 1.662540853280006,
      "language_loss": 0.77706259,
      "learning_rate": 9.726208257172697e-08,
      "loss": 0.79882997,
      "num_input_tokens_seen": 162488710,
      "step": 7512,
      "time_per_iteration": 2.5492446422576904
    },
    {
      "auxiliary_loss_clip": 0.01163065,
      "auxiliary_loss_mlp": 0.01024551,
      "balance_loss_clip": 1.04717135,
      "balance_loss_mlp": 1.01743138,
      "epoch": 0.9033848373714904,
      "flos": 21178821196800.0,
      "grad_norm": 1.8620945613384838,
      "language_loss": 0.74410975,
      "learning_rate": 9.702226177335115e-08,
      "loss": 0.76598585,
      "num_input_tokens_seen": 162507205,
      "step": 7513,
      "time_per_iteration": 2.479057550430298
    },
    {
      "auxiliary_loss_clip": 0.01135288,
      "auxiliary_loss_mlp": 0.0102751,
      "balance_loss_clip": 1.04686522,
      "balance_loss_mlp": 1.02027977,
      "epoch": 0.9035050802621295,
      "flos": 26286359702400.0,
      "grad_norm": 1.5865305521496562,
      "language_loss": 0.72612453,
      "learning_rate": 9.67827296551853e-08,
      "loss": 0.74775255,
      "num_input_tokens_seen": 162528490,
      "step": 7514,
      "time_per_iteration": 2.555086851119995
    },
    {
      "auxiliary_loss_clip": 0.0112577,
      "auxiliary_loss_mlp": 0.00760089,
      "balance_loss_clip": 1.04067826,
      "balance_loss_mlp": 1.00036478,
      "epoch": 0.9036253231527686,
      "flos": 24204546224640.0,
      "grad_norm": 1.912592880445081,
      "language_loss": 0.68280953,
      "learning_rate": 9.65434862535659e-08,
      "loss": 0.70166814,
      "num_input_tokens_seen": 162547860,
      "step": 7515,
      "time_per_iteration": 2.533799409866333
    },
    {
      "auxiliary_loss_clip": 0.01140731,
      "auxiliary_loss_mlp": 0.0102699,
      "balance_loss_clip": 1.04582834,
      "balance_loss_mlp": 1.01943207,
      "epoch": 0.9037455660434077,
      "flos": 18072655660800.0,
      "grad_norm": 2.711142443370385,
      "language_loss": 0.65098083,
      "learning_rate": 9.630453160478635e-08,
      "loss": 0.67265797,
      "num_input_tokens_seen": 162563215,
      "step": 7516,
      "time_per_iteration": 2.478158950805664
    },
    {
      "auxiliary_loss_clip": 0.01108287,
      "auxiliary_loss_mlp": 0.01030156,
      "balance_loss_clip": 1.0411948,
      "balance_loss_mlp": 1.02347183,
      "epoch": 0.9038658089340468,
      "flos": 24060795995520.0,
      "grad_norm": 1.607858542849952,
      "language_loss": 0.82282364,
      "learning_rate": 9.60658657450959e-08,
      "loss": 0.84420806,
      "num_input_tokens_seen": 162583515,
      "step": 7517,
      "time_per_iteration": 2.613612174987793
    },
    {
      "auxiliary_loss_clip": 0.0112146,
      "auxiliary_loss_mlp": 0.01022674,
      "balance_loss_clip": 1.0411222,
      "balance_loss_mlp": 1.01551819,
      "epoch": 0.9039860518246858,
      "flos": 21834298535040.0,
      "grad_norm": 1.6204822065489703,
      "language_loss": 0.79641712,
      "learning_rate": 9.582748871069979e-08,
      "loss": 0.81785846,
      "num_input_tokens_seen": 162602955,
      "step": 7518,
      "time_per_iteration": 3.3233110904693604
    },
    {
      "auxiliary_loss_clip": 0.01137355,
      "auxiliary_loss_mlp": 0.007602,
      "balance_loss_clip": 1.04342592,
      "balance_loss_mlp": 1.00032616,
      "epoch": 0.904106294715325,
      "flos": 26614870513920.0,
      "grad_norm": 1.870590205917733,
      "language_loss": 0.83027351,
      "learning_rate": 9.558940053775954e-08,
      "loss": 0.84924906,
      "num_input_tokens_seen": 162621595,
      "step": 7519,
      "time_per_iteration": 2.5785863399505615
    },
    {
      "auxiliary_loss_clip": 0.01151513,
      "auxiliary_loss_mlp": 0.01028688,
      "balance_loss_clip": 1.04853773,
      "balance_loss_mlp": 1.02175021,
      "epoch": 0.904226537605964,
      "flos": 17785693906560.0,
      "grad_norm": 2.1657524778564294,
      "language_loss": 0.67661536,
      "learning_rate": 9.535160126239294e-08,
      "loss": 0.69841731,
      "num_input_tokens_seen": 162638220,
      "step": 7520,
      "time_per_iteration": 3.266892910003662
    },
    {
      "auxiliary_loss_clip": 0.01148134,
      "auxiliary_loss_mlp": 0.01025196,
      "balance_loss_clip": 1.04645658,
      "balance_loss_mlp": 1.01859772,
      "epoch": 0.9043467804966031,
      "flos": 24790428961920.0,
      "grad_norm": 1.5060776854808053,
      "language_loss": 0.70682383,
      "learning_rate": 9.511409092067424e-08,
      "loss": 0.72855717,
      "num_input_tokens_seen": 162658575,
      "step": 7521,
      "time_per_iteration": 3.2174365520477295
    },
    {
      "auxiliary_loss_clip": 0.01138973,
      "auxiliary_loss_mlp": 0.01025597,
      "balance_loss_clip": 1.0475347,
      "balance_loss_mlp": 1.01856637,
      "epoch": 0.9044670233872423,
      "flos": 22632125472000.0,
      "grad_norm": 1.8186389945338424,
      "language_loss": 0.67198241,
      "learning_rate": 9.487686954863327e-08,
      "loss": 0.69362807,
      "num_input_tokens_seen": 162678295,
      "step": 7522,
      "time_per_iteration": 2.5293962955474854
    },
    {
      "auxiliary_loss_clip": 0.0114983,
      "auxiliary_loss_mlp": 0.01031605,
      "balance_loss_clip": 1.04782748,
      "balance_loss_mlp": 1.02510488,
      "epoch": 0.9045872662778813,
      "flos": 23771320289280.0,
      "grad_norm": 1.8205543415507535,
      "language_loss": 0.7695173,
      "learning_rate": 9.46399371822566e-08,
      "loss": 0.79133153,
      "num_input_tokens_seen": 162698070,
      "step": 7523,
      "time_per_iteration": 2.5196595191955566
    },
    {
      "auxiliary_loss_clip": 0.01163308,
      "auxiliary_loss_mlp": 0.01027828,
      "balance_loss_clip": 1.04664922,
      "balance_loss_mlp": 1.02060938,
      "epoch": 0.9047075091685204,
      "flos": 15191039998080.0,
      "grad_norm": 2.229582675228149,
      "language_loss": 0.72821891,
      "learning_rate": 9.440329385748657e-08,
      "loss": 0.7501303,
      "num_input_tokens_seen": 162715140,
      "step": 7524,
      "time_per_iteration": 2.434082269668579
    },
    {
      "auxiliary_loss_clip": 0.01123108,
      "auxiliary_loss_mlp": 0.01018678,
      "balance_loss_clip": 1.04491329,
      "balance_loss_mlp": 1.01294672,
      "epoch": 0.9048277520591596,
      "flos": 18003707504640.0,
      "grad_norm": 1.840842167576267,
      "language_loss": 0.70582676,
      "learning_rate": 9.416693961022137e-08,
      "loss": 0.72724462,
      "num_input_tokens_seen": 162733390,
      "step": 7525,
      "time_per_iteration": 2.5202794075012207
    },
    {
      "auxiliary_loss_clip": 0.01083129,
      "auxiliary_loss_mlp": 0.01024806,
      "balance_loss_clip": 1.03909945,
      "balance_loss_mlp": 1.0178982,
      "epoch": 0.9049479949497986,
      "flos": 21872471713920.0,
      "grad_norm": 1.7449261672084901,
      "language_loss": 0.77196246,
      "learning_rate": 9.393087447631654e-08,
      "loss": 0.79304183,
      "num_input_tokens_seen": 162751670,
      "step": 7526,
      "time_per_iteration": 2.636296510696411
    },
    {
      "auxiliary_loss_clip": 0.01132883,
      "auxiliary_loss_mlp": 0.01025887,
      "balance_loss_clip": 1.04326558,
      "balance_loss_mlp": 1.01952469,
      "epoch": 0.9050682378404377,
      "flos": 20773928113920.0,
      "grad_norm": 2.1402821765207625,
      "language_loss": 0.72586107,
      "learning_rate": 9.36950984915823e-08,
      "loss": 0.7474488,
      "num_input_tokens_seen": 162770025,
      "step": 7527,
      "time_per_iteration": 2.523245334625244
    },
    {
      "auxiliary_loss_clip": 0.01165154,
      "auxiliary_loss_mlp": 0.0102427,
      "balance_loss_clip": 1.04834485,
      "balance_loss_mlp": 1.01737952,
      "epoch": 0.9051884807310768,
      "flos": 21580015178880.0,
      "grad_norm": 1.7336743434797648,
      "language_loss": 0.69104588,
      "learning_rate": 9.345961169178607e-08,
      "loss": 0.7129401,
      "num_input_tokens_seen": 162789710,
      "step": 7528,
      "time_per_iteration": 2.478794574737549
    },
    {
      "auxiliary_loss_clip": 0.01107854,
      "auxiliary_loss_mlp": 0.01026674,
      "balance_loss_clip": 1.04412365,
      "balance_loss_mlp": 1.01927733,
      "epoch": 0.9053087236217159,
      "flos": 21908059113600.0,
      "grad_norm": 1.423533797434014,
      "language_loss": 0.72768432,
      "learning_rate": 9.322441411265081e-08,
      "loss": 0.74902958,
      "num_input_tokens_seen": 162810695,
      "step": 7529,
      "time_per_iteration": 2.5558722019195557
    },
    {
      "auxiliary_loss_clip": 0.01130476,
      "auxiliary_loss_mlp": 0.0102897,
      "balance_loss_clip": 1.04458952,
      "balance_loss_mlp": 1.02193379,
      "epoch": 0.9054289665123549,
      "flos": 17055809544960.0,
      "grad_norm": 3.1735858840069917,
      "language_loss": 0.73347962,
      "learning_rate": 9.298950578985554e-08,
      "loss": 0.75507408,
      "num_input_tokens_seen": 162827770,
      "step": 7530,
      "time_per_iteration": 2.506119966506958
    },
    {
      "auxiliary_loss_clip": 0.01146269,
      "auxiliary_loss_mlp": 0.0076101,
      "balance_loss_clip": 1.04767561,
      "balance_loss_mlp": 1.00027633,
      "epoch": 0.905549209402994,
      "flos": 20777268078720.0,
      "grad_norm": 1.6220055736747767,
      "language_loss": 0.70979416,
      "learning_rate": 9.275488675903665e-08,
      "loss": 0.72886693,
      "num_input_tokens_seen": 162846715,
      "step": 7531,
      "time_per_iteration": 2.483637571334839
    },
    {
      "auxiliary_loss_clip": 0.01101013,
      "auxiliary_loss_mlp": 0.01026166,
      "balance_loss_clip": 1.04042864,
      "balance_loss_mlp": 1.01949286,
      "epoch": 0.9056694522936332,
      "flos": 21686813291520.0,
      "grad_norm": 2.3981714014860085,
      "language_loss": 0.73598546,
      "learning_rate": 9.252055705578454e-08,
      "loss": 0.75725722,
      "num_input_tokens_seen": 162866215,
      "step": 7532,
      "time_per_iteration": 2.6007566452026367
    },
    {
      "auxiliary_loss_clip": 0.01149821,
      "auxiliary_loss_mlp": 0.01023329,
      "balance_loss_clip": 1.0455687,
      "balance_loss_mlp": 1.01702881,
      "epoch": 0.9057896951842722,
      "flos": 29569133433600.0,
      "grad_norm": 1.606993608403782,
      "language_loss": 0.72470391,
      "learning_rate": 9.228651671564747e-08,
      "loss": 0.7464354,
      "num_input_tokens_seen": 162888245,
      "step": 7533,
      "time_per_iteration": 3.3082590103149414
    },
    {
      "auxiliary_loss_clip": 0.01105317,
      "auxiliary_loss_mlp": 0.01035284,
      "balance_loss_clip": 1.04433846,
      "balance_loss_mlp": 1.02868271,
      "epoch": 0.9059099380749113,
      "flos": 27892248952320.0,
      "grad_norm": 1.5158012277626711,
      "language_loss": 0.77756178,
      "learning_rate": 9.205276577412901e-08,
      "loss": 0.79896772,
      "num_input_tokens_seen": 162911025,
      "step": 7534,
      "time_per_iteration": 2.639434814453125
    },
    {
      "auxiliary_loss_clip": 0.0114068,
      "auxiliary_loss_mlp": 0.0076091,
      "balance_loss_clip": 1.04350185,
      "balance_loss_mlp": 1.00029182,
      "epoch": 0.9060301809655504,
      "flos": 17748993185280.0,
      "grad_norm": 2.407995880019666,
      "language_loss": 0.76448059,
      "learning_rate": 9.181930426668905e-08,
      "loss": 0.78349644,
      "num_input_tokens_seen": 162927820,
      "step": 7535,
      "time_per_iteration": 2.494900703430176
    },
    {
      "auxiliary_loss_clip": 0.01096602,
      "auxiliary_loss_mlp": 0.01029124,
      "balance_loss_clip": 1.03869987,
      "balance_loss_mlp": 1.02246869,
      "epoch": 0.9061504238561895,
      "flos": 31759432963200.0,
      "grad_norm": 1.5263092272394707,
      "language_loss": 0.67867208,
      "learning_rate": 9.158613222874346e-08,
      "loss": 0.69992936,
      "num_input_tokens_seen": 162949445,
      "step": 7536,
      "time_per_iteration": 2.65517258644104
    },
    {
      "auxiliary_loss_clip": 0.01134373,
      "auxiliary_loss_mlp": 0.01025108,
      "balance_loss_clip": 1.04233241,
      "balance_loss_mlp": 1.0181849,
      "epoch": 0.9062706667468285,
      "flos": 20048066075520.0,
      "grad_norm": 1.446165840771521,
      "language_loss": 0.81793916,
      "learning_rate": 9.135324969566394e-08,
      "loss": 0.83953393,
      "num_input_tokens_seen": 162968945,
      "step": 7537,
      "time_per_iteration": 2.5262868404388428
    },
    {
      "auxiliary_loss_clip": 0.01156796,
      "auxiliary_loss_mlp": 0.01029442,
      "balance_loss_clip": 1.04909849,
      "balance_loss_mlp": 1.02260256,
      "epoch": 0.9063909096374677,
      "flos": 18437292576000.0,
      "grad_norm": 1.823513552141641,
      "language_loss": 0.75465423,
      "learning_rate": 9.112065670277913e-08,
      "loss": 0.77651656,
      "num_input_tokens_seen": 162985310,
      "step": 7538,
      "time_per_iteration": 2.4454779624938965
    },
    {
      "auxiliary_loss_clip": 0.01130129,
      "auxiliary_loss_mlp": 0.01029316,
      "balance_loss_clip": 1.04104197,
      "balance_loss_mlp": 1.02326334,
      "epoch": 0.9065111525281068,
      "flos": 33547353361920.0,
      "grad_norm": 2.327920122028646,
      "language_loss": 0.72749496,
      "learning_rate": 9.088835328537303e-08,
      "loss": 0.74908936,
      "num_input_tokens_seen": 163006900,
      "step": 7539,
      "time_per_iteration": 2.621633768081665
    },
    {
      "auxiliary_loss_clip": 0.01138969,
      "auxiliary_loss_mlp": 0.01028803,
      "balance_loss_clip": 1.04578805,
      "balance_loss_mlp": 1.02209103,
      "epoch": 0.9066313954187458,
      "flos": 23367863750400.0,
      "grad_norm": 4.072086667351616,
      "language_loss": 0.71408296,
      "learning_rate": 9.065633947868568e-08,
      "loss": 0.73576069,
      "num_input_tokens_seen": 163026505,
      "step": 7540,
      "time_per_iteration": 2.535513162612915
    },
    {
      "auxiliary_loss_clip": 0.01121636,
      "auxiliary_loss_mlp": 0.0076052,
      "balance_loss_clip": 1.04599166,
      "balance_loss_mlp": 1.0003221,
      "epoch": 0.906751638309385,
      "flos": 26249623067520.0,
      "grad_norm": 2.282306810981983,
      "language_loss": 0.80220795,
      "learning_rate": 9.042461531791379e-08,
      "loss": 0.82102948,
      "num_input_tokens_seen": 163044925,
      "step": 7541,
      "time_per_iteration": 2.593374013900757
    },
    {
      "auxiliary_loss_clip": 0.01161202,
      "auxiliary_loss_mlp": 0.01021919,
      "balance_loss_clip": 1.04690135,
      "balance_loss_mlp": 1.01529741,
      "epoch": 0.906871881200024,
      "flos": 16544477485440.0,
      "grad_norm": 1.975492595357323,
      "language_loss": 0.77750731,
      "learning_rate": 9.019318083820903e-08,
      "loss": 0.79933858,
      "num_input_tokens_seen": 163063505,
      "step": 7542,
      "time_per_iteration": 2.436610221862793
    },
    {
      "auxiliary_loss_clip": 0.01147791,
      "auxiliary_loss_mlp": 0.01027463,
      "balance_loss_clip": 1.04545748,
      "balance_loss_mlp": 1.0201019,
      "epoch": 0.9069921240906631,
      "flos": 24605129675520.0,
      "grad_norm": 1.5727841745463118,
      "language_loss": 0.85113698,
      "learning_rate": 8.996203607468045e-08,
      "loss": 0.87288958,
      "num_input_tokens_seen": 163082505,
      "step": 7543,
      "time_per_iteration": 2.534026622772217
    },
    {
      "auxiliary_loss_clip": 0.01144078,
      "auxiliary_loss_mlp": 0.01023249,
      "balance_loss_clip": 1.0432986,
      "balance_loss_mlp": 1.01621902,
      "epoch": 0.9071123669813023,
      "flos": 25374731500800.0,
      "grad_norm": 1.4193650147364238,
      "language_loss": 0.75306237,
      "learning_rate": 8.973118106239241e-08,
      "loss": 0.77473569,
      "num_input_tokens_seen": 163105110,
      "step": 7544,
      "time_per_iteration": 3.2642009258270264
    },
    {
      "auxiliary_loss_clip": 0.01090416,
      "auxiliary_loss_mlp": 0.01024592,
      "balance_loss_clip": 1.03621578,
      "balance_loss_mlp": 1.01818156,
      "epoch": 0.9072326098719413,
      "flos": 26725798690560.0,
      "grad_norm": 2.1578681035509804,
      "language_loss": 0.9473592,
      "learning_rate": 8.95006158363656e-08,
      "loss": 0.96850926,
      "num_input_tokens_seen": 163125295,
      "step": 7545,
      "time_per_iteration": 2.6673049926757812
    },
    {
      "auxiliary_loss_clip": 0.01149132,
      "auxiliary_loss_mlp": 0.01027212,
      "balance_loss_clip": 1.04726994,
      "balance_loss_mlp": 1.02003872,
      "epoch": 0.9073528527625804,
      "flos": 23878800760320.0,
      "grad_norm": 1.7200648792727475,
      "language_loss": 0.77602291,
      "learning_rate": 8.9270340431576e-08,
      "loss": 0.79778636,
      "num_input_tokens_seen": 163144385,
      "step": 7546,
      "time_per_iteration": 3.2761638164520264
    },
    {
      "auxiliary_loss_clip": 0.01150345,
      "auxiliary_loss_mlp": 0.01027081,
      "balance_loss_clip": 1.04522121,
      "balance_loss_mlp": 1.02053618,
      "epoch": 0.9074730956532195,
      "flos": 37852144767360.0,
      "grad_norm": 2.510602769705035,
      "language_loss": 0.73221153,
      "learning_rate": 8.904035488295658e-08,
      "loss": 0.75398576,
      "num_input_tokens_seen": 163163885,
      "step": 7547,
      "time_per_iteration": 2.6206932067871094
    },
    {
      "auxiliary_loss_clip": 0.01045243,
      "auxiliary_loss_mlp": 0.00751252,
      "balance_loss_clip": 1.00834501,
      "balance_loss_mlp": 1.00020397,
      "epoch": 0.9075933385438586,
      "flos": 65173307385600.0,
      "grad_norm": 0.7852770751793497,
      "language_loss": 0.53266501,
      "learning_rate": 8.881065922539632e-08,
      "loss": 0.55062997,
      "num_input_tokens_seen": 163224325,
      "step": 7548,
      "time_per_iteration": 3.6873855590820312
    },
    {
      "auxiliary_loss_clip": 0.01117293,
      "auxiliary_loss_mlp": 0.01019854,
      "balance_loss_clip": 1.04580522,
      "balance_loss_mlp": 1.01326776,
      "epoch": 0.9077135814344977,
      "flos": 19931571290880.0,
      "grad_norm": 1.5490261587677663,
      "language_loss": 0.73296881,
      "learning_rate": 8.85812534937389e-08,
      "loss": 0.75434029,
      "num_input_tokens_seen": 163242425,
      "step": 7549,
      "time_per_iteration": 2.5423717498779297
    },
    {
      "auxiliary_loss_clip": 0.01151272,
      "auxiliary_loss_mlp": 0.01029209,
      "balance_loss_clip": 1.04697967,
      "balance_loss_mlp": 1.02211297,
      "epoch": 0.9078338243251368,
      "flos": 17529650784000.0,
      "grad_norm": 3.0055969639560853,
      "language_loss": 0.67277849,
      "learning_rate": 8.835213772278583e-08,
      "loss": 0.6945833,
      "num_input_tokens_seen": 163259280,
      "step": 7550,
      "time_per_iteration": 2.44325590133667
    },
    {
      "auxiliary_loss_clip": 0.01111639,
      "auxiliary_loss_mlp": 0.01024116,
      "balance_loss_clip": 1.04295588,
      "balance_loss_mlp": 1.0177834,
      "epoch": 0.9079540672157759,
      "flos": 28803410277120.0,
      "grad_norm": 1.6366405887197504,
      "language_loss": 0.7880832,
      "learning_rate": 8.812331194729373e-08,
      "loss": 0.80944073,
      "num_input_tokens_seen": 163278925,
      "step": 7551,
      "time_per_iteration": 2.6109108924865723
    },
    {
      "auxiliary_loss_clip": 0.01169213,
      "auxiliary_loss_mlp": 0.01026976,
      "balance_loss_clip": 1.0513829,
      "balance_loss_mlp": 1.01949823,
      "epoch": 0.9080743101064149,
      "flos": 23513840622720.0,
      "grad_norm": 2.611961233924004,
      "language_loss": 0.72062993,
      "learning_rate": 8.789477620197461e-08,
      "loss": 0.74259186,
      "num_input_tokens_seen": 163298450,
      "step": 7552,
      "time_per_iteration": 2.4790303707122803
    },
    {
      "auxiliary_loss_clip": 0.01134125,
      "auxiliary_loss_mlp": 0.01026913,
      "balance_loss_clip": 1.0432651,
      "balance_loss_mlp": 1.01968288,
      "epoch": 0.9081945529970541,
      "flos": 22778102344320.0,
      "grad_norm": 2.3104927277797547,
      "language_loss": 0.78947598,
      "learning_rate": 8.766653052149831e-08,
      "loss": 0.81108636,
      "num_input_tokens_seen": 163313635,
      "step": 7553,
      "time_per_iteration": 2.4895284175872803
    },
    {
      "auxiliary_loss_clip": 0.01134531,
      "auxiliary_loss_mlp": 0.01028529,
      "balance_loss_clip": 1.04576635,
      "balance_loss_mlp": 1.02182019,
      "epoch": 0.9083147958876931,
      "flos": 18873714821760.0,
      "grad_norm": 1.9361150589004221,
      "language_loss": 0.74410826,
      "learning_rate": 8.743857494048823e-08,
      "loss": 0.76573884,
      "num_input_tokens_seen": 163330450,
      "step": 7554,
      "time_per_iteration": 2.501981735229492
    },
    {
      "auxiliary_loss_clip": 0.01120887,
      "auxiliary_loss_mlp": 0.01025541,
      "balance_loss_clip": 1.04382849,
      "balance_loss_mlp": 1.01859975,
      "epoch": 0.9084350387783322,
      "flos": 18909374048640.0,
      "grad_norm": 2.0410445202792586,
      "language_loss": 0.62748158,
      "learning_rate": 8.721090949352605e-08,
      "loss": 0.64894581,
      "num_input_tokens_seen": 163346690,
      "step": 7555,
      "time_per_iteration": 2.526177167892456
    },
    {
      "auxiliary_loss_clip": 0.01154262,
      "auxiliary_loss_mlp": 0.01026022,
      "balance_loss_clip": 1.04744649,
      "balance_loss_mlp": 1.01896501,
      "epoch": 0.9085552816689714,
      "flos": 20595488325120.0,
      "grad_norm": 2.0122490505668376,
      "language_loss": 0.72797829,
      "learning_rate": 8.698353421514793e-08,
      "loss": 0.74978113,
      "num_input_tokens_seen": 163365065,
      "step": 7556,
      "time_per_iteration": 2.4722578525543213
    },
    {
      "auxiliary_loss_clip": 0.01148186,
      "auxiliary_loss_mlp": 0.01025716,
      "balance_loss_clip": 1.04579067,
      "balance_loss_mlp": 1.01889396,
      "epoch": 0.9086755245596104,
      "flos": 18113163223680.0,
      "grad_norm": 2.171532126397396,
      "language_loss": 0.80733806,
      "learning_rate": 8.67564491398467e-08,
      "loss": 0.82907706,
      "num_input_tokens_seen": 163382070,
      "step": 7557,
      "time_per_iteration": 2.4457597732543945
    },
    {
      "auxiliary_loss_clip": 0.01151811,
      "auxiliary_loss_mlp": 0.01028934,
      "balance_loss_clip": 1.04480004,
      "balance_loss_mlp": 1.02179325,
      "epoch": 0.9087957674502495,
      "flos": 19129793857920.0,
      "grad_norm": 2.3899239024787176,
      "language_loss": 0.73316777,
      "learning_rate": 8.652965430207104e-08,
      "loss": 0.7549752,
      "num_input_tokens_seen": 163399975,
      "step": 7558,
      "time_per_iteration": 2.475050926208496
    },
    {
      "auxiliary_loss_clip": 0.01151165,
      "auxiliary_loss_mlp": 0.01025351,
      "balance_loss_clip": 1.04496205,
      "balance_loss_mlp": 1.01823425,
      "epoch": 0.9089160103408886,
      "flos": 18109930999680.0,
      "grad_norm": 1.8815850001306704,
      "language_loss": 0.65349221,
      "learning_rate": 8.630314973622521e-08,
      "loss": 0.67525738,
      "num_input_tokens_seen": 163417520,
      "step": 7559,
      "time_per_iteration": 3.26001238822937
    },
    {
      "auxiliary_loss_clip": 0.01147953,
      "auxiliary_loss_mlp": 0.01028371,
      "balance_loss_clip": 1.04617262,
      "balance_loss_mlp": 1.02212417,
      "epoch": 0.9090362532315277,
      "flos": 33364855336320.0,
      "grad_norm": 2.026271983969225,
      "language_loss": 0.70879865,
      "learning_rate": 8.607693547666995e-08,
      "loss": 0.73056185,
      "num_input_tokens_seen": 163440060,
      "step": 7560,
      "time_per_iteration": 2.586787462234497
    },
    {
      "auxiliary_loss_clip": 0.01025958,
      "auxiliary_loss_mlp": 0.01002715,
      "balance_loss_clip": 1.00737965,
      "balance_loss_mlp": 1.00158262,
      "epoch": 0.9091564961221668,
      "flos": 71480585082240.0,
      "grad_norm": 0.8726116464099366,
      "language_loss": 0.57981509,
      "learning_rate": 8.585101155772201e-08,
      "loss": 0.60010183,
      "num_input_tokens_seen": 163502180,
      "step": 7561,
      "time_per_iteration": 3.21933650970459
    },
    {
      "auxiliary_loss_clip": 0.01125591,
      "auxiliary_loss_mlp": 0.01025783,
      "balance_loss_clip": 1.0395143,
      "balance_loss_mlp": 1.01854062,
      "epoch": 0.9092767390128058,
      "flos": 24712574232960.0,
      "grad_norm": 1.778374627565462,
      "language_loss": 0.68543702,
      "learning_rate": 8.562537801365377e-08,
      "loss": 0.70695078,
      "num_input_tokens_seen": 163521915,
      "step": 7562,
      "time_per_iteration": 2.546579360961914
    },
    {
      "auxiliary_loss_clip": 0.01162379,
      "auxiliary_loss_mlp": 0.01027932,
      "balance_loss_clip": 1.0461601,
      "balance_loss_mlp": 1.01995742,
      "epoch": 0.909396981903445,
      "flos": 23586487879680.0,
      "grad_norm": 1.9257046278900358,
      "language_loss": 0.69977242,
      "learning_rate": 8.540003487869362e-08,
      "loss": 0.72167552,
      "num_input_tokens_seen": 163543585,
      "step": 7563,
      "time_per_iteration": 2.484055280685425
    },
    {
      "auxiliary_loss_clip": 0.01112529,
      "auxiliary_loss_mlp": 0.010251,
      "balance_loss_clip": 1.04301,
      "balance_loss_mlp": 1.01788497,
      "epoch": 0.909517224794084,
      "flos": 23404169422080.0,
      "grad_norm": 1.844099305913209,
      "language_loss": 0.79580057,
      "learning_rate": 8.517498218702557e-08,
      "loss": 0.81717682,
      "num_input_tokens_seen": 163561515,
      "step": 7564,
      "time_per_iteration": 2.5626189708709717
    },
    {
      "auxiliary_loss_clip": 0.01117337,
      "auxiliary_loss_mlp": 0.01029199,
      "balance_loss_clip": 1.04170656,
      "balance_loss_mlp": 1.02232933,
      "epoch": 0.9096374676847231,
      "flos": 19208618254080.0,
      "grad_norm": 1.6948992639755918,
      "language_loss": 0.69345891,
      "learning_rate": 8.49502199727905e-08,
      "loss": 0.71492434,
      "num_input_tokens_seen": 163579540,
      "step": 7565,
      "time_per_iteration": 2.542144775390625
    },
    {
      "auxiliary_loss_clip": 0.01145289,
      "auxiliary_loss_mlp": 0.01027902,
      "balance_loss_clip": 1.04350722,
      "balance_loss_mlp": 1.02067232,
      "epoch": 0.9097577105753623,
      "flos": 33292495388160.0,
      "grad_norm": 5.589564210642016,
      "language_loss": 0.66102481,
      "learning_rate": 8.472574827008428e-08,
      "loss": 0.68275678,
      "num_input_tokens_seen": 163600425,
      "step": 7566,
      "time_per_iteration": 2.6063830852508545
    },
    {
      "auxiliary_loss_clip": 0.01148016,
      "auxiliary_loss_mlp": 0.0102421,
      "balance_loss_clip": 1.04351974,
      "balance_loss_mlp": 1.01717114,
      "epoch": 0.9098779534660013,
      "flos": 21906443001600.0,
      "grad_norm": 1.6929746349108834,
      "language_loss": 0.83766961,
      "learning_rate": 8.450156711295942e-08,
      "loss": 0.85939181,
      "num_input_tokens_seen": 163620595,
      "step": 7567,
      "time_per_iteration": 2.497229814529419
    },
    {
      "auxiliary_loss_clip": 0.01133517,
      "auxiliary_loss_mlp": 0.01034438,
      "balance_loss_clip": 1.04612648,
      "balance_loss_mlp": 1.02771449,
      "epoch": 0.9099981963566404,
      "flos": 25730354102400.0,
      "grad_norm": 2.1338451038583965,
      "language_loss": 0.86242801,
      "learning_rate": 8.427767653542383e-08,
      "loss": 0.88410759,
      "num_input_tokens_seen": 163635765,
      "step": 7568,
      "time_per_iteration": 2.5368642807006836
    },
    {
      "auxiliary_loss_clip": 0.01103682,
      "auxiliary_loss_mlp": 0.01026653,
      "balance_loss_clip": 1.03922594,
      "balance_loss_mlp": 1.02015591,
      "epoch": 0.9101184392472795,
      "flos": 21069437304960.0,
      "grad_norm": 1.8211968564229994,
      "language_loss": 0.70042145,
      "learning_rate": 8.405407657144125e-08,
      "loss": 0.72172481,
      "num_input_tokens_seen": 163654925,
      "step": 7569,
      "time_per_iteration": 2.6055727005004883
    },
    {
      "auxiliary_loss_clip": 0.01128917,
      "auxiliary_loss_mlp": 0.01024834,
      "balance_loss_clip": 1.04210973,
      "balance_loss_mlp": 1.01840305,
      "epoch": 0.9102386821379186,
      "flos": 24752614919040.0,
      "grad_norm": 1.7920105393130261,
      "language_loss": 0.72402555,
      "learning_rate": 8.383076725493232e-08,
      "loss": 0.74556309,
      "num_input_tokens_seen": 163672245,
      "step": 7570,
      "time_per_iteration": 3.356362819671631
    },
    {
      "auxiliary_loss_clip": 0.01150461,
      "auxiliary_loss_mlp": 0.01030769,
      "balance_loss_clip": 1.04566216,
      "balance_loss_mlp": 1.02394724,
      "epoch": 0.9103589250285576,
      "flos": 22562818179840.0,
      "grad_norm": 2.438517523463585,
      "language_loss": 0.67788219,
      "learning_rate": 8.360774861977216e-08,
      "loss": 0.69969445,
      "num_input_tokens_seen": 163691365,
      "step": 7571,
      "time_per_iteration": 2.515854835510254
    },
    {
      "auxiliary_loss_clip": 0.01133068,
      "auxiliary_loss_mlp": 0.01022958,
      "balance_loss_clip": 1.04054654,
      "balance_loss_mlp": 1.01591611,
      "epoch": 0.9104791679191968,
      "flos": 25373474524800.0,
      "grad_norm": 2.6053419148419805,
      "language_loss": 0.74603403,
      "learning_rate": 8.338502069979281e-08,
      "loss": 0.76759422,
      "num_input_tokens_seen": 163711675,
      "step": 7572,
      "time_per_iteration": 3.3281149864196777
    },
    {
      "auxiliary_loss_clip": 0.01148677,
      "auxiliary_loss_mlp": 0.01027907,
      "balance_loss_clip": 1.04203296,
      "balance_loss_mlp": 1.02125812,
      "epoch": 0.9105994108098359,
      "flos": 14426681558400.0,
      "grad_norm": 2.7164739864052287,
      "language_loss": 0.79584217,
      "learning_rate": 8.316258352878214e-08,
      "loss": 0.817608,
      "num_input_tokens_seen": 163728095,
      "step": 7573,
      "time_per_iteration": 3.200885534286499
    },
    {
      "auxiliary_loss_clip": 0.01151119,
      "auxiliary_loss_mlp": 0.01027292,
      "balance_loss_clip": 1.04419303,
      "balance_loss_mlp": 1.02024376,
      "epoch": 0.9107196537004749,
      "flos": 26718292748160.0,
      "grad_norm": 2.131237362347271,
      "language_loss": 0.71113896,
      "learning_rate": 8.294043714048338e-08,
      "loss": 0.73292309,
      "num_input_tokens_seen": 163747175,
      "step": 7574,
      "time_per_iteration": 2.5337460041046143
    },
    {
      "auxiliary_loss_clip": 0.01035554,
      "auxiliary_loss_mlp": 0.01002711,
      "balance_loss_clip": 1.00780129,
      "balance_loss_mlp": 1.00164366,
      "epoch": 0.9108398965911141,
      "flos": 66532634703360.0,
      "grad_norm": 0.7484370302359629,
      "language_loss": 0.60466337,
      "learning_rate": 8.271858156859624e-08,
      "loss": 0.62504601,
      "num_input_tokens_seen": 163812545,
      "step": 7575,
      "time_per_iteration": 3.1795601844787598
    },
    {
      "auxiliary_loss_clip": 0.01161013,
      "auxiliary_loss_mlp": 0.01026157,
      "balance_loss_clip": 1.04716873,
      "balance_loss_mlp": 1.01930821,
      "epoch": 0.9109601394817531,
      "flos": 25411073086080.0,
      "grad_norm": 1.6803964587103293,
      "language_loss": 0.73678803,
      "learning_rate": 8.249701684677557e-08,
      "loss": 0.75865972,
      "num_input_tokens_seen": 163833870,
      "step": 7576,
      "time_per_iteration": 2.514474868774414
    },
    {
      "auxiliary_loss_clip": 0.01151259,
      "auxiliary_loss_mlp": 0.01021193,
      "balance_loss_clip": 1.04952598,
      "balance_loss_mlp": 1.01452661,
      "epoch": 0.9110803823723922,
      "flos": 22747794243840.0,
      "grad_norm": 1.876702403247439,
      "language_loss": 0.80815327,
      "learning_rate": 8.227574300863294e-08,
      "loss": 0.82987779,
      "num_input_tokens_seen": 163854040,
      "step": 7577,
      "time_per_iteration": 2.5656697750091553
    },
    {
      "auxiliary_loss_clip": 0.01142499,
      "auxiliary_loss_mlp": 0.01025657,
      "balance_loss_clip": 1.04783595,
      "balance_loss_mlp": 1.01811075,
      "epoch": 0.9112006252630314,
      "flos": 48469924131840.0,
      "grad_norm": 1.9086796943853797,
      "language_loss": 0.69513214,
      "learning_rate": 8.205476008773548e-08,
      "loss": 0.71681362,
      "num_input_tokens_seen": 163878040,
      "step": 7578,
      "time_per_iteration": 2.7710695266723633
    },
    {
      "auxiliary_loss_clip": 0.01117324,
      "auxiliary_loss_mlp": 0.01027804,
      "balance_loss_clip": 1.04531395,
      "balance_loss_mlp": 1.02112579,
      "epoch": 0.9113208681536704,
      "flos": 30009649829760.0,
      "grad_norm": 1.922384569197043,
      "language_loss": 0.82464808,
      "learning_rate": 8.183406811760596e-08,
      "loss": 0.84609938,
      "num_input_tokens_seen": 163897770,
      "step": 7579,
      "time_per_iteration": 2.6167683601379395
    },
    {
      "auxiliary_loss_clip": 0.01108468,
      "auxiliary_loss_mlp": 0.01026596,
      "balance_loss_clip": 1.04010248,
      "balance_loss_mlp": 1.0198245,
      "epoch": 0.9114411110443095,
      "flos": 25594971742080.0,
      "grad_norm": 1.5082556565134873,
      "language_loss": 0.74320632,
      "learning_rate": 8.161366713172313e-08,
      "loss": 0.76455694,
      "num_input_tokens_seen": 163920160,
      "step": 7580,
      "time_per_iteration": 2.612069606781006
    },
    {
      "auxiliary_loss_clip": 0.01124495,
      "auxiliary_loss_mlp": 0.01026737,
      "balance_loss_clip": 1.04134321,
      "balance_loss_mlp": 1.01938796,
      "epoch": 0.9115613539349486,
      "flos": 18399729928320.0,
      "grad_norm": 4.163482342106094,
      "language_loss": 0.83741069,
      "learning_rate": 8.139355716352137e-08,
      "loss": 0.85892296,
      "num_input_tokens_seen": 163935000,
      "step": 7581,
      "time_per_iteration": 2.508976459503174
    },
    {
      "auxiliary_loss_clip": 0.01132288,
      "auxiliary_loss_mlp": 0.01020814,
      "balance_loss_clip": 1.04162025,
      "balance_loss_mlp": 1.0139451,
      "epoch": 0.9116815968255877,
      "flos": 21726171619200.0,
      "grad_norm": 1.5723226242354638,
      "language_loss": 0.69902802,
      "learning_rate": 8.117373824639196e-08,
      "loss": 0.720559,
      "num_input_tokens_seen": 163955265,
      "step": 7582,
      "time_per_iteration": 2.5303585529327393
    },
    {
      "auxiliary_loss_clip": 0.01054885,
      "auxiliary_loss_mlp": 0.01002451,
      "balance_loss_clip": 1.00805807,
      "balance_loss_mlp": 1.00134873,
      "epoch": 0.9118018397162267,
      "flos": 65363526835200.0,
      "grad_norm": 0.7196073539075255,
      "language_loss": 0.59311998,
      "learning_rate": 8.095421041368067e-08,
      "loss": 0.61369324,
      "num_input_tokens_seen": 164014680,
      "step": 7583,
      "time_per_iteration": 2.9779539108276367
    },
    {
      "auxiliary_loss_clip": 0.01134467,
      "auxiliary_loss_mlp": 0.00760285,
      "balance_loss_clip": 1.04592323,
      "balance_loss_mlp": 1.00033593,
      "epoch": 0.9119220826068659,
      "flos": 20922885815040.0,
      "grad_norm": 2.4930885094509123,
      "language_loss": 0.70695943,
      "learning_rate": 8.073497369868999e-08,
      "loss": 0.72590697,
      "num_input_tokens_seen": 164033140,
      "step": 7584,
      "time_per_iteration": 2.5402300357818604
    },
    {
      "auxiliary_loss_clip": 0.01140915,
      "auxiliary_loss_mlp": 0.01023157,
      "balance_loss_clip": 1.04343152,
      "balance_loss_mlp": 1.01683617,
      "epoch": 0.912042325497505,
      "flos": 28366449327360.0,
      "grad_norm": 1.612863218258099,
      "language_loss": 0.75454444,
      "learning_rate": 8.051602813467772e-08,
      "loss": 0.77618515,
      "num_input_tokens_seen": 164054995,
      "step": 7585,
      "time_per_iteration": 3.4162771701812744
    },
    {
      "auxiliary_loss_clip": 0.01149014,
      "auxiliary_loss_mlp": 0.01023531,
      "balance_loss_clip": 1.04425538,
      "balance_loss_mlp": 1.01683128,
      "epoch": 0.912162568388144,
      "flos": 17566782468480.0,
      "grad_norm": 1.668319483739733,
      "language_loss": 0.71101958,
      "learning_rate": 8.029737375485756e-08,
      "loss": 0.73274505,
      "num_input_tokens_seen": 164074225,
      "step": 7586,
      "time_per_iteration": 2.477029800415039
    },
    {
      "auxiliary_loss_clip": 0.01162993,
      "auxiliary_loss_mlp": 0.01027529,
      "balance_loss_clip": 1.04717922,
      "balance_loss_mlp": 1.02051342,
      "epoch": 0.9122828112787832,
      "flos": 19827897661440.0,
      "grad_norm": 1.653902586310946,
      "language_loss": 0.72636092,
      "learning_rate": 8.007901059239986e-08,
      "loss": 0.74826616,
      "num_input_tokens_seen": 164093505,
      "step": 7587,
      "time_per_iteration": 2.448899507522583
    },
    {
      "auxiliary_loss_clip": 0.01133115,
      "auxiliary_loss_mlp": 0.01024162,
      "balance_loss_clip": 1.04195976,
      "balance_loss_mlp": 1.01746583,
      "epoch": 0.9124030541694222,
      "flos": 20813789232000.0,
      "grad_norm": 1.5544616259969288,
      "language_loss": 0.80106419,
      "learning_rate": 7.986093868042964e-08,
      "loss": 0.82263696,
      "num_input_tokens_seen": 164113750,
      "step": 7588,
      "time_per_iteration": 2.532444477081299
    },
    {
      "auxiliary_loss_clip": 0.01147864,
      "auxiliary_loss_mlp": 0.01021184,
      "balance_loss_clip": 1.04648232,
      "balance_loss_mlp": 1.01471996,
      "epoch": 0.9125232970600613,
      "flos": 25192305302400.0,
      "grad_norm": 1.674883673837388,
      "language_loss": 0.67329061,
      "learning_rate": 7.964315805202826e-08,
      "loss": 0.6949811,
      "num_input_tokens_seen": 164134330,
      "step": 7589,
      "time_per_iteration": 2.520984172821045
    },
    {
      "auxiliary_loss_clip": 0.01134078,
      "auxiliary_loss_mlp": 0.01025898,
      "balance_loss_clip": 1.0443126,
      "balance_loss_mlp": 1.01915395,
      "epoch": 0.9126435399507005,
      "flos": 19719591177600.0,
      "grad_norm": 1.7308612547128135,
      "language_loss": 0.73333454,
      "learning_rate": 7.942566874023304e-08,
      "loss": 0.75493431,
      "num_input_tokens_seen": 164153515,
      "step": 7590,
      "time_per_iteration": 2.4996166229248047
    },
    {
      "auxiliary_loss_clip": 0.01127169,
      "auxiliary_loss_mlp": 0.01033028,
      "balance_loss_clip": 1.04226947,
      "balance_loss_mlp": 1.02483225,
      "epoch": 0.9127637828413395,
      "flos": 19573614305280.0,
      "grad_norm": 2.0085203203352964,
      "language_loss": 0.6975863,
      "learning_rate": 7.920847077803649e-08,
      "loss": 0.71918827,
      "num_input_tokens_seen": 164171305,
      "step": 7591,
      "time_per_iteration": 2.4885244369506836
    },
    {
      "auxiliary_loss_clip": 0.01092627,
      "auxiliary_loss_mlp": 0.01019204,
      "balance_loss_clip": 1.03519893,
      "balance_loss_mlp": 1.01259983,
      "epoch": 0.9128840257319786,
      "flos": 20230635928320.0,
      "grad_norm": 2.362125182599027,
      "language_loss": 0.81808072,
      "learning_rate": 7.899156419838826e-08,
      "loss": 0.83919901,
      "num_input_tokens_seen": 164190275,
      "step": 7592,
      "time_per_iteration": 2.5745208263397217
    },
    {
      "auxiliary_loss_clip": 0.01118931,
      "auxiliary_loss_mlp": 0.01030498,
      "balance_loss_clip": 1.04184592,
      "balance_loss_mlp": 1.02328312,
      "epoch": 0.9130042686226177,
      "flos": 24858658846080.0,
      "grad_norm": 1.7234350189664545,
      "language_loss": 0.656106,
      "learning_rate": 7.87749490341918e-08,
      "loss": 0.67760026,
      "num_input_tokens_seen": 164210550,
      "step": 7593,
      "time_per_iteration": 2.58345103263855
    },
    {
      "auxiliary_loss_clip": 0.01168036,
      "auxiliary_loss_mlp": 0.01028403,
      "balance_loss_clip": 1.04951072,
      "balance_loss_mlp": 1.02085698,
      "epoch": 0.9131245115132568,
      "flos": 23581747284480.0,
      "grad_norm": 2.0449800844392514,
      "language_loss": 0.83369398,
      "learning_rate": 7.855862531830836e-08,
      "loss": 0.85565835,
      "num_input_tokens_seen": 164226660,
      "step": 7594,
      "time_per_iteration": 2.4445040225982666
    },
    {
      "auxiliary_loss_clip": 0.01148846,
      "auxiliary_loss_mlp": 0.01027627,
      "balance_loss_clip": 1.04569328,
      "balance_loss_mlp": 1.02048695,
      "epoch": 0.9132447544038959,
      "flos": 19931607204480.0,
      "grad_norm": 1.5804590516655308,
      "language_loss": 0.72351074,
      "learning_rate": 7.834259308355373e-08,
      "loss": 0.74527544,
      "num_input_tokens_seen": 164245425,
      "step": 7595,
      "time_per_iteration": 2.4833874702453613
    },
    {
      "auxiliary_loss_clip": 0.0108027,
      "auxiliary_loss_mlp": 0.01023183,
      "balance_loss_clip": 1.03818226,
      "balance_loss_mlp": 1.01564908,
      "epoch": 0.9133649972945349,
      "flos": 21981747864960.0,
      "grad_norm": 2.2696034878088307,
      "language_loss": 0.75150251,
      "learning_rate": 7.812685236269989e-08,
      "loss": 0.77253711,
      "num_input_tokens_seen": 164264085,
      "step": 7596,
      "time_per_iteration": 3.397035837173462
    },
    {
      "auxiliary_loss_clip": 0.01020404,
      "auxiliary_loss_mlp": 0.0100237,
      "balance_loss_clip": 1.00854731,
      "balance_loss_mlp": 1.00130928,
      "epoch": 0.9134852401851741,
      "flos": 71240523511680.0,
      "grad_norm": 0.7844215204050851,
      "language_loss": 0.58630592,
      "learning_rate": 7.791140318847445e-08,
      "loss": 0.60653365,
      "num_input_tokens_seen": 164322220,
      "step": 7597,
      "time_per_iteration": 3.146777391433716
    },
    {
      "auxiliary_loss_clip": 0.01128526,
      "auxiliary_loss_mlp": 0.01029518,
      "balance_loss_clip": 1.04531741,
      "balance_loss_mlp": 1.02323258,
      "epoch": 0.9136054830758131,
      "flos": 23626923615360.0,
      "grad_norm": 1.3714477827516112,
      "language_loss": 0.80482507,
      "learning_rate": 7.769624559356081e-08,
      "loss": 0.82640553,
      "num_input_tokens_seen": 164345615,
      "step": 7598,
      "time_per_iteration": 3.358421564102173
    },
    {
      "auxiliary_loss_clip": 0.01147613,
      "auxiliary_loss_mlp": 0.01029275,
      "balance_loss_clip": 1.04457498,
      "balance_loss_mlp": 1.02186918,
      "epoch": 0.9137257259664522,
      "flos": 23438858981760.0,
      "grad_norm": 3.4236457144369212,
      "language_loss": 0.7511667,
      "learning_rate": 7.748137961059842e-08,
      "loss": 0.77293557,
      "num_input_tokens_seen": 164359595,
      "step": 7599,
      "time_per_iteration": 3.2068989276885986
    },
    {
      "auxiliary_loss_clip": 0.01160585,
      "auxiliary_loss_mlp": 0.01029418,
      "balance_loss_clip": 1.04772818,
      "balance_loss_mlp": 1.02249467,
      "epoch": 0.9138459688570914,
      "flos": 19127854523520.0,
      "grad_norm": 2.4270296233738606,
      "language_loss": 0.65085131,
      "learning_rate": 7.726680527218211e-08,
      "loss": 0.67275131,
      "num_input_tokens_seen": 164376635,
      "step": 7600,
      "time_per_iteration": 2.444509983062744
    },
    {
      "auxiliary_loss_clip": 0.01162236,
      "auxiliary_loss_mlp": 0.01023511,
      "balance_loss_clip": 1.04538989,
      "balance_loss_mlp": 1.01621008,
      "epoch": 0.9139662117477304,
      "flos": 46281240714240.0,
      "grad_norm": 1.6821811758077685,
      "language_loss": 0.75212038,
      "learning_rate": 7.70525226108627e-08,
      "loss": 0.77397788,
      "num_input_tokens_seen": 164400305,
      "step": 7601,
      "time_per_iteration": 2.659435272216797
    },
    {
      "auxiliary_loss_clip": 0.01152963,
      "auxiliary_loss_mlp": 0.01033582,
      "balance_loss_clip": 1.0503062,
      "balance_loss_mlp": 1.02727878,
      "epoch": 0.9140864546383695,
      "flos": 22273198819200.0,
      "grad_norm": 2.3001563851001188,
      "language_loss": 0.80010855,
      "learning_rate": 7.683853165914666e-08,
      "loss": 0.82197392,
      "num_input_tokens_seen": 164418075,
      "step": 7602,
      "time_per_iteration": 2.4902050495147705
    },
    {
      "auxiliary_loss_clip": 0.01109763,
      "auxiliary_loss_mlp": 0.01025391,
      "balance_loss_clip": 1.04364681,
      "balance_loss_mlp": 1.01871502,
      "epoch": 0.9142066975290086,
      "flos": 17530009920000.0,
      "grad_norm": 1.6347512659649743,
      "language_loss": 0.76883543,
      "learning_rate": 7.662483244949602e-08,
      "loss": 0.790187,
      "num_input_tokens_seen": 164435335,
      "step": 7603,
      "time_per_iteration": 2.5740301609039307
    },
    {
      "auxiliary_loss_clip": 0.01115067,
      "auxiliary_loss_mlp": 0.01026546,
      "balance_loss_clip": 1.04444349,
      "balance_loss_mlp": 1.02031398,
      "epoch": 0.9143269404196477,
      "flos": 17712148809600.0,
      "grad_norm": 2.694413738064927,
      "language_loss": 0.80692017,
      "learning_rate": 7.641142501432951e-08,
      "loss": 0.82833636,
      "num_input_tokens_seen": 164451530,
      "step": 7604,
      "time_per_iteration": 2.530841588973999
    },
    {
      "auxiliary_loss_clip": 0.01130488,
      "auxiliary_loss_mlp": 0.0102947,
      "balance_loss_clip": 1.04268813,
      "balance_loss_mlp": 1.02282429,
      "epoch": 0.9144471833102867,
      "flos": 33323414019840.0,
      "grad_norm": 1.6552615819023204,
      "language_loss": 0.73690373,
      "learning_rate": 7.619830938602013e-08,
      "loss": 0.75850326,
      "num_input_tokens_seen": 164472755,
      "step": 7605,
      "time_per_iteration": 2.6221320629119873
    },
    {
      "auxiliary_loss_clip": 0.01145039,
      "auxiliary_loss_mlp": 0.01024009,
      "balance_loss_clip": 1.04577768,
      "balance_loss_mlp": 1.01713085,
      "epoch": 0.9145674262009259,
      "flos": 21068970428160.0,
      "grad_norm": 1.8434438042807417,
      "language_loss": 0.82353497,
      "learning_rate": 7.598548559689777e-08,
      "loss": 0.84522545,
      "num_input_tokens_seen": 164491155,
      "step": 7606,
      "time_per_iteration": 2.4995813369750977
    },
    {
      "auxiliary_loss_clip": 0.0111373,
      "auxiliary_loss_mlp": 0.01023209,
      "balance_loss_clip": 1.04017198,
      "balance_loss_mlp": 1.01641154,
      "epoch": 0.914687669091565,
      "flos": 16800269212800.0,
      "grad_norm": 2.291711731950592,
      "language_loss": 0.80965096,
      "learning_rate": 7.577295367924751e-08,
      "loss": 0.83102036,
      "num_input_tokens_seen": 164507555,
      "step": 7607,
      "time_per_iteration": 2.5469439029693604
    },
    {
      "auxiliary_loss_clip": 0.01142697,
      "auxiliary_loss_mlp": 0.010285,
      "balance_loss_clip": 1.04737306,
      "balance_loss_mlp": 1.02126372,
      "epoch": 0.914807911982204,
      "flos": 25773627012480.0,
      "grad_norm": 2.0029119937614417,
      "language_loss": 0.82331216,
      "learning_rate": 7.556071366531002e-08,
      "loss": 0.84502417,
      "num_input_tokens_seen": 164528525,
      "step": 7608,
      "time_per_iteration": 2.6050970554351807
    },
    {
      "auxiliary_loss_clip": 0.01151521,
      "auxiliary_loss_mlp": 0.0103294,
      "balance_loss_clip": 1.04830337,
      "balance_loss_mlp": 1.02551305,
      "epoch": 0.9149281548728432,
      "flos": 19208043636480.0,
      "grad_norm": 2.528880563200342,
      "language_loss": 0.79028058,
      "learning_rate": 7.53487655872822e-08,
      "loss": 0.81212521,
      "num_input_tokens_seen": 164547695,
      "step": 7609,
      "time_per_iteration": 2.5132739543914795
    },
    {
      "auxiliary_loss_clip": 0.01109561,
      "auxiliary_loss_mlp": 0.01023197,
      "balance_loss_clip": 1.03917956,
      "balance_loss_mlp": 1.01619077,
      "epoch": 0.9150483977634822,
      "flos": 26870554500480.0,
      "grad_norm": 1.781831466687112,
      "language_loss": 0.73945796,
      "learning_rate": 7.513710947731656e-08,
      "loss": 0.76078558,
      "num_input_tokens_seen": 164568905,
      "step": 7610,
      "time_per_iteration": 2.6211729049682617
    },
    {
      "auxiliary_loss_clip": 0.01127431,
      "auxiliary_loss_mlp": 0.01026982,
      "balance_loss_clip": 1.0442096,
      "balance_loss_mlp": 1.02016664,
      "epoch": 0.9151686406541213,
      "flos": 21908956953600.0,
      "grad_norm": 1.8031782647416035,
      "language_loss": 0.85209954,
      "learning_rate": 7.492574536752095e-08,
      "loss": 0.87364364,
      "num_input_tokens_seen": 164588895,
      "step": 7611,
      "time_per_iteration": 3.2480111122131348
    },
    {
      "auxiliary_loss_clip": 0.01146702,
      "auxiliary_loss_mlp": 0.01027465,
      "balance_loss_clip": 1.04769027,
      "balance_loss_mlp": 1.02095938,
      "epoch": 0.9152888835447605,
      "flos": 27308556944640.0,
      "grad_norm": 6.364382044526774,
      "language_loss": 0.77900672,
      "learning_rate": 7.471467328995907e-08,
      "loss": 0.80074847,
      "num_input_tokens_seen": 164607705,
      "step": 7612,
      "time_per_iteration": 2.53262996673584
    },
    {
      "auxiliary_loss_clip": 0.01070618,
      "auxiliary_loss_mlp": 0.01028649,
      "balance_loss_clip": 1.03813684,
      "balance_loss_mlp": 1.02139759,
      "epoch": 0.9154091264353995,
      "flos": 13370728510080.0,
      "grad_norm": 2.347366691655838,
      "language_loss": 0.6079179,
      "learning_rate": 7.450389327665018e-08,
      "loss": 0.62891054,
      "num_input_tokens_seen": 164625540,
      "step": 7613,
      "time_per_iteration": 2.8319075107574463
    },
    {
      "auxiliary_loss_clip": 0.01120756,
      "auxiliary_loss_mlp": 0.01027243,
      "balance_loss_clip": 1.04529476,
      "balance_loss_mlp": 1.01942563,
      "epoch": 0.9155293693260386,
      "flos": 20193037367040.0,
      "grad_norm": 2.7700185991227695,
      "language_loss": 0.68109465,
      "learning_rate": 7.429340535957029e-08,
      "loss": 0.70257461,
      "num_input_tokens_seen": 164640735,
      "step": 7614,
      "time_per_iteration": 2.690930128097534
    },
    {
      "auxiliary_loss_clip": 0.01135555,
      "auxiliary_loss_mlp": 0.0101869,
      "balance_loss_clip": 1.04243076,
      "balance_loss_mlp": 1.01211572,
      "epoch": 0.9156496122166777,
      "flos": 19354990176000.0,
      "grad_norm": 2.7225583354817284,
      "language_loss": 0.7059145,
      "learning_rate": 7.40832095706494e-08,
      "loss": 0.72745693,
      "num_input_tokens_seen": 164657430,
      "step": 7615,
      "time_per_iteration": 2.521418809890747
    },
    {
      "auxiliary_loss_clip": 0.01126548,
      "auxiliary_loss_mlp": 0.0102514,
      "balance_loss_clip": 1.04341173,
      "balance_loss_mlp": 1.01855385,
      "epoch": 0.9157698551073168,
      "flos": 21107287261440.0,
      "grad_norm": 2.6464187009315374,
      "language_loss": 0.80182397,
      "learning_rate": 7.387330594177443e-08,
      "loss": 0.82334089,
      "num_input_tokens_seen": 164679505,
      "step": 7616,
      "time_per_iteration": 2.581125259399414
    },
    {
      "auxiliary_loss_clip": 0.01115453,
      "auxiliary_loss_mlp": 0.01028869,
      "balance_loss_clip": 1.04231119,
      "balance_loss_mlp": 1.02160907,
      "epoch": 0.9158900979979558,
      "flos": 25193167228800.0,
      "grad_norm": 1.4893695217982585,
      "language_loss": 0.79170477,
      "learning_rate": 7.366369450478749e-08,
      "loss": 0.81314802,
      "num_input_tokens_seen": 164700615,
      "step": 7617,
      "time_per_iteration": 2.5845160484313965
    },
    {
      "auxiliary_loss_clip": 0.01116327,
      "auxiliary_loss_mlp": 0.01025437,
      "balance_loss_clip": 1.04126167,
      "balance_loss_mlp": 1.01874983,
      "epoch": 0.916010340888595,
      "flos": 30146648302080.0,
      "grad_norm": 1.6218259885877768,
      "language_loss": 0.65835381,
      "learning_rate": 7.345437529148646e-08,
      "loss": 0.67977142,
      "num_input_tokens_seen": 164719625,
      "step": 7618,
      "time_per_iteration": 2.596672534942627
    },
    {
      "auxiliary_loss_clip": 0.011206,
      "auxiliary_loss_mlp": 0.01025212,
      "balance_loss_clip": 1.04291439,
      "balance_loss_mlp": 1.01829493,
      "epoch": 0.9161305837792341,
      "flos": 17091827907840.0,
      "grad_norm": 1.911550740348573,
      "language_loss": 0.7288968,
      "learning_rate": 7.324534833362483e-08,
      "loss": 0.75035489,
      "num_input_tokens_seen": 164737200,
      "step": 7619,
      "time_per_iteration": 2.5913751125335693
    },
    {
      "auxiliary_loss_clip": 0.01134808,
      "auxiliary_loss_mlp": 0.01023366,
      "balance_loss_clip": 1.04469776,
      "balance_loss_mlp": 1.01701534,
      "epoch": 0.9162508266698731,
      "flos": 22893699288960.0,
      "grad_norm": 1.651393194593592,
      "language_loss": 0.68594521,
      "learning_rate": 7.303661366291192e-08,
      "loss": 0.70752692,
      "num_input_tokens_seen": 164757870,
      "step": 7620,
      "time_per_iteration": 2.6063106060028076
    },
    {
      "auxiliary_loss_clip": 0.01104819,
      "auxiliary_loss_mlp": 0.01030129,
      "balance_loss_clip": 1.04145694,
      "balance_loss_mlp": 1.02268767,
      "epoch": 0.9163710695605123,
      "flos": 19974808287360.0,
      "grad_norm": 1.9198077000930927,
      "language_loss": 0.81618559,
      "learning_rate": 7.28281713110126e-08,
      "loss": 0.83753502,
      "num_input_tokens_seen": 164775945,
      "step": 7621,
      "time_per_iteration": 2.597595453262329
    },
    {
      "auxiliary_loss_clip": 0.01130016,
      "auxiliary_loss_mlp": 0.01025551,
      "balance_loss_clip": 1.04380679,
      "balance_loss_mlp": 1.0192275,
      "epoch": 0.9164913124511513,
      "flos": 22783812606720.0,
      "grad_norm": 2.793982684816331,
      "language_loss": 0.76973128,
      "learning_rate": 7.262002130954759e-08,
      "loss": 0.79128695,
      "num_input_tokens_seen": 164794400,
      "step": 7622,
      "time_per_iteration": 3.491931676864624
    },
    {
      "auxiliary_loss_clip": 0.01109372,
      "auxiliary_loss_mlp": 0.01032092,
      "balance_loss_clip": 1.0401125,
      "balance_loss_mlp": 1.02505565,
      "epoch": 0.9166115553417904,
      "flos": 24900854348160.0,
      "grad_norm": 1.6972904929415682,
      "language_loss": 0.78923273,
      "learning_rate": 7.241216369009296e-08,
      "loss": 0.81064731,
      "num_input_tokens_seen": 164814585,
      "step": 7623,
      "time_per_iteration": 2.653470754623413
    },
    {
      "auxiliary_loss_clip": 0.01163369,
      "auxiliary_loss_mlp": 0.01020043,
      "balance_loss_clip": 1.04645801,
      "balance_loss_mlp": 1.01263428,
      "epoch": 0.9167317982324296,
      "flos": 25702919089920.0,
      "grad_norm": 1.7446276334253201,
      "language_loss": 0.66181791,
      "learning_rate": 7.220459848418037e-08,
      "loss": 0.68365204,
      "num_input_tokens_seen": 164834660,
      "step": 7624,
      "time_per_iteration": 3.3519842624664307
    },
    {
      "auxiliary_loss_clip": 0.01163257,
      "auxiliary_loss_mlp": 0.01033081,
      "balance_loss_clip": 1.04965782,
      "balance_loss_mlp": 1.02661633,
      "epoch": 0.9168520411230686,
      "flos": 15632813370240.0,
      "grad_norm": 1.7172007192672034,
      "language_loss": 0.79445904,
      "learning_rate": 7.199732572329708e-08,
      "loss": 0.81642246,
      "num_input_tokens_seen": 164852560,
      "step": 7625,
      "time_per_iteration": 3.1799309253692627
    },
    {
      "auxiliary_loss_clip": 0.01120547,
      "auxiliary_loss_mlp": 0.01023164,
      "balance_loss_clip": 1.04170394,
      "balance_loss_mlp": 1.01595187,
      "epoch": 0.9169722840137077,
      "flos": 30258151096320.0,
      "grad_norm": 2.3849200558109747,
      "language_loss": 0.75836414,
      "learning_rate": 7.179034543888684e-08,
      "loss": 0.77980125,
      "num_input_tokens_seen": 164872065,
      "step": 7626,
      "time_per_iteration": 2.611210823059082
    },
    {
      "auxiliary_loss_clip": 0.01151292,
      "auxiliary_loss_mlp": 0.01023841,
      "balance_loss_clip": 1.04535723,
      "balance_loss_mlp": 1.01707017,
      "epoch": 0.9170925269043467,
      "flos": 22491643380480.0,
      "grad_norm": 1.8098665338264077,
      "language_loss": 0.77830321,
      "learning_rate": 7.158365766234808e-08,
      "loss": 0.80005455,
      "num_input_tokens_seen": 164890915,
      "step": 7627,
      "time_per_iteration": 2.489579916000366
    },
    {
      "auxiliary_loss_clip": 0.01112197,
      "auxiliary_loss_mlp": 0.01028722,
      "balance_loss_clip": 1.03877556,
      "balance_loss_mlp": 1.02153707,
      "epoch": 0.9172127697949859,
      "flos": 22893914770560.0,
      "grad_norm": 2.0221178805315145,
      "language_loss": 0.7216112,
      "learning_rate": 7.137726242503527e-08,
      "loss": 0.74302042,
      "num_input_tokens_seen": 164909835,
      "step": 7628,
      "time_per_iteration": 2.557910680770874
    },
    {
      "auxiliary_loss_clip": 0.01150318,
      "auxiliary_loss_mlp": 0.00761213,
      "balance_loss_clip": 1.04813313,
      "balance_loss_mlp": 1.00035548,
      "epoch": 0.917333012685625,
      "flos": 17451867882240.0,
      "grad_norm": 2.3480818567928874,
      "language_loss": 0.78097928,
      "learning_rate": 7.11711597582585e-08,
      "loss": 0.8000946,
      "num_input_tokens_seen": 164927195,
      "step": 7629,
      "time_per_iteration": 2.446298122406006
    },
    {
      "auxiliary_loss_clip": 0.01119533,
      "auxiliary_loss_mlp": 0.01024136,
      "balance_loss_clip": 1.03844666,
      "balance_loss_mlp": 1.01770782,
      "epoch": 0.917453255576264,
      "flos": 14318949692160.0,
      "grad_norm": 1.654209540867679,
      "language_loss": 0.79634386,
      "learning_rate": 7.096534969328271e-08,
      "loss": 0.81778055,
      "num_input_tokens_seen": 164944640,
      "step": 7630,
      "time_per_iteration": 2.530721426010132
    },
    {
      "auxiliary_loss_clip": 0.01137156,
      "auxiliary_loss_mlp": 0.01025454,
      "balance_loss_clip": 1.04204464,
      "balance_loss_mlp": 1.01889157,
      "epoch": 0.9175734984669032,
      "flos": 20741177888640.0,
      "grad_norm": 1.932714441210669,
      "language_loss": 0.84176052,
      "learning_rate": 7.075983226132987e-08,
      "loss": 0.86338663,
      "num_input_tokens_seen": 164963570,
      "step": 7631,
      "time_per_iteration": 2.520890474319458
    },
    {
      "auxiliary_loss_clip": 0.01139553,
      "auxiliary_loss_mlp": 0.00761617,
      "balance_loss_clip": 1.04341424,
      "balance_loss_mlp": 1.00027621,
      "epoch": 0.9176937413575422,
      "flos": 14830497233280.0,
      "grad_norm": 2.3541079266607574,
      "language_loss": 0.7901963,
      "learning_rate": 7.055460749357656e-08,
      "loss": 0.80920804,
      "num_input_tokens_seen": 164979850,
      "step": 7632,
      "time_per_iteration": 2.5131471157073975
    },
    {
      "auxiliary_loss_clip": 0.01136688,
      "auxiliary_loss_mlp": 0.01032458,
      "balance_loss_clip": 1.0470295,
      "balance_loss_mlp": 1.02491856,
      "epoch": 0.9178139842481813,
      "flos": 18474603828480.0,
      "grad_norm": 1.699099630802657,
      "language_loss": 0.70207202,
      "learning_rate": 7.034967542115521e-08,
      "loss": 0.72376353,
      "num_input_tokens_seen": 164998115,
      "step": 7633,
      "time_per_iteration": 2.4944441318511963
    },
    {
      "auxiliary_loss_clip": 0.01138893,
      "auxiliary_loss_mlp": 0.00760725,
      "balance_loss_clip": 1.04306471,
      "balance_loss_mlp": 1.00033927,
      "epoch": 0.9179342271388204,
      "flos": 20047455544320.0,
      "grad_norm": 1.9738854432377848,
      "language_loss": 0.75105989,
      "learning_rate": 7.014503607515388e-08,
      "loss": 0.77005601,
      "num_input_tokens_seen": 165017420,
      "step": 7634,
      "time_per_iteration": 2.497682571411133
    },
    {
      "auxiliary_loss_clip": 0.0113733,
      "auxiliary_loss_mlp": 0.01029411,
      "balance_loss_clip": 1.04787517,
      "balance_loss_mlp": 1.0220325,
      "epoch": 0.9180544700294595,
      "flos": 24676232647680.0,
      "grad_norm": 1.9087921325872663,
      "language_loss": 0.68544447,
      "learning_rate": 6.994068948661592e-08,
      "loss": 0.70711184,
      "num_input_tokens_seen": 165035575,
      "step": 7635,
      "time_per_iteration": 2.540978193283081
    },
    {
      "auxiliary_loss_clip": 0.01149167,
      "auxiliary_loss_mlp": 0.01031273,
      "balance_loss_clip": 1.04533041,
      "balance_loss_mlp": 1.02329838,
      "epoch": 0.9181747129200986,
      "flos": 16727478301440.0,
      "grad_norm": 2.4350864347472987,
      "language_loss": 0.7638514,
      "learning_rate": 6.973663568654142e-08,
      "loss": 0.78565586,
      "num_input_tokens_seen": 165053280,
      "step": 7636,
      "time_per_iteration": 2.4792795181274414
    },
    {
      "auxiliary_loss_clip": 0.01163302,
      "auxiliary_loss_mlp": 0.01026916,
      "balance_loss_clip": 1.04792368,
      "balance_loss_mlp": 1.0201509,
      "epoch": 0.9182949558107377,
      "flos": 24271626873600.0,
      "grad_norm": 1.8881347905107015,
      "language_loss": 0.64834678,
      "learning_rate": 6.953287470588386e-08,
      "loss": 0.67024893,
      "num_input_tokens_seen": 165071235,
      "step": 7637,
      "time_per_iteration": 3.2862117290496826
    },
    {
      "auxiliary_loss_clip": 0.0115111,
      "auxiliary_loss_mlp": 0.01024622,
      "balance_loss_clip": 1.04413092,
      "balance_loss_mlp": 1.01743364,
      "epoch": 0.9184151987013768,
      "flos": 22082117443200.0,
      "grad_norm": 2.052889639735495,
      "language_loss": 0.86048305,
      "learning_rate": 6.932940657555452e-08,
      "loss": 0.88224036,
      "num_input_tokens_seen": 165087365,
      "step": 7638,
      "time_per_iteration": 2.4721901416778564
    },
    {
      "auxiliary_loss_clip": 0.01159272,
      "auxiliary_loss_mlp": 0.0102763,
      "balance_loss_clip": 1.04758525,
      "balance_loss_mlp": 1.02170861,
      "epoch": 0.9185354415920158,
      "flos": 32166732257280.0,
      "grad_norm": 1.4310083669181788,
      "language_loss": 0.76293689,
      "learning_rate": 6.912623132641938e-08,
      "loss": 0.78480589,
      "num_input_tokens_seen": 165112455,
      "step": 7639,
      "time_per_iteration": 2.6046388149261475
    },
    {
      "auxiliary_loss_clip": 0.01137234,
      "auxiliary_loss_mlp": 0.01029021,
      "balance_loss_clip": 1.04549706,
      "balance_loss_mlp": 1.0220294,
      "epoch": 0.918655684482655,
      "flos": 20997831542400.0,
      "grad_norm": 2.063562705998082,
      "language_loss": 0.76478601,
      "learning_rate": 6.892334898929952e-08,
      "loss": 0.7864486,
      "num_input_tokens_seen": 165132700,
      "step": 7640,
      "time_per_iteration": 2.5810108184814453
    },
    {
      "auxiliary_loss_clip": 0.01143908,
      "auxiliary_loss_mlp": 0.01029151,
      "balance_loss_clip": 1.04497612,
      "balance_loss_mlp": 1.02258897,
      "epoch": 0.918775927373294,
      "flos": 15560704817280.0,
      "grad_norm": 1.7825278264456614,
      "language_loss": 0.84419233,
      "learning_rate": 6.872075959497236e-08,
      "loss": 0.86592293,
      "num_input_tokens_seen": 165151475,
      "step": 7641,
      "time_per_iteration": 2.568774700164795
    },
    {
      "auxiliary_loss_clip": 0.01146538,
      "auxiliary_loss_mlp": 0.01022037,
      "balance_loss_clip": 1.04335272,
      "balance_loss_mlp": 1.01515305,
      "epoch": 0.9188961702639331,
      "flos": 29934057657600.0,
      "grad_norm": 1.8068815669015137,
      "language_loss": 0.827196,
      "learning_rate": 6.85184631741702e-08,
      "loss": 0.84888184,
      "num_input_tokens_seen": 165172040,
      "step": 7642,
      "time_per_iteration": 2.5405378341674805
    },
    {
      "auxiliary_loss_clip": 0.01148639,
      "auxiliary_loss_mlp": 0.01028518,
      "balance_loss_clip": 1.04562688,
      "balance_loss_mlp": 1.02168393,
      "epoch": 0.9190164131545723,
      "flos": 20701244943360.0,
      "grad_norm": 1.7555574032176964,
      "language_loss": 0.77377486,
      "learning_rate": 6.831645975758161e-08,
      "loss": 0.79554641,
      "num_input_tokens_seen": 165189980,
      "step": 7643,
      "time_per_iteration": 2.5203957557678223
    },
    {
      "auxiliary_loss_clip": 0.0112788,
      "auxiliary_loss_mlp": 0.01029337,
      "balance_loss_clip": 1.04475319,
      "balance_loss_mlp": 1.02151072,
      "epoch": 0.9191366560452113,
      "flos": 25629912696960.0,
      "grad_norm": 1.8401938935219608,
      "language_loss": 0.67278928,
      "learning_rate": 6.811474937585026e-08,
      "loss": 0.69436145,
      "num_input_tokens_seen": 165209770,
      "step": 7644,
      "time_per_iteration": 2.6572232246398926
    },
    {
      "auxiliary_loss_clip": 0.01116884,
      "auxiliary_loss_mlp": 0.01023769,
      "balance_loss_clip": 1.04226041,
      "balance_loss_mlp": 1.01792765,
      "epoch": 0.9192568989358504,
      "flos": 21434325615360.0,
      "grad_norm": 1.530598345694022,
      "language_loss": 0.79114401,
      "learning_rate": 6.79133320595755e-08,
      "loss": 0.81255054,
      "num_input_tokens_seen": 165229690,
      "step": 7645,
      "time_per_iteration": 2.5722286701202393
    },
    {
      "auxiliary_loss_clip": 0.01138763,
      "auxiliary_loss_mlp": 0.01026953,
      "balance_loss_clip": 1.04649782,
      "balance_loss_mlp": 1.02017033,
      "epoch": 0.9193771418264896,
      "flos": 23185078416000.0,
      "grad_norm": 1.641734230474859,
      "language_loss": 0.75438553,
      "learning_rate": 6.771220783931198e-08,
      "loss": 0.7760427,
      "num_input_tokens_seen": 165249850,
      "step": 7646,
      "time_per_iteration": 2.5558085441589355
    },
    {
      "auxiliary_loss_clip": 0.00980346,
      "auxiliary_loss_mlp": 0.00751022,
      "balance_loss_clip": 1.00983763,
      "balance_loss_mlp": 1.00013161,
      "epoch": 0.9194973847171286,
      "flos": 70582963184640.0,
      "grad_norm": 1.0445576493932889,
      "language_loss": 0.64613622,
      "learning_rate": 6.751137674556994e-08,
      "loss": 0.66344988,
      "num_input_tokens_seen": 165310235,
      "step": 7647,
      "time_per_iteration": 3.5503108501434326
    },
    {
      "auxiliary_loss_clip": 0.01151318,
      "auxiliary_loss_mlp": 0.01024447,
      "balance_loss_clip": 1.04374385,
      "balance_loss_mlp": 1.01704395,
      "epoch": 0.9196176276077677,
      "flos": 14720682378240.0,
      "grad_norm": 1.9933636741664869,
      "language_loss": 0.77341312,
      "learning_rate": 6.731083880881572e-08,
      "loss": 0.79517072,
      "num_input_tokens_seen": 165326455,
      "step": 7648,
      "time_per_iteration": 3.7403910160064697
    },
    {
      "auxiliary_loss_clip": 0.01136027,
      "auxiliary_loss_mlp": 0.01022362,
      "balance_loss_clip": 1.0446409,
      "balance_loss_mlp": 1.01582348,
      "epoch": 0.9197378704984068,
      "flos": 23294893271040.0,
      "grad_norm": 2.1386908998783363,
      "language_loss": 0.8096754,
      "learning_rate": 6.711059405947072e-08,
      "loss": 0.83125925,
      "num_input_tokens_seen": 165344645,
      "step": 7649,
      "time_per_iteration": 2.536228895187378
    },
    {
      "auxiliary_loss_clip": 0.01118412,
      "auxiliary_loss_mlp": 0.01029819,
      "balance_loss_clip": 1.04436958,
      "balance_loss_mlp": 1.02311361,
      "epoch": 0.9198581133890459,
      "flos": 20302564913280.0,
      "grad_norm": 1.7655479541017438,
      "language_loss": 0.76958871,
      "learning_rate": 6.691064252791156e-08,
      "loss": 0.791071,
      "num_input_tokens_seen": 165364120,
      "step": 7650,
      "time_per_iteration": 3.616182565689087
    },
    {
      "auxiliary_loss_clip": 0.01099499,
      "auxiliary_loss_mlp": 0.01028985,
      "balance_loss_clip": 1.04271019,
      "balance_loss_mlp": 1.02212405,
      "epoch": 0.9199783562796849,
      "flos": 17675663569920.0,
      "grad_norm": 1.5454846052045323,
      "language_loss": 0.77973354,
      "learning_rate": 6.67109842444713e-08,
      "loss": 0.80101836,
      "num_input_tokens_seen": 165383050,
      "step": 7651,
      "time_per_iteration": 3.2830073833465576
    },
    {
      "auxiliary_loss_clip": 0.01144519,
      "auxiliary_loss_mlp": 0.0076055,
      "balance_loss_clip": 1.0459187,
      "balance_loss_mlp": 1.0002923,
      "epoch": 0.9200985991703241,
      "flos": 17676022705920.0,
      "grad_norm": 1.8069501323315724,
      "language_loss": 0.76650584,
      "learning_rate": 6.651161923943704e-08,
      "loss": 0.78555655,
      "num_input_tokens_seen": 165400955,
      "step": 7652,
      "time_per_iteration": 2.4761130809783936
    },
    {
      "auxiliary_loss_clip": 0.01142858,
      "auxiliary_loss_mlp": 0.01027117,
      "balance_loss_clip": 1.04346418,
      "balance_loss_mlp": 1.01994646,
      "epoch": 0.9202188420609632,
      "flos": 20996574566400.0,
      "grad_norm": 1.7565639733266192,
      "language_loss": 0.77056587,
      "learning_rate": 6.631254754305326e-08,
      "loss": 0.79226565,
      "num_input_tokens_seen": 165420415,
      "step": 7653,
      "time_per_iteration": 2.488304376602173
    },
    {
      "auxiliary_loss_clip": 0.01162746,
      "auxiliary_loss_mlp": 0.01025143,
      "balance_loss_clip": 1.04471147,
      "balance_loss_mlp": 1.01813638,
      "epoch": 0.9203390849516022,
      "flos": 13918222586880.0,
      "grad_norm": 1.8481650229219,
      "language_loss": 0.78227305,
      "learning_rate": 6.611376918551848e-08,
      "loss": 0.80415189,
      "num_input_tokens_seen": 165439200,
      "step": 7654,
      "time_per_iteration": 2.453503370285034
    },
    {
      "auxiliary_loss_clip": 0.01113769,
      "auxiliary_loss_mlp": 0.00761005,
      "balance_loss_clip": 1.04037952,
      "balance_loss_mlp": 1.00028729,
      "epoch": 0.9204593278422414,
      "flos": 21175912195200.0,
      "grad_norm": 2.9978394828784096,
      "language_loss": 0.79303151,
      "learning_rate": 6.591528419698744e-08,
      "loss": 0.81177926,
      "num_input_tokens_seen": 165458985,
      "step": 7655,
      "time_per_iteration": 2.5455081462860107
    },
    {
      "auxiliary_loss_clip": 0.01135557,
      "auxiliary_loss_mlp": 0.01022893,
      "balance_loss_clip": 1.04204512,
      "balance_loss_mlp": 1.01635468,
      "epoch": 0.9205795707328804,
      "flos": 14501375890560.0,
      "grad_norm": 2.3235160184181334,
      "language_loss": 0.83418882,
      "learning_rate": 6.571709260756986e-08,
      "loss": 0.85577333,
      "num_input_tokens_seen": 165475630,
      "step": 7656,
      "time_per_iteration": 2.5028228759765625
    },
    {
      "auxiliary_loss_clip": 0.0115188,
      "auxiliary_loss_mlp": 0.01028458,
      "balance_loss_clip": 1.04898262,
      "balance_loss_mlp": 1.02177036,
      "epoch": 0.9206998136235195,
      "flos": 22417559579520.0,
      "grad_norm": 2.559173880076541,
      "language_loss": 0.76446748,
      "learning_rate": 6.551919444733122e-08,
      "loss": 0.78627086,
      "num_input_tokens_seen": 165493445,
      "step": 7657,
      "time_per_iteration": 2.484921455383301
    },
    {
      "auxiliary_loss_clip": 0.01136669,
      "auxiliary_loss_mlp": 0.01021647,
      "balance_loss_clip": 1.0469178,
      "balance_loss_mlp": 1.01493263,
      "epoch": 0.9208200565141585,
      "flos": 53358407544960.0,
      "grad_norm": 1.8815481446419366,
      "language_loss": 0.6600818,
      "learning_rate": 6.53215897462931e-08,
      "loss": 0.681665,
      "num_input_tokens_seen": 165517200,
      "step": 7658,
      "time_per_iteration": 2.827063798904419
    },
    {
      "auxiliary_loss_clip": 0.01147051,
      "auxiliary_loss_mlp": 0.01034329,
      "balance_loss_clip": 1.04507172,
      "balance_loss_mlp": 1.02699709,
      "epoch": 0.9209402994047977,
      "flos": 30589139946240.0,
      "grad_norm": 2.654986365830183,
      "language_loss": 0.74559379,
      "learning_rate": 6.512427853443103e-08,
      "loss": 0.76740766,
      "num_input_tokens_seen": 165539280,
      "step": 7659,
      "time_per_iteration": 2.5660204887390137
    },
    {
      "auxiliary_loss_clip": 0.01149746,
      "auxiliary_loss_mlp": 0.01022222,
      "balance_loss_clip": 1.0452745,
      "balance_loss_mlp": 1.01537967,
      "epoch": 0.9210605422954368,
      "flos": 29132711187840.0,
      "grad_norm": 1.531669066675027,
      "language_loss": 0.75934571,
      "learning_rate": 6.492726084167799e-08,
      "loss": 0.7810654,
      "num_input_tokens_seen": 165561395,
      "step": 7660,
      "time_per_iteration": 2.5592429637908936
    },
    {
      "auxiliary_loss_clip": 0.01054367,
      "auxiliary_loss_mlp": 0.01002413,
      "balance_loss_clip": 1.00750971,
      "balance_loss_mlp": 1.00133991,
      "epoch": 0.9211807851860758,
      "flos": 54853838472960.0,
      "grad_norm": 0.8361272992307527,
      "language_loss": 0.57546914,
      "learning_rate": 6.473053669792072e-08,
      "loss": 0.59603697,
      "num_input_tokens_seen": 165616085,
      "step": 7661,
      "time_per_iteration": 2.932178020477295
    },
    {
      "auxiliary_loss_clip": 0.0114688,
      "auxiliary_loss_mlp": 0.01024243,
      "balance_loss_clip": 1.04416251,
      "balance_loss_mlp": 1.01713848,
      "epoch": 0.921301028076715,
      "flos": 19201974238080.0,
      "grad_norm": 2.1867030196138173,
      "language_loss": 0.73093599,
      "learning_rate": 6.453410613300248e-08,
      "loss": 0.75264728,
      "num_input_tokens_seen": 165634015,
      "step": 7662,
      "time_per_iteration": 2.475830316543579
    },
    {
      "auxiliary_loss_clip": 0.01095204,
      "auxiliary_loss_mlp": 0.01027805,
      "balance_loss_clip": 1.04213417,
      "balance_loss_mlp": 1.02088761,
      "epoch": 0.921421270967354,
      "flos": 27526893765120.0,
      "grad_norm": 1.5267614312039506,
      "language_loss": 0.57827556,
      "learning_rate": 6.43379691767214e-08,
      "loss": 0.59950566,
      "num_input_tokens_seen": 165653220,
      "step": 7663,
      "time_per_iteration": 3.497004985809326
    },
    {
      "auxiliary_loss_clip": 0.01014622,
      "auxiliary_loss_mlp": 0.01001189,
      "balance_loss_clip": 1.00709343,
      "balance_loss_mlp": 1.00010383,
      "epoch": 0.9215415138579931,
      "flos": 70209311955840.0,
      "grad_norm": 0.7203311196117483,
      "language_loss": 0.55173892,
      "learning_rate": 6.414212585883105e-08,
      "loss": 0.57189703,
      "num_input_tokens_seen": 165715850,
      "step": 7664,
      "time_per_iteration": 3.227815866470337
    },
    {
      "auxiliary_loss_clip": 0.01137951,
      "auxiliary_loss_mlp": 0.01025273,
      "balance_loss_clip": 1.04566455,
      "balance_loss_mlp": 1.01812339,
      "epoch": 0.9216617567486323,
      "flos": 35553107790720.0,
      "grad_norm": 1.5303492803399512,
      "language_loss": 0.69705069,
      "learning_rate": 6.394657620904143e-08,
      "loss": 0.71868289,
      "num_input_tokens_seen": 165738960,
      "step": 7665,
      "time_per_iteration": 2.634403944015503
    },
    {
      "auxiliary_loss_clip": 0.01164804,
      "auxiliary_loss_mlp": 0.01027977,
      "balance_loss_clip": 1.04753089,
      "balance_loss_mlp": 1.02046633,
      "epoch": 0.9217819996392713,
      "flos": 29533330552320.0,
      "grad_norm": 1.7269338344895377,
      "language_loss": 0.72010994,
      "learning_rate": 6.375132025701657e-08,
      "loss": 0.74203771,
      "num_input_tokens_seen": 165761260,
      "step": 7666,
      "time_per_iteration": 2.5288872718811035
    },
    {
      "auxiliary_loss_clip": 0.01166009,
      "auxiliary_loss_mlp": 0.01028483,
      "balance_loss_clip": 1.04954195,
      "balance_loss_mlp": 1.02062988,
      "epoch": 0.9219022425299104,
      "flos": 14574669592320.0,
      "grad_norm": 2.163519874451551,
      "language_loss": 0.68889356,
      "learning_rate": 6.355635803237724e-08,
      "loss": 0.71083844,
      "num_input_tokens_seen": 165776960,
      "step": 7667,
      "time_per_iteration": 2.4355993270874023
    },
    {
      "auxiliary_loss_clip": 0.01146817,
      "auxiliary_loss_mlp": 0.01027173,
      "balance_loss_clip": 1.04396963,
      "balance_loss_mlp": 1.01987791,
      "epoch": 0.9220224854205495,
      "flos": 18077503996800.0,
      "grad_norm": 2.4882929743183095,
      "language_loss": 0.7968623,
      "learning_rate": 6.336168956469867e-08,
      "loss": 0.8186022,
      "num_input_tokens_seen": 165795435,
      "step": 7668,
      "time_per_iteration": 2.4568698406219482
    },
    {
      "auxiliary_loss_clip": 0.01128038,
      "auxiliary_loss_mlp": 0.01031168,
      "balance_loss_clip": 1.04400909,
      "balance_loss_mlp": 1.02473962,
      "epoch": 0.9221427283111886,
      "flos": 24790464875520.0,
      "grad_norm": 1.5655812884996205,
      "language_loss": 0.72036123,
      "learning_rate": 6.316731488351168e-08,
      "loss": 0.74195331,
      "num_input_tokens_seen": 165816625,
      "step": 7669,
      "time_per_iteration": 2.537635326385498
    },
    {
      "auxiliary_loss_clip": 0.01149486,
      "auxiliary_loss_mlp": 0.01022331,
      "balance_loss_clip": 1.04684746,
      "balance_loss_mlp": 1.01532197,
      "epoch": 0.9222629712018277,
      "flos": 13845036625920.0,
      "grad_norm": 2.6677993551257435,
      "language_loss": 0.63433349,
      "learning_rate": 6.297323401830334e-08,
      "loss": 0.65605164,
      "num_input_tokens_seen": 165835410,
      "step": 7670,
      "time_per_iteration": 2.469575881958008
    },
    {
      "auxiliary_loss_clip": 0.01150467,
      "auxiliary_loss_mlp": 0.0103165,
      "balance_loss_clip": 1.04553425,
      "balance_loss_mlp": 1.02476549,
      "epoch": 0.9223832140924668,
      "flos": 21616177196160.0,
      "grad_norm": 1.872931138501021,
      "language_loss": 0.68843591,
      "learning_rate": 6.277944699851523e-08,
      "loss": 0.71025711,
      "num_input_tokens_seen": 165854930,
      "step": 7671,
      "time_per_iteration": 2.4808924198150635
    },
    {
      "auxiliary_loss_clip": 0.01161546,
      "auxiliary_loss_mlp": 0.01028702,
      "balance_loss_clip": 1.04596746,
      "balance_loss_mlp": 1.02195752,
      "epoch": 0.9225034569831059,
      "flos": 21142084561920.0,
      "grad_norm": 1.7435335082504113,
      "language_loss": 0.73247039,
      "learning_rate": 6.25859538535447e-08,
      "loss": 0.75437284,
      "num_input_tokens_seen": 165875725,
      "step": 7672,
      "time_per_iteration": 2.4509432315826416
    },
    {
      "auxiliary_loss_clip": 0.01136767,
      "auxiliary_loss_mlp": 0.01026948,
      "balance_loss_clip": 1.04678154,
      "balance_loss_mlp": 1.02005434,
      "epoch": 0.9226236998737449,
      "flos": 12495046844160.0,
      "grad_norm": 2.7573546752160736,
      "language_loss": 0.78039718,
      "learning_rate": 6.239275461274474e-08,
      "loss": 0.80203438,
      "num_input_tokens_seen": 165892100,
      "step": 7673,
      "time_per_iteration": 2.4919064044952393
    },
    {
      "auxiliary_loss_clip": 0.01150165,
      "auxiliary_loss_mlp": 0.01026909,
      "balance_loss_clip": 1.04740798,
      "balance_loss_mlp": 1.02021241,
      "epoch": 0.9227439427643841,
      "flos": 26214071581440.0,
      "grad_norm": 1.7452956926787306,
      "language_loss": 0.8596037,
      "learning_rate": 6.219984930542299e-08,
      "loss": 0.88137448,
      "num_input_tokens_seen": 165912840,
      "step": 7674,
      "time_per_iteration": 3.305701971054077
    },
    {
      "auxiliary_loss_clip": 0.01152825,
      "auxiliary_loss_mlp": 0.01027955,
      "balance_loss_clip": 1.04665768,
      "balance_loss_mlp": 1.02040935,
      "epoch": 0.9228641856550232,
      "flos": 17967581400960.0,
      "grad_norm": 2.4401691037694073,
      "language_loss": 0.75740826,
      "learning_rate": 6.200723796084383e-08,
      "loss": 0.77921605,
      "num_input_tokens_seen": 165930935,
      "step": 7675,
      "time_per_iteration": 2.4476311206817627
    },
    {
      "auxiliary_loss_clip": 0.01030115,
      "auxiliary_loss_mlp": 0.01001803,
      "balance_loss_clip": 1.01000547,
      "balance_loss_mlp": 1.00061703,
      "epoch": 0.9229844285456622,
      "flos": 70420609710720.0,
      "grad_norm": 0.7593497909037988,
      "language_loss": 0.63055813,
      "learning_rate": 6.181492060822546e-08,
      "loss": 0.65087736,
      "num_input_tokens_seen": 165991110,
      "step": 7676,
      "time_per_iteration": 3.8189635276794434
    },
    {
      "auxiliary_loss_clip": 0.01104482,
      "auxiliary_loss_mlp": 0.01024712,
      "balance_loss_clip": 1.03963017,
      "balance_loss_mlp": 1.01816702,
      "epoch": 0.9231046714363014,
      "flos": 17967832796160.0,
      "grad_norm": 2.944341498075886,
      "language_loss": 0.82152832,
      "learning_rate": 6.162289727674274e-08,
      "loss": 0.84282029,
      "num_input_tokens_seen": 166008790,
      "step": 7677,
      "time_per_iteration": 3.2444796562194824
    },
    {
      "auxiliary_loss_clip": 0.01119827,
      "auxiliary_loss_mlp": 0.01023433,
      "balance_loss_clip": 1.04198241,
      "balance_loss_mlp": 1.01734114,
      "epoch": 0.9232249143269404,
      "flos": 17858233422720.0,
      "grad_norm": 2.233484917530925,
      "language_loss": 0.87798786,
      "learning_rate": 6.143116799552527e-08,
      "loss": 0.8994205,
      "num_input_tokens_seen": 166025035,
      "step": 7678,
      "time_per_iteration": 2.523056745529175
    },
    {
      "auxiliary_loss_clip": 0.01151207,
      "auxiliary_loss_mlp": 0.01024648,
      "balance_loss_clip": 1.04581594,
      "balance_loss_mlp": 1.01764464,
      "epoch": 0.9233451572175795,
      "flos": 23404384903680.0,
      "grad_norm": 2.5184850434258146,
      "language_loss": 0.55793488,
      "learning_rate": 6.123973279365802e-08,
      "loss": 0.57969338,
      "num_input_tokens_seen": 166044010,
      "step": 7679,
      "time_per_iteration": 2.5386502742767334
    },
    {
      "auxiliary_loss_clip": 0.01152278,
      "auxiliary_loss_mlp": 0.01036492,
      "balance_loss_clip": 1.04766941,
      "balance_loss_mlp": 1.0297296,
      "epoch": 0.9234654001082186,
      "flos": 17999326045440.0,
      "grad_norm": 1.7298326821859793,
      "language_loss": 0.77812696,
      "learning_rate": 6.10485917001824e-08,
      "loss": 0.80001462,
      "num_input_tokens_seen": 166061865,
      "step": 7680,
      "time_per_iteration": 2.478100061416626
    },
    {
      "auxiliary_loss_clip": 0.01132932,
      "auxiliary_loss_mlp": 0.01026849,
      "balance_loss_clip": 1.04174495,
      "balance_loss_mlp": 1.02018499,
      "epoch": 0.9235856429988577,
      "flos": 24750747411840.0,
      "grad_norm": 1.492200027632409,
      "language_loss": 0.80973506,
      "learning_rate": 6.085774474409322e-08,
      "loss": 0.8313328,
      "num_input_tokens_seen": 166082425,
      "step": 7681,
      "time_per_iteration": 2.618298292160034
    },
    {
      "auxiliary_loss_clip": 0.01137618,
      "auxiliary_loss_mlp": 0.01028144,
      "balance_loss_clip": 1.04773974,
      "balance_loss_mlp": 1.02126586,
      "epoch": 0.9237058858894968,
      "flos": 14099894599680.0,
      "grad_norm": 1.7709358630121466,
      "language_loss": 0.6976428,
      "learning_rate": 6.066719195434267e-08,
      "loss": 0.71930045,
      "num_input_tokens_seen": 166100225,
      "step": 7682,
      "time_per_iteration": 2.497868061065674
    },
    {
      "auxiliary_loss_clip": 0.0115231,
      "auxiliary_loss_mlp": 0.01030494,
      "balance_loss_clip": 1.0473659,
      "balance_loss_mlp": 1.02299595,
      "epoch": 0.9238261287801359,
      "flos": 28694529175680.0,
      "grad_norm": 2.1840117427667165,
      "language_loss": 0.66323727,
      "learning_rate": 6.047693335983717e-08,
      "loss": 0.68506533,
      "num_input_tokens_seen": 166122570,
      "step": 7683,
      "time_per_iteration": 2.5644378662109375
    },
    {
      "auxiliary_loss_clip": 0.01152606,
      "auxiliary_loss_mlp": 0.01030392,
      "balance_loss_clip": 1.04630017,
      "balance_loss_mlp": 1.02244711,
      "epoch": 0.923946371670775,
      "flos": 23111856541440.0,
      "grad_norm": 2.33429427601771,
      "language_loss": 0.82335985,
      "learning_rate": 6.028696898943853e-08,
      "loss": 0.84518981,
      "num_input_tokens_seen": 166141630,
      "step": 7684,
      "time_per_iteration": 2.503946542739868
    },
    {
      "auxiliary_loss_clip": 0.01136309,
      "auxiliary_loss_mlp": 0.00761067,
      "balance_loss_clip": 1.04324627,
      "balance_loss_mlp": 1.00031281,
      "epoch": 0.924066614561414,
      "flos": 21867120587520.0,
      "grad_norm": 1.9877791811572185,
      "language_loss": 0.71079314,
      "learning_rate": 6.00972988719648e-08,
      "loss": 0.72976685,
      "num_input_tokens_seen": 166159865,
      "step": 7685,
      "time_per_iteration": 2.5226075649261475
    },
    {
      "auxiliary_loss_clip": 0.01123058,
      "auxiliary_loss_mlp": 0.0076083,
      "balance_loss_clip": 1.04249895,
      "balance_loss_mlp": 1.00030708,
      "epoch": 0.9241868574520532,
      "flos": 28511887495680.0,
      "grad_norm": 2.3109812188963095,
      "language_loss": 0.70687747,
      "learning_rate": 5.990792303618807e-08,
      "loss": 0.72571629,
      "num_input_tokens_seen": 166179445,
      "step": 7686,
      "time_per_iteration": 2.602893114089966
    },
    {
      "auxiliary_loss_clip": 0.01118792,
      "auxiliary_loss_mlp": 0.01020924,
      "balance_loss_clip": 1.04255605,
      "balance_loss_mlp": 1.01460934,
      "epoch": 0.9243071003426923,
      "flos": 30518324282880.0,
      "grad_norm": 1.8341891207524899,
      "language_loss": 0.69449145,
      "learning_rate": 5.971884151083695e-08,
      "loss": 0.71588862,
      "num_input_tokens_seen": 166201855,
      "step": 7687,
      "time_per_iteration": 2.647028684616089
    },
    {
      "auxiliary_loss_clip": 0.011395,
      "auxiliary_loss_mlp": 0.01027212,
      "balance_loss_clip": 1.04611504,
      "balance_loss_mlp": 1.02057505,
      "epoch": 0.9244273432333313,
      "flos": 28658331244800.0,
      "grad_norm": 1.778803184095275,
      "language_loss": 0.74242657,
      "learning_rate": 5.9530054324595124e-08,
      "loss": 0.76409376,
      "num_input_tokens_seen": 166221970,
      "step": 7688,
      "time_per_iteration": 2.587827205657959
    },
    {
      "auxiliary_loss_clip": 0.01041105,
      "auxiliary_loss_mlp": 0.00751281,
      "balance_loss_clip": 1.00688946,
      "balance_loss_mlp": 1.0001992,
      "epoch": 0.9245475861239704,
      "flos": 66230589237120.0,
      "grad_norm": 0.72228600628726,
      "language_loss": 0.57581925,
      "learning_rate": 5.934156150610103e-08,
      "loss": 0.59374309,
      "num_input_tokens_seen": 166279335,
      "step": 7689,
      "time_per_iteration": 3.8479158878326416
    },
    {
      "auxiliary_loss_clip": 0.01133574,
      "auxiliary_loss_mlp": 0.01024408,
      "balance_loss_clip": 1.0433979,
      "balance_loss_mlp": 1.01757705,
      "epoch": 0.9246678290146095,
      "flos": 24239918142720.0,
      "grad_norm": 2.240030638223351,
      "language_loss": 0.79165953,
      "learning_rate": 5.915336308394914e-08,
      "loss": 0.81323934,
      "num_input_tokens_seen": 166298170,
      "step": 7690,
      "time_per_iteration": 2.5376858711242676
    },
    {
      "auxiliary_loss_clip": 0.01145682,
      "auxiliary_loss_mlp": 0.01025048,
      "balance_loss_clip": 1.04516292,
      "balance_loss_mlp": 1.01835418,
      "epoch": 0.9247880719052486,
      "flos": 18988808976000.0,
      "grad_norm": 1.5318764621139471,
      "language_loss": 0.7683115,
      "learning_rate": 5.89654590866886e-08,
      "loss": 0.7900188,
      "num_input_tokens_seen": 166317670,
      "step": 7691,
      "time_per_iteration": 2.500533103942871
    },
    {
      "auxiliary_loss_clip": 0.01096829,
      "auxiliary_loss_mlp": 0.01023055,
      "balance_loss_clip": 1.04382539,
      "balance_loss_mlp": 1.01563454,
      "epoch": 0.9249083147958876,
      "flos": 24024095274240.0,
      "grad_norm": 2.040693574887011,
      "language_loss": 0.88080895,
      "learning_rate": 5.877784954282483e-08,
      "loss": 0.90200782,
      "num_input_tokens_seen": 166337010,
      "step": 7692,
      "time_per_iteration": 2.6274616718292236
    },
    {
      "auxiliary_loss_clip": 0.01152969,
      "auxiliary_loss_mlp": 0.01022513,
      "balance_loss_clip": 1.0463562,
      "balance_loss_mlp": 1.01519954,
      "epoch": 0.9250285576865268,
      "flos": 30773972355840.0,
      "grad_norm": 1.9847138999261853,
      "language_loss": 0.72499931,
      "learning_rate": 5.8590534480817963e-08,
      "loss": 0.74675417,
      "num_input_tokens_seen": 166358735,
      "step": 7693,
      "time_per_iteration": 2.5713694095611572
    },
    {
      "auxiliary_loss_clip": 0.01166021,
      "auxiliary_loss_mlp": 0.01025598,
      "balance_loss_clip": 1.04954076,
      "balance_loss_mlp": 1.01806998,
      "epoch": 0.9251488005771659,
      "flos": 10633581348480.0,
      "grad_norm": 2.277819387373324,
      "language_loss": 0.72236764,
      "learning_rate": 5.840351392908349e-08,
      "loss": 0.7442838,
      "num_input_tokens_seen": 166374455,
      "step": 7694,
      "time_per_iteration": 2.433427572250366
    },
    {
      "auxiliary_loss_clip": 0.01140136,
      "auxiliary_loss_mlp": 0.00760635,
      "balance_loss_clip": 1.04330206,
      "balance_loss_mlp": 1.00028861,
      "epoch": 0.9252690434678049,
      "flos": 23586416052480.0,
      "grad_norm": 2.833469425040787,
      "language_loss": 0.70580804,
      "learning_rate": 5.821678791599205e-08,
      "loss": 0.72481573,
      "num_input_tokens_seen": 166393900,
      "step": 7695,
      "time_per_iteration": 2.5653109550476074
    },
    {
      "auxiliary_loss_clip": 0.01137014,
      "auxiliary_loss_mlp": 0.01027772,
      "balance_loss_clip": 1.04621196,
      "balance_loss_mlp": 1.02102828,
      "epoch": 0.9253892863584441,
      "flos": 21469158829440.0,
      "grad_norm": 1.878808402566749,
      "language_loss": 0.80588621,
      "learning_rate": 5.803035646986965e-08,
      "loss": 0.82753408,
      "num_input_tokens_seen": 166413235,
      "step": 7696,
      "time_per_iteration": 2.5599372386932373
    },
    {
      "auxiliary_loss_clip": 0.01166877,
      "auxiliary_loss_mlp": 0.01025303,
      "balance_loss_clip": 1.04754817,
      "balance_loss_mlp": 1.01762569,
      "epoch": 0.9255095292490831,
      "flos": 17456680304640.0,
      "grad_norm": 2.3349298933423266,
      "language_loss": 0.67308688,
      "learning_rate": 5.7844219618998766e-08,
      "loss": 0.6950087,
      "num_input_tokens_seen": 166427560,
      "step": 7697,
      "time_per_iteration": 2.4302470684051514
    },
    {
      "auxiliary_loss_clip": 0.01106442,
      "auxiliary_loss_mlp": 0.01025955,
      "balance_loss_clip": 1.03803778,
      "balance_loss_mlp": 1.01894546,
      "epoch": 0.9256297721397222,
      "flos": 24750675584640.0,
      "grad_norm": 2.1474153535272236,
      "language_loss": 0.71497786,
      "learning_rate": 5.765837739161505e-08,
      "loss": 0.7363019,
      "num_input_tokens_seen": 166446680,
      "step": 7698,
      "time_per_iteration": 2.5706520080566406
    },
    {
      "auxiliary_loss_clip": 0.01121558,
      "auxiliary_loss_mlp": 0.01028679,
      "balance_loss_clip": 1.04336882,
      "balance_loss_mlp": 1.02218246,
      "epoch": 0.9257500150303614,
      "flos": 23112215677440.0,
      "grad_norm": 1.5731078966653718,
      "language_loss": 0.7434563,
      "learning_rate": 5.7472829815911504e-08,
      "loss": 0.76495862,
      "num_input_tokens_seen": 166465505,
      "step": 7699,
      "time_per_iteration": 2.558093786239624
    },
    {
      "auxiliary_loss_clip": 0.01131411,
      "auxiliary_loss_mlp": 0.01024095,
      "balance_loss_clip": 1.0439713,
      "balance_loss_mlp": 1.01727343,
      "epoch": 0.9258702579210004,
      "flos": 22564685687040.0,
      "grad_norm": 1.8267780707858625,
      "language_loss": 0.81573522,
      "learning_rate": 5.7287576920035164e-08,
      "loss": 0.83729029,
      "num_input_tokens_seen": 166484520,
      "step": 7700,
      "time_per_iteration": 3.320439100265503
    },
    {
      "auxiliary_loss_clip": 0.01118183,
      "auxiliary_loss_mlp": 0.01022343,
      "balance_loss_clip": 1.04314661,
      "balance_loss_mlp": 1.01566458,
      "epoch": 0.9259905008116395,
      "flos": 30004298703360.0,
      "grad_norm": 1.771719580741003,
      "language_loss": 0.768305,
      "learning_rate": 5.7102618732088435e-08,
      "loss": 0.78971028,
      "num_input_tokens_seen": 166503850,
      "step": 7701,
      "time_per_iteration": 2.600755214691162
    },
    {
      "auxiliary_loss_clip": 0.01137605,
      "auxiliary_loss_mlp": 0.01025472,
      "balance_loss_clip": 1.04596698,
      "balance_loss_mlp": 1.01834607,
      "epoch": 0.9261107437022786,
      "flos": 24572128055040.0,
      "grad_norm": 1.5709040201079039,
      "language_loss": 0.74462175,
      "learning_rate": 5.6917955280130216e-08,
      "loss": 0.76625252,
      "num_input_tokens_seen": 166525330,
      "step": 7702,
      "time_per_iteration": 4.111975431442261
    },
    {
      "auxiliary_loss_clip": 0.01145589,
      "auxiliary_loss_mlp": 0.01027265,
      "balance_loss_clip": 1.04545581,
      "balance_loss_mlp": 1.02010059,
      "epoch": 0.9262309865929177,
      "flos": 22018448586240.0,
      "grad_norm": 2.9937019391656765,
      "language_loss": 0.7193619,
      "learning_rate": 5.6733586592172755e-08,
      "loss": 0.74109054,
      "num_input_tokens_seen": 166544825,
      "step": 7703,
      "time_per_iteration": 2.493257999420166
    },
    {
      "auxiliary_loss_clip": 0.0112948,
      "auxiliary_loss_mlp": 0.00760277,
      "balance_loss_clip": 1.04116893,
      "balance_loss_mlp": 1.0003171,
      "epoch": 0.9263512294835567,
      "flos": 20339481116160.0,
      "grad_norm": 1.8827491365675897,
      "language_loss": 0.79998046,
      "learning_rate": 5.6549512696185244e-08,
      "loss": 0.81887805,
      "num_input_tokens_seen": 166563325,
      "step": 7704,
      "time_per_iteration": 2.5413458347320557
    },
    {
      "auxiliary_loss_clip": 0.01163565,
      "auxiliary_loss_mlp": 0.01026835,
      "balance_loss_clip": 1.04940867,
      "balance_loss_mlp": 1.02008772,
      "epoch": 0.9264714723741959,
      "flos": 21215378263680.0,
      "grad_norm": 1.6228881400248352,
      "language_loss": 0.67940199,
      "learning_rate": 5.636573362009156e-08,
      "loss": 0.70130599,
      "num_input_tokens_seen": 166583385,
      "step": 7705,
      "time_per_iteration": 2.4741077423095703
    },
    {
      "auxiliary_loss_clip": 0.01166428,
      "auxiliary_loss_mlp": 0.01030251,
      "balance_loss_clip": 1.04773843,
      "balance_loss_mlp": 1.02265787,
      "epoch": 0.926591715264835,
      "flos": 18004964480640.0,
      "grad_norm": 2.1476169126112743,
      "language_loss": 0.77088559,
      "learning_rate": 5.618224939177074e-08,
      "loss": 0.7928524,
      "num_input_tokens_seen": 166601290,
      "step": 7706,
      "time_per_iteration": 2.427630662918091
    },
    {
      "auxiliary_loss_clip": 0.01124429,
      "auxiliary_loss_mlp": 0.01027445,
      "balance_loss_clip": 1.04173565,
      "balance_loss_mlp": 1.02034664,
      "epoch": 0.926711958155474,
      "flos": 36167969825280.0,
      "grad_norm": 1.8727036330982585,
      "language_loss": 0.70504987,
      "learning_rate": 5.599906003905719e-08,
      "loss": 0.72656858,
      "num_input_tokens_seen": 166623835,
      "step": 7707,
      "time_per_iteration": 2.6413724422454834
    },
    {
      "auxiliary_loss_clip": 0.0114242,
      "auxiliary_loss_mlp": 0.01025685,
      "balance_loss_clip": 1.04546571,
      "balance_loss_mlp": 1.01860142,
      "epoch": 0.9268322010461132,
      "flos": 21032736583680.0,
      "grad_norm": 2.0856011696089514,
      "language_loss": 0.81887889,
      "learning_rate": 5.581616558974023e-08,
      "loss": 0.84055996,
      "num_input_tokens_seen": 166642400,
      "step": 7708,
      "time_per_iteration": 2.4676475524902344
    },
    {
      "auxiliary_loss_clip": 0.01155274,
      "auxiliary_loss_mlp": 0.0076077,
      "balance_loss_clip": 1.04701793,
      "balance_loss_mlp": 1.00028646,
      "epoch": 0.9269524439367522,
      "flos": 22964838174720.0,
      "grad_norm": 1.720788183393249,
      "language_loss": 0.79004109,
      "learning_rate": 5.5633566071565444e-08,
      "loss": 0.80920154,
      "num_input_tokens_seen": 166661640,
      "step": 7709,
      "time_per_iteration": 2.5171151161193848
    },
    {
      "auxiliary_loss_clip": 0.01096315,
      "auxiliary_loss_mlp": 0.01019848,
      "balance_loss_clip": 1.03965235,
      "balance_loss_mlp": 1.01311827,
      "epoch": 0.9270726868273913,
      "flos": 41975551468800.0,
      "grad_norm": 1.9725667982392892,
      "language_loss": 0.70748883,
      "learning_rate": 5.5451261512232896e-08,
      "loss": 0.72865051,
      "num_input_tokens_seen": 166684320,
      "step": 7710,
      "time_per_iteration": 2.784224033355713
    },
    {
      "auxiliary_loss_clip": 0.01153071,
      "auxiliary_loss_mlp": 0.01024097,
      "balance_loss_clip": 1.04520941,
      "balance_loss_mlp": 1.01670933,
      "epoch": 0.9271929297180305,
      "flos": 19791771557760.0,
      "grad_norm": 2.8486836527600614,
      "language_loss": 0.62283218,
      "learning_rate": 5.5269251939397576e-08,
      "loss": 0.64460385,
      "num_input_tokens_seen": 166703835,
      "step": 7711,
      "time_per_iteration": 2.4868571758270264
    },
    {
      "auxiliary_loss_clip": 0.01119859,
      "auxiliary_loss_mlp": 0.01021878,
      "balance_loss_clip": 1.0399375,
      "balance_loss_mlp": 1.01452851,
      "epoch": 0.9273131726086695,
      "flos": 19968343839360.0,
      "grad_norm": 2.041828099144128,
      "language_loss": 0.76562399,
      "learning_rate": 5.508753738067073e-08,
      "loss": 0.78704131,
      "num_input_tokens_seen": 166723375,
      "step": 7712,
      "time_per_iteration": 2.56257963180542
    },
    {
      "auxiliary_loss_clip": 0.0115276,
      "auxiliary_loss_mlp": 0.01027221,
      "balance_loss_clip": 1.0446558,
      "balance_loss_mlp": 1.02045321,
      "epoch": 0.9274334154993086,
      "flos": 23258587599360.0,
      "grad_norm": 1.8597029596727213,
      "language_loss": 0.78984201,
      "learning_rate": 5.4906117863617875e-08,
      "loss": 0.81164181,
      "num_input_tokens_seen": 166742760,
      "step": 7713,
      "time_per_iteration": 2.505225419998169
    },
    {
      "auxiliary_loss_clip": 0.01116757,
      "auxiliary_loss_mlp": 0.01025082,
      "balance_loss_clip": 1.04055202,
      "balance_loss_mlp": 1.01877856,
      "epoch": 0.9275536583899477,
      "flos": 31795343585280.0,
      "grad_norm": 1.8905904802750306,
      "language_loss": 0.78255057,
      "learning_rate": 5.4724993415760533e-08,
      "loss": 0.80396897,
      "num_input_tokens_seen": 166761115,
      "step": 7714,
      "time_per_iteration": 2.6505308151245117
    },
    {
      "auxiliary_loss_clip": 0.01128112,
      "auxiliary_loss_mlp": 0.00761256,
      "balance_loss_clip": 1.04175448,
      "balance_loss_mlp": 1.00036287,
      "epoch": 0.9276739012805868,
      "flos": 18696998885760.0,
      "grad_norm": 3.0696474371003415,
      "language_loss": 0.74441141,
      "learning_rate": 5.454416406457496e-08,
      "loss": 0.76330507,
      "num_input_tokens_seen": 166780210,
      "step": 7715,
      "time_per_iteration": 3.3030738830566406
    },
    {
      "auxiliary_loss_clip": 0.01147781,
      "auxiliary_loss_mlp": 0.01025295,
      "balance_loss_clip": 1.04393578,
      "balance_loss_mlp": 1.01914144,
      "epoch": 0.9277941441712259,
      "flos": 13879079740800.0,
      "grad_norm": 2.161976477522184,
      "language_loss": 0.73557997,
      "learning_rate": 5.436362983749299e-08,
      "loss": 0.75731075,
      "num_input_tokens_seen": 166795380,
      "step": 7716,
      "time_per_iteration": 2.45060658454895
    },
    {
      "auxiliary_loss_clip": 0.01121247,
      "auxiliary_loss_mlp": 0.01033835,
      "balance_loss_clip": 1.04898512,
      "balance_loss_mlp": 1.02683115,
      "epoch": 0.927914387061865,
      "flos": 23258659426560.0,
      "grad_norm": 1.8377320720767139,
      "language_loss": 0.64173901,
      "learning_rate": 5.418339076190137e-08,
      "loss": 0.66328979,
      "num_input_tokens_seen": 166814890,
      "step": 7717,
      "time_per_iteration": 2.5724127292633057
    },
    {
      "auxiliary_loss_clip": 0.01130361,
      "auxiliary_loss_mlp": 0.01025723,
      "balance_loss_clip": 1.04531145,
      "balance_loss_mlp": 1.01882648,
      "epoch": 0.9280346299525041,
      "flos": 18073733068800.0,
      "grad_norm": 1.7910359483526372,
      "language_loss": 0.88377035,
      "learning_rate": 5.400344686514202e-08,
      "loss": 0.90533113,
      "num_input_tokens_seen": 166832475,
      "step": 7718,
      "time_per_iteration": 2.501126289367676
    },
    {
      "auxiliary_loss_clip": 0.01148608,
      "auxiliary_loss_mlp": 0.01025393,
      "balance_loss_clip": 1.0469898,
      "balance_loss_mlp": 1.01867223,
      "epoch": 0.9281548728431431,
      "flos": 22342901160960.0,
      "grad_norm": 1.735497392725897,
      "language_loss": 0.66801947,
      "learning_rate": 5.38237981745131e-08,
      "loss": 0.68975949,
      "num_input_tokens_seen": 166850590,
      "step": 7719,
      "time_per_iteration": 2.5678069591522217
    },
    {
      "auxiliary_loss_clip": 0.01150361,
      "auxiliary_loss_mlp": 0.00760557,
      "balance_loss_clip": 1.04482603,
      "balance_loss_mlp": 1.00031722,
      "epoch": 0.9282751157337822,
      "flos": 18843765857280.0,
      "grad_norm": 1.6454047507522218,
      "language_loss": 0.81210983,
      "learning_rate": 5.364444471726592e-08,
      "loss": 0.83121902,
      "num_input_tokens_seen": 166869795,
      "step": 7720,
      "time_per_iteration": 2.4806504249572754
    },
    {
      "auxiliary_loss_clip": 0.01147436,
      "auxiliary_loss_mlp": 0.01029964,
      "balance_loss_clip": 1.04454458,
      "balance_loss_mlp": 1.0226922,
      "epoch": 0.9283953586244214,
      "flos": 25556834476800.0,
      "grad_norm": 1.886145385643796,
      "language_loss": 0.7971949,
      "learning_rate": 5.346538652060939e-08,
      "loss": 0.81896889,
      "num_input_tokens_seen": 166891150,
      "step": 7721,
      "time_per_iteration": 2.544504404067993
    },
    {
      "auxiliary_loss_clip": 0.01134072,
      "auxiliary_loss_mlp": 0.01024202,
      "balance_loss_clip": 1.04620826,
      "balance_loss_mlp": 1.01746118,
      "epoch": 0.9285156015150604,
      "flos": 18223480869120.0,
      "grad_norm": 1.9313693634514075,
      "language_loss": 0.70133543,
      "learning_rate": 5.3286623611705994e-08,
      "loss": 0.72291815,
      "num_input_tokens_seen": 166909195,
      "step": 7722,
      "time_per_iteration": 2.5068166255950928
    },
    {
      "auxiliary_loss_clip": 0.01054436,
      "auxiliary_loss_mlp": 0.01002842,
      "balance_loss_clip": 1.00761294,
      "balance_loss_mlp": 1.00175762,
      "epoch": 0.9286358444056995,
      "flos": 66400017690240.0,
      "grad_norm": 0.8063935666207264,
      "language_loss": 0.60662413,
      "learning_rate": 5.3108156017673824e-08,
      "loss": 0.62719691,
      "num_input_tokens_seen": 166970955,
      "step": 7723,
      "time_per_iteration": 3.179189682006836
    },
    {
      "auxiliary_loss_clip": 0.01137996,
      "auxiliary_loss_mlp": 0.01022916,
      "balance_loss_clip": 1.04571247,
      "balance_loss_mlp": 1.01568568,
      "epoch": 0.9287560872963386,
      "flos": 22345630594560.0,
      "grad_norm": 1.61048877994738,
      "language_loss": 0.71664333,
      "learning_rate": 5.2929983765586775e-08,
      "loss": 0.7382524,
      "num_input_tokens_seen": 166989735,
      "step": 7724,
      "time_per_iteration": 2.522639751434326
    },
    {
      "auxiliary_loss_clip": 0.01161505,
      "auxiliary_loss_mlp": 0.01027627,
      "balance_loss_clip": 1.04671192,
      "balance_loss_mlp": 1.02043271,
      "epoch": 0.9288763301869777,
      "flos": 25700225569920.0,
      "grad_norm": 1.908975374021451,
      "language_loss": 0.62505043,
      "learning_rate": 5.275210688247278e-08,
      "loss": 0.64694172,
      "num_input_tokens_seen": 167010060,
      "step": 7725,
      "time_per_iteration": 2.501647710800171
    },
    {
      "auxiliary_loss_clip": 0.01108437,
      "auxiliary_loss_mlp": 0.01026802,
      "balance_loss_clip": 1.04328227,
      "balance_loss_mlp": 1.02012026,
      "epoch": 0.9289965730776167,
      "flos": 12312046028160.0,
      "grad_norm": 2.4261390793196203,
      "language_loss": 0.84782779,
      "learning_rate": 5.257452539531604e-08,
      "loss": 0.86918014,
      "num_input_tokens_seen": 167027130,
      "step": 7726,
      "time_per_iteration": 3.3678996562957764
    },
    {
      "auxiliary_loss_clip": 0.01148736,
      "auxiliary_loss_mlp": 0.01026689,
      "balance_loss_clip": 1.04380679,
      "balance_loss_mlp": 1.01991224,
      "epoch": 0.9291168159682559,
      "flos": 26685973486080.0,
      "grad_norm": 1.5029415882509454,
      "language_loss": 0.68195033,
      "learning_rate": 5.2397239331055445e-08,
      "loss": 0.7037046,
      "num_input_tokens_seen": 167049130,
      "step": 7727,
      "time_per_iteration": 2.528411626815796
    },
    {
      "auxiliary_loss_clip": 0.01134682,
      "auxiliary_loss_mlp": 0.0102382,
      "balance_loss_clip": 1.04584861,
      "balance_loss_mlp": 1.01668549,
      "epoch": 0.929237058858895,
      "flos": 14538256179840.0,
      "grad_norm": 2.474510087553388,
      "language_loss": 0.81125176,
      "learning_rate": 5.2220248716585036e-08,
      "loss": 0.83283675,
      "num_input_tokens_seen": 167066810,
      "step": 7728,
      "time_per_iteration": 3.2860183715820312
    },
    {
      "auxiliary_loss_clip": 0.01140257,
      "auxiliary_loss_mlp": 0.01030105,
      "balance_loss_clip": 1.04310811,
      "balance_loss_mlp": 1.02283311,
      "epoch": 0.929357301749534,
      "flos": 23835456023040.0,
      "grad_norm": 2.149270020831387,
      "language_loss": 0.75692213,
      "learning_rate": 5.204355357875445e-08,
      "loss": 0.77862579,
      "num_input_tokens_seen": 167085155,
      "step": 7729,
      "time_per_iteration": 3.240177631378174
    },
    {
      "auxiliary_loss_clip": 0.0113127,
      "auxiliary_loss_mlp": 0.01025825,
      "balance_loss_clip": 1.04245806,
      "balance_loss_mlp": 1.0187645,
      "epoch": 0.9294775446401732,
      "flos": 12969319046400.0,
      "grad_norm": 2.7873887043018235,
      "language_loss": 0.70580828,
      "learning_rate": 5.1867153944367584e-08,
      "loss": 0.7273792,
      "num_input_tokens_seen": 167101545,
      "step": 7730,
      "time_per_iteration": 2.4915401935577393
    },
    {
      "auxiliary_loss_clip": 0.01125966,
      "auxiliary_loss_mlp": 0.0102342,
      "balance_loss_clip": 1.04390728,
      "balance_loss_mlp": 1.01726866,
      "epoch": 0.9295977875308122,
      "flos": 26211809024640.0,
      "grad_norm": 1.621498344115582,
      "language_loss": 0.73423123,
      "learning_rate": 5.16910498401848e-08,
      "loss": 0.75572509,
      "num_input_tokens_seen": 167120995,
      "step": 7731,
      "time_per_iteration": 2.5965700149536133
    },
    {
      "auxiliary_loss_clip": 0.01163306,
      "auxiliary_loss_mlp": 0.01023307,
      "balance_loss_clip": 1.05016613,
      "balance_loss_mlp": 1.01729846,
      "epoch": 0.9297180304214513,
      "flos": 16472297105280.0,
      "grad_norm": 1.9677871855528999,
      "language_loss": 0.83362234,
      "learning_rate": 5.151524129292073e-08,
      "loss": 0.85548848,
      "num_input_tokens_seen": 167138890,
      "step": 7732,
      "time_per_iteration": 2.4481592178344727
    },
    {
      "auxiliary_loss_clip": 0.01148261,
      "auxiliary_loss_mlp": 0.01027415,
      "balance_loss_clip": 1.04580581,
      "balance_loss_mlp": 1.02006912,
      "epoch": 0.9298382733120905,
      "flos": 24060436859520.0,
      "grad_norm": 2.7728504800955083,
      "language_loss": 0.66440046,
      "learning_rate": 5.1339728329245155e-08,
      "loss": 0.68615717,
      "num_input_tokens_seen": 167159455,
      "step": 7733,
      "time_per_iteration": 2.5072975158691406
    },
    {
      "auxiliary_loss_clip": 0.01167288,
      "auxiliary_loss_mlp": 0.01027086,
      "balance_loss_clip": 1.04825735,
      "balance_loss_mlp": 1.01994812,
      "epoch": 0.9299585162027295,
      "flos": 22127652910080.0,
      "grad_norm": 3.873609447709687,
      "language_loss": 0.79018366,
      "learning_rate": 5.116451097578367e-08,
      "loss": 0.81212735,
      "num_input_tokens_seen": 167178495,
      "step": 7734,
      "time_per_iteration": 2.4784107208251953
    },
    {
      "auxiliary_loss_clip": 0.01119161,
      "auxiliary_loss_mlp": 0.01024082,
      "balance_loss_clip": 1.04180408,
      "balance_loss_mlp": 1.01782405,
      "epoch": 0.9300787590933686,
      "flos": 21471780522240.0,
      "grad_norm": 1.6151443284423441,
      "language_loss": 0.74342847,
      "learning_rate": 5.0989589259115895e-08,
      "loss": 0.76486087,
      "num_input_tokens_seen": 167199380,
      "step": 7735,
      "time_per_iteration": 2.5633625984191895
    },
    {
      "auxiliary_loss_clip": 0.01146548,
      "auxiliary_loss_mlp": 0.01032822,
      "balance_loss_clip": 1.04379189,
      "balance_loss_mlp": 1.02453685,
      "epoch": 0.9301990019840077,
      "flos": 17779588594560.0,
      "grad_norm": 1.700742103400936,
      "language_loss": 0.71393633,
      "learning_rate": 5.081496320577816e-08,
      "loss": 0.73573005,
      "num_input_tokens_seen": 167216500,
      "step": 7736,
      "time_per_iteration": 2.4816906452178955
    },
    {
      "auxiliary_loss_clip": 0.01029997,
      "auxiliary_loss_mlp": 0.01001485,
      "balance_loss_clip": 1.00840437,
      "balance_loss_mlp": 1.00043607,
      "epoch": 0.9303192448746468,
      "flos": 58896122307840.0,
      "grad_norm": 0.9314083684430908,
      "language_loss": 0.61227691,
      "learning_rate": 5.0640632842260835e-08,
      "loss": 0.63259172,
      "num_input_tokens_seen": 167276760,
      "step": 7737,
      "time_per_iteration": 3.1791553497314453
    },
    {
      "auxiliary_loss_clip": 0.01118608,
      "auxiliary_loss_mlp": 0.00760692,
      "balance_loss_clip": 1.04487288,
      "balance_loss_mlp": 1.00029159,
      "epoch": 0.9304394877652858,
      "flos": 57663522172800.0,
      "grad_norm": 1.4979989358425179,
      "language_loss": 0.72614515,
      "learning_rate": 5.0466598195009426e-08,
      "loss": 0.74493814,
      "num_input_tokens_seen": 167303630,
      "step": 7738,
      "time_per_iteration": 2.880998134613037
    },
    {
      "auxiliary_loss_clip": 0.01120227,
      "auxiliary_loss_mlp": 0.01028151,
      "balance_loss_clip": 1.0444324,
      "balance_loss_mlp": 1.02154946,
      "epoch": 0.930559730655925,
      "flos": 20996143603200.0,
      "grad_norm": 1.9790486910340126,
      "language_loss": 0.70076948,
      "learning_rate": 5.0292859290425036e-08,
      "loss": 0.72225326,
      "num_input_tokens_seen": 167321500,
      "step": 7739,
      "time_per_iteration": 2.5399749279022217
    },
    {
      "auxiliary_loss_clip": 0.01162835,
      "auxiliary_loss_mlp": 0.01024244,
      "balance_loss_clip": 1.04820991,
      "balance_loss_mlp": 1.01736569,
      "epoch": 0.9306799735465641,
      "flos": 23258264376960.0,
      "grad_norm": 1.8212460020888777,
      "language_loss": 0.77775645,
      "learning_rate": 5.011941615486348e-08,
      "loss": 0.79962718,
      "num_input_tokens_seen": 167340615,
      "step": 7740,
      "time_per_iteration": 2.4674181938171387
    },
    {
      "auxiliary_loss_clip": 0.01163745,
      "auxiliary_loss_mlp": 0.01029169,
      "balance_loss_clip": 1.04799652,
      "balance_loss_mlp": 1.02260387,
      "epoch": 0.9308002164372031,
      "flos": 15231547560960.0,
      "grad_norm": 1.8696310088684718,
      "language_loss": 0.84142101,
      "learning_rate": 4.994626881463659e-08,
      "loss": 0.86335015,
      "num_input_tokens_seen": 167356870,
      "step": 7741,
      "time_per_iteration": 3.1739439964294434
    },
    {
      "auxiliary_loss_clip": 0.0109442,
      "auxiliary_loss_mlp": 0.01025255,
      "balance_loss_clip": 1.04034519,
      "balance_loss_mlp": 1.01826072,
      "epoch": 0.9309204593278423,
      "flos": 30847481539200.0,
      "grad_norm": 1.7029864142210447,
      "language_loss": 0.7085762,
      "learning_rate": 4.9773417296009814e-08,
      "loss": 0.72977293,
      "num_input_tokens_seen": 167378390,
      "step": 7742,
      "time_per_iteration": 2.648183822631836
    },
    {
      "auxiliary_loss_clip": 0.01154028,
      "auxiliary_loss_mlp": 0.01034364,
      "balance_loss_clip": 1.04691935,
      "balance_loss_mlp": 1.02682984,
      "epoch": 0.9310407022184813,
      "flos": 23037269950080.0,
      "grad_norm": 2.584485081293265,
      "language_loss": 0.65569925,
      "learning_rate": 4.960086162520527e-08,
      "loss": 0.67758322,
      "num_input_tokens_seen": 167398480,
      "step": 7743,
      "time_per_iteration": 2.517779588699341
    },
    {
      "auxiliary_loss_clip": 0.01110439,
      "auxiliary_loss_mlp": 0.01022779,
      "balance_loss_clip": 1.04256392,
      "balance_loss_mlp": 1.01588321,
      "epoch": 0.9311609451091204,
      "flos": 22127976132480.0,
      "grad_norm": 1.8714804163950711,
      "language_loss": 0.82659018,
      "learning_rate": 4.942860182839936e-08,
      "loss": 0.84792233,
      "num_input_tokens_seen": 167416825,
      "step": 7744,
      "time_per_iteration": 2.5890183448791504
    },
    {
      "auxiliary_loss_clip": 0.01133768,
      "auxiliary_loss_mlp": 0.01024269,
      "balance_loss_clip": 1.04354405,
      "balance_loss_mlp": 1.0178051,
      "epoch": 0.9312811879997596,
      "flos": 21099206701440.0,
      "grad_norm": 1.950409155177218,
      "language_loss": 0.79266948,
      "learning_rate": 4.925663793172341e-08,
      "loss": 0.81424987,
      "num_input_tokens_seen": 167434785,
      "step": 7745,
      "time_per_iteration": 2.533735752105713
    },
    {
      "auxiliary_loss_clip": 0.01032605,
      "auxiliary_loss_mlp": 0.0075119,
      "balance_loss_clip": 1.00870395,
      "balance_loss_mlp": 1.00022125,
      "epoch": 0.9314014308903986,
      "flos": 67148179096320.0,
      "grad_norm": 0.7824167776902249,
      "language_loss": 0.56576383,
      "learning_rate": 4.908496996126477e-08,
      "loss": 0.58360171,
      "num_input_tokens_seen": 167498245,
      "step": 7746,
      "time_per_iteration": 3.1496355533599854
    },
    {
      "auxiliary_loss_clip": 0.01147202,
      "auxiliary_loss_mlp": 0.01026757,
      "balance_loss_clip": 1.04868126,
      "balance_loss_mlp": 1.01961017,
      "epoch": 0.9315216737810377,
      "flos": 22565583527040.0,
      "grad_norm": 1.9292519357996785,
      "language_loss": 0.7617619,
      "learning_rate": 4.89135979430646e-08,
      "loss": 0.78350151,
      "num_input_tokens_seen": 167518290,
      "step": 7747,
      "time_per_iteration": 2.5138373374938965
    },
    {
      "auxiliary_loss_clip": 0.0116477,
      "auxiliary_loss_mlp": 0.01027298,
      "balance_loss_clip": 1.04839754,
      "balance_loss_mlp": 1.01990676,
      "epoch": 0.9316419166716768,
      "flos": 23984054588160.0,
      "grad_norm": 1.6544903470416328,
      "language_loss": 0.85374177,
      "learning_rate": 4.874252190312078e-08,
      "loss": 0.87566245,
      "num_input_tokens_seen": 167538675,
      "step": 7748,
      "time_per_iteration": 2.49051833152771
    },
    {
      "auxiliary_loss_clip": 0.01149136,
      "auxiliary_loss_mlp": 0.01021703,
      "balance_loss_clip": 1.04599464,
      "balance_loss_mlp": 1.01438022,
      "epoch": 0.9317621595623159,
      "flos": 30230464688640.0,
      "grad_norm": 1.5506196554919556,
      "language_loss": 0.64666671,
      "learning_rate": 4.857174186738477e-08,
      "loss": 0.66837507,
      "num_input_tokens_seen": 167562025,
      "step": 7749,
      "time_per_iteration": 2.577256917953491
    },
    {
      "auxiliary_loss_clip": 0.01166749,
      "auxiliary_loss_mlp": 0.010286,
      "balance_loss_clip": 1.0502038,
      "balance_loss_mlp": 1.02171016,
      "epoch": 0.931882402452955,
      "flos": 15742735966080.0,
      "grad_norm": 2.5416754432047846,
      "language_loss": 0.73147076,
      "learning_rate": 4.840125786176408e-08,
      "loss": 0.75342429,
      "num_input_tokens_seen": 167578230,
      "step": 7750,
      "time_per_iteration": 2.468268394470215
    },
    {
      "auxiliary_loss_clip": 0.01133408,
      "auxiliary_loss_mlp": 0.01025171,
      "balance_loss_clip": 1.04608583,
      "balance_loss_mlp": 1.018507,
      "epoch": 0.932002645343594,
      "flos": 28366521154560.0,
      "grad_norm": 1.7593041787217303,
      "language_loss": 0.77466881,
      "learning_rate": 4.823106991212067e-08,
      "loss": 0.79625463,
      "num_input_tokens_seen": 167597470,
      "step": 7751,
      "time_per_iteration": 2.574995279312134
    },
    {
      "auxiliary_loss_clip": 0.01151196,
      "auxiliary_loss_mlp": 0.01025163,
      "balance_loss_clip": 1.04577446,
      "balance_loss_mlp": 1.01869869,
      "epoch": 0.9321228882342332,
      "flos": 15341146934400.0,
      "grad_norm": 1.8711473743901335,
      "language_loss": 0.83435476,
      "learning_rate": 4.806117804427212e-08,
      "loss": 0.85611838,
      "num_input_tokens_seen": 167615405,
      "step": 7752,
      "time_per_iteration": 3.1868414878845215
    },
    {
      "auxiliary_loss_clip": 0.01142911,
      "auxiliary_loss_mlp": 0.01024263,
      "balance_loss_clip": 1.04370093,
      "balance_loss_mlp": 1.01671672,
      "epoch": 0.9322431311248722,
      "flos": 17895365107200.0,
      "grad_norm": 2.277888452349502,
      "language_loss": 0.64905047,
      "learning_rate": 4.7891582283990926e-08,
      "loss": 0.67072225,
      "num_input_tokens_seen": 167634130,
      "step": 7753,
      "time_per_iteration": 2.45401930809021
    },
    {
      "auxiliary_loss_clip": 0.01118022,
      "auxiliary_loss_mlp": 0.01025222,
      "balance_loss_clip": 1.04224992,
      "balance_loss_mlp": 1.01862395,
      "epoch": 0.9323633740155113,
      "flos": 24169713010560.0,
      "grad_norm": 1.5034441095051343,
      "language_loss": 0.7292164,
      "learning_rate": 4.772228265700473e-08,
      "loss": 0.75064886,
      "num_input_tokens_seen": 167654990,
      "step": 7754,
      "time_per_iteration": 3.3526763916015625
    },
    {
      "auxiliary_loss_clip": 0.01150806,
      "auxiliary_loss_mlp": 0.01025376,
      "balance_loss_clip": 1.0458349,
      "balance_loss_mlp": 1.01850677,
      "epoch": 0.9324836169061504,
      "flos": 15043482927360.0,
      "grad_norm": 1.9799520635297339,
      "language_loss": 0.75821394,
      "learning_rate": 4.75532791889961e-08,
      "loss": 0.77997577,
      "num_input_tokens_seen": 167671690,
      "step": 7755,
      "time_per_iteration": 3.2119526863098145
    },
    {
      "auxiliary_loss_clip": 0.01148761,
      "auxiliary_loss_mlp": 0.01025598,
      "balance_loss_clip": 1.04475999,
      "balance_loss_mlp": 1.01864815,
      "epoch": 0.9326038597967895,
      "flos": 18624890332800.0,
      "grad_norm": 1.7255725392126338,
      "language_loss": 0.65477359,
      "learning_rate": 4.738457190560252e-08,
      "loss": 0.67651719,
      "num_input_tokens_seen": 167690800,
      "step": 7756,
      "time_per_iteration": 2.4793736934661865
    },
    {
      "auxiliary_loss_clip": 0.01113877,
      "auxiliary_loss_mlp": 0.01026473,
      "balance_loss_clip": 1.04706132,
      "balance_loss_mlp": 1.01977921,
      "epoch": 0.9327241026874286,
      "flos": 18952646958720.0,
      "grad_norm": 2.845748143523889,
      "language_loss": 0.78581214,
      "learning_rate": 4.721616083241664e-08,
      "loss": 0.80721563,
      "num_input_tokens_seen": 167709055,
      "step": 7757,
      "time_per_iteration": 2.5709803104400635
    },
    {
      "auxiliary_loss_clip": 0.0114475,
      "auxiliary_loss_mlp": 0.01029283,
      "balance_loss_clip": 1.0442642,
      "balance_loss_mlp": 1.02217197,
      "epoch": 0.9328443455780677,
      "flos": 29570282668800.0,
      "grad_norm": 1.651694790742778,
      "language_loss": 0.77375817,
      "learning_rate": 4.7048045994986684e-08,
      "loss": 0.79549849,
      "num_input_tokens_seen": 167729915,
      "step": 7758,
      "time_per_iteration": 2.5611448287963867
    },
    {
      "auxiliary_loss_clip": 0.0115335,
      "auxiliary_loss_mlp": 0.01027993,
      "balance_loss_clip": 1.04844129,
      "balance_loss_mlp": 1.02097762,
      "epoch": 0.9329645884687068,
      "flos": 30081722469120.0,
      "grad_norm": 1.9955302244263122,
      "language_loss": 0.90985668,
      "learning_rate": 4.688022741881559e-08,
      "loss": 0.93167007,
      "num_input_tokens_seen": 167750440,
      "step": 7759,
      "time_per_iteration": 2.5574941635131836
    },
    {
      "auxiliary_loss_clip": 0.01144913,
      "auxiliary_loss_mlp": 0.0102206,
      "balance_loss_clip": 1.04447818,
      "balance_loss_mlp": 1.01537824,
      "epoch": 0.9330848313593458,
      "flos": 21867982513920.0,
      "grad_norm": 1.5016490263663547,
      "language_loss": 0.75145042,
      "learning_rate": 4.671270512936076e-08,
      "loss": 0.77312016,
      "num_input_tokens_seen": 167769600,
      "step": 7760,
      "time_per_iteration": 2.4879982471466064
    },
    {
      "auxiliary_loss_clip": 0.01114244,
      "auxiliary_loss_mlp": 0.01022858,
      "balance_loss_clip": 1.04112589,
      "balance_loss_mlp": 1.01598525,
      "epoch": 0.933205074249985,
      "flos": 22127221946880.0,
      "grad_norm": 1.873066469720172,
      "language_loss": 0.82614625,
      "learning_rate": 4.6545479152035884e-08,
      "loss": 0.84751725,
      "num_input_tokens_seen": 167788770,
      "step": 7761,
      "time_per_iteration": 2.5767979621887207
    },
    {
      "auxiliary_loss_clip": 0.0115066,
      "auxiliary_loss_mlp": 0.01027656,
      "balance_loss_clip": 1.04665244,
      "balance_loss_mlp": 1.02069712,
      "epoch": 0.9333253171406241,
      "flos": 15341254675200.0,
      "grad_norm": 1.873599681868824,
      "language_loss": 0.75915402,
      "learning_rate": 4.637854951220821e-08,
      "loss": 0.78093714,
      "num_input_tokens_seen": 167805555,
      "step": 7762,
      "time_per_iteration": 2.477757215499878
    },
    {
      "auxiliary_loss_clip": 0.01117717,
      "auxiliary_loss_mlp": 0.01026543,
      "balance_loss_clip": 1.04323828,
      "balance_loss_mlp": 1.01971197,
      "epoch": 0.9334455600312631,
      "flos": 15706142985600.0,
      "grad_norm": 3.3462118164088803,
      "language_loss": 0.74944675,
      "learning_rate": 4.621191623520171e-08,
      "loss": 0.7708894,
      "num_input_tokens_seen": 167823985,
      "step": 7763,
      "time_per_iteration": 2.5756447315216064
    },
    {
      "auxiliary_loss_clip": 0.01102752,
      "auxiliary_loss_mlp": 0.01023461,
      "balance_loss_clip": 1.04175389,
      "balance_loss_mlp": 1.01677883,
      "epoch": 0.9335658029219023,
      "flos": 22163563532160.0,
      "grad_norm": 2.2280610450207248,
      "language_loss": 0.84517509,
      "learning_rate": 4.604557934629372e-08,
      "loss": 0.8664372,
      "num_input_tokens_seen": 167843060,
      "step": 7764,
      "time_per_iteration": 2.6184310913085938
    },
    {
      "auxiliary_loss_clip": 0.01131768,
      "auxiliary_loss_mlp": 0.01026237,
      "balance_loss_clip": 1.04545546,
      "balance_loss_mlp": 1.01921272,
      "epoch": 0.9336860458125413,
      "flos": 20266833859200.0,
      "grad_norm": 1.6412641832078307,
      "language_loss": 0.80325556,
      "learning_rate": 4.587953887071805e-08,
      "loss": 0.8248356,
      "num_input_tokens_seen": 167862880,
      "step": 7765,
      "time_per_iteration": 2.5134949684143066
    },
    {
      "auxiliary_loss_clip": 0.01131619,
      "auxiliary_loss_mlp": 0.01027316,
      "balance_loss_clip": 1.04200387,
      "balance_loss_mlp": 1.02030027,
      "epoch": 0.9338062887031804,
      "flos": 20919689504640.0,
      "grad_norm": 1.7467057428700739,
      "language_loss": 0.85794318,
      "learning_rate": 4.5713794833662554e-08,
      "loss": 0.87953246,
      "num_input_tokens_seen": 167882095,
      "step": 7766,
      "time_per_iteration": 3.3173017501831055
    },
    {
      "auxiliary_loss_clip": 0.01162132,
      "auxiliary_loss_mlp": 0.01025967,
      "balance_loss_clip": 1.0461607,
      "balance_loss_mlp": 1.01852512,
      "epoch": 0.9339265315938196,
      "flos": 23221635482880.0,
      "grad_norm": 1.7987644488154866,
      "language_loss": 0.63457632,
      "learning_rate": 4.5548347260270236e-08,
      "loss": 0.6564573,
      "num_input_tokens_seen": 167901385,
      "step": 7767,
      "time_per_iteration": 2.4785733222961426
    },
    {
      "auxiliary_loss_clip": 0.01115186,
      "auxiliary_loss_mlp": 0.01024204,
      "balance_loss_clip": 1.04157865,
      "balance_loss_mlp": 1.01776338,
      "epoch": 0.9340467744844586,
      "flos": 22820261932800.0,
      "grad_norm": 1.966121539265018,
      "language_loss": 0.69391418,
      "learning_rate": 4.538319617564012e-08,
      "loss": 0.71530807,
      "num_input_tokens_seen": 167920405,
      "step": 7768,
      "time_per_iteration": 2.5658674240112305
    },
    {
      "auxiliary_loss_clip": 0.01132188,
      "auxiliary_loss_mlp": 0.01028042,
      "balance_loss_clip": 1.04178953,
      "balance_loss_mlp": 1.02092528,
      "epoch": 0.9341670173750977,
      "flos": 23660428026240.0,
      "grad_norm": 1.9478903965202232,
      "language_loss": 0.74476135,
      "learning_rate": 4.521834160482485e-08,
      "loss": 0.76636374,
      "num_input_tokens_seen": 167939145,
      "step": 7769,
      "time_per_iteration": 2.554039239883423
    },
    {
      "auxiliary_loss_clip": 0.01151828,
      "auxiliary_loss_mlp": 0.01024948,
      "balance_loss_clip": 1.04632258,
      "balance_loss_mlp": 1.0182904,
      "epoch": 0.9342872602657368,
      "flos": 24824256595200.0,
      "grad_norm": 1.520274059708202,
      "language_loss": 0.82096165,
      "learning_rate": 4.5053783572832846e-08,
      "loss": 0.84272939,
      "num_input_tokens_seen": 167959325,
      "step": 7770,
      "time_per_iteration": 2.519944667816162
    },
    {
      "auxiliary_loss_clip": 0.01149142,
      "auxiliary_loss_mlp": 0.01027543,
      "balance_loss_clip": 1.04608059,
      "balance_loss_mlp": 1.02027392,
      "epoch": 0.9344075031563759,
      "flos": 25771831332480.0,
      "grad_norm": 1.6241241863133782,
      "language_loss": 0.7635957,
      "learning_rate": 4.488952210462771e-08,
      "loss": 0.78536254,
      "num_input_tokens_seen": 167979530,
      "step": 7771,
      "time_per_iteration": 2.5844430923461914
    },
    {
      "auxiliary_loss_clip": 0.01162367,
      "auxiliary_loss_mlp": 0.01028155,
      "balance_loss_clip": 1.04822242,
      "balance_loss_mlp": 1.0211606,
      "epoch": 0.9345277460470149,
      "flos": 25551303782400.0,
      "grad_norm": 1.8383850076205175,
      "language_loss": 0.85252535,
      "learning_rate": 4.4725557225127495e-08,
      "loss": 0.87443054,
      "num_input_tokens_seen": 167997870,
      "step": 7772,
      "time_per_iteration": 2.484492778778076
    },
    {
      "auxiliary_loss_clip": 0.01148989,
      "auxiliary_loss_mlp": 0.01021634,
      "balance_loss_clip": 1.04506743,
      "balance_loss_mlp": 1.01507759,
      "epoch": 0.9346479889376541,
      "flos": 34313112432000.0,
      "grad_norm": 1.5573592336352793,
      "language_loss": 0.79641294,
      "learning_rate": 4.456188895920565e-08,
      "loss": 0.81811917,
      "num_input_tokens_seen": 168019625,
      "step": 7773,
      "time_per_iteration": 2.6160342693328857
    },
    {
      "auxiliary_loss_clip": 0.0116321,
      "auxiliary_loss_mlp": 0.01024626,
      "balance_loss_clip": 1.04747319,
      "balance_loss_mlp": 1.01747358,
      "epoch": 0.9347682318282932,
      "flos": 19093739581440.0,
      "grad_norm": 1.969806395182221,
      "language_loss": 0.85327303,
      "learning_rate": 4.439851733169031e-08,
      "loss": 0.8751514,
      "num_input_tokens_seen": 168037415,
      "step": 7774,
      "time_per_iteration": 2.422485113143921
    },
    {
      "auxiliary_loss_clip": 0.01121421,
      "auxiliary_loss_mlp": 0.01023331,
      "balance_loss_clip": 1.04379892,
      "balance_loss_mlp": 1.01656938,
      "epoch": 0.9348884747189322,
      "flos": 26249587153920.0,
      "grad_norm": 1.9282969018591154,
      "language_loss": 0.69955033,
      "learning_rate": 4.4235442367365204e-08,
      "loss": 0.72099781,
      "num_input_tokens_seen": 168057725,
      "step": 7775,
      "time_per_iteration": 2.622904062271118
    },
    {
      "auxiliary_loss_clip": 0.01130851,
      "auxiliary_loss_mlp": 0.01032055,
      "balance_loss_clip": 1.04132187,
      "balance_loss_mlp": 1.02499211,
      "epoch": 0.9350087176095714,
      "flos": 18333080242560.0,
      "grad_norm": 2.014559490300582,
      "language_loss": 0.78831863,
      "learning_rate": 4.4072664090968545e-08,
      "loss": 0.80994767,
      "num_input_tokens_seen": 168076110,
      "step": 7776,
      "time_per_iteration": 2.5083165168762207
    },
    {
      "auxiliary_loss_clip": 0.01135271,
      "auxiliary_loss_mlp": 0.01027865,
      "balance_loss_clip": 1.04197037,
      "balance_loss_mlp": 1.02069139,
      "epoch": 0.9351289605002104,
      "flos": 19318253541120.0,
      "grad_norm": 2.5132918775497615,
      "language_loss": 0.84829175,
      "learning_rate": 4.391018252719347e-08,
      "loss": 0.86992311,
      "num_input_tokens_seen": 168095905,
      "step": 7777,
      "time_per_iteration": 2.5433552265167236
    },
    {
      "auxiliary_loss_clip": 0.01137964,
      "auxiliary_loss_mlp": 0.01028407,
      "balance_loss_clip": 1.04346514,
      "balance_loss_mlp": 1.0212307,
      "epoch": 0.9352492033908495,
      "flos": 18799990156800.0,
      "grad_norm": 2.0894379620874024,
      "language_loss": 0.6897161,
      "learning_rate": 4.374799770068849e-08,
      "loss": 0.71137983,
      "num_input_tokens_seen": 168112580,
      "step": 7778,
      "time_per_iteration": 3.2365236282348633
    },
    {
      "auxiliary_loss_clip": 0.01146822,
      "auxiliary_loss_mlp": 0.01024442,
      "balance_loss_clip": 1.0470407,
      "balance_loss_mlp": 1.01725674,
      "epoch": 0.9353694462814887,
      "flos": 29530134241920.0,
      "grad_norm": 1.9290050966835304,
      "language_loss": 0.75001127,
      "learning_rate": 4.358610963605658e-08,
      "loss": 0.77172387,
      "num_input_tokens_seen": 168133030,
      "step": 7779,
      "time_per_iteration": 2.5390923023223877
    },
    {
      "auxiliary_loss_clip": 0.01164248,
      "auxiliary_loss_mlp": 0.0102886,
      "balance_loss_clip": 1.04797077,
      "balance_loss_mlp": 1.02139425,
      "epoch": 0.9354896891721277,
      "flos": 30665450390400.0,
      "grad_norm": 2.445966841185058,
      "language_loss": 0.6881392,
      "learning_rate": 4.342451835785677e-08,
      "loss": 0.71007031,
      "num_input_tokens_seen": 168153940,
      "step": 7780,
      "time_per_iteration": 3.327244281768799
    },
    {
      "auxiliary_loss_clip": 0.01133896,
      "auxiliary_loss_mlp": 0.010287,
      "balance_loss_clip": 1.04370999,
      "balance_loss_mlp": 1.02200031,
      "epoch": 0.9356099320627668,
      "flos": 19463907191040.0,
      "grad_norm": 1.5359818113044614,
      "language_loss": 0.74871731,
      "learning_rate": 4.3263223890601665e-08,
      "loss": 0.7703433,
      "num_input_tokens_seen": 168172650,
      "step": 7781,
      "time_per_iteration": 3.2248542308807373
    },
    {
      "auxiliary_loss_clip": 0.01145367,
      "auxiliary_loss_mlp": 0.00760088,
      "balance_loss_clip": 1.04758203,
      "balance_loss_mlp": 1.00029933,
      "epoch": 0.9357301749534058,
      "flos": 19098156954240.0,
      "grad_norm": 1.6981972886785448,
      "language_loss": 0.79575706,
      "learning_rate": 4.31022262587597e-08,
      "loss": 0.81481165,
      "num_input_tokens_seen": 168191325,
      "step": 7782,
      "time_per_iteration": 2.494405508041382
    },
    {
      "auxiliary_loss_clip": 0.01152227,
      "auxiliary_loss_mlp": 0.0103116,
      "balance_loss_clip": 1.04856575,
      "balance_loss_mlp": 1.02352166,
      "epoch": 0.935850417844045,
      "flos": 23550361776000.0,
      "grad_norm": 1.4704919420395046,
      "language_loss": 0.65921724,
      "learning_rate": 4.2941525486754225e-08,
      "loss": 0.68105119,
      "num_input_tokens_seen": 168211645,
      "step": 7783,
      "time_per_iteration": 2.5131821632385254
    },
    {
      "auxiliary_loss_clip": 0.01113249,
      "auxiliary_loss_mlp": 0.01021642,
      "balance_loss_clip": 1.04275298,
      "balance_loss_mlp": 1.01546443,
      "epoch": 0.935970660734684,
      "flos": 18588333265920.0,
      "grad_norm": 1.76058896685364,
      "language_loss": 0.79721928,
      "learning_rate": 4.278112159896286e-08,
      "loss": 0.81856823,
      "num_input_tokens_seen": 168229485,
      "step": 7784,
      "time_per_iteration": 2.547502040863037
    },
    {
      "auxiliary_loss_clip": 0.01126852,
      "auxiliary_loss_mlp": 0.01023794,
      "balance_loss_clip": 1.03972912,
      "balance_loss_mlp": 1.01749647,
      "epoch": 0.9360909036253231,
      "flos": 20631255292800.0,
      "grad_norm": 1.6983895071419197,
      "language_loss": 0.67569554,
      "learning_rate": 4.2621014619719896e-08,
      "loss": 0.69720203,
      "num_input_tokens_seen": 168247250,
      "step": 7785,
      "time_per_iteration": 2.520031690597534
    },
    {
      "auxiliary_loss_clip": 0.0103403,
      "auxiliary_loss_mlp": 0.01002512,
      "balance_loss_clip": 1.00812817,
      "balance_loss_mlp": 1.0014571,
      "epoch": 0.9362111465159623,
      "flos": 61791421052160.0,
      "grad_norm": 0.7169455460037275,
      "language_loss": 0.58633542,
      "learning_rate": 4.246120457331215e-08,
      "loss": 0.6067009,
      "num_input_tokens_seen": 168309425,
      "step": 7786,
      "time_per_iteration": 3.1474897861480713
    },
    {
      "auxiliary_loss_clip": 0.01128615,
      "auxiliary_loss_mlp": 0.01030183,
      "balance_loss_clip": 1.04542637,
      "balance_loss_mlp": 1.02298307,
      "epoch": 0.9363313894066013,
      "flos": 24170395368960.0,
      "grad_norm": 1.7682128259737324,
      "language_loss": 0.72013676,
      "learning_rate": 4.2301691483983325e-08,
      "loss": 0.74172473,
      "num_input_tokens_seen": 168329545,
      "step": 7787,
      "time_per_iteration": 2.558661699295044
    },
    {
      "auxiliary_loss_clip": 0.01153173,
      "auxiliary_loss_mlp": 0.01025515,
      "balance_loss_clip": 1.04620755,
      "balance_loss_mlp": 1.0186367,
      "epoch": 0.9364516322972404,
      "flos": 20120354196480.0,
      "grad_norm": 3.631844610563545,
      "language_loss": 0.76207221,
      "learning_rate": 4.214247537593163e-08,
      "loss": 0.78385907,
      "num_input_tokens_seen": 168348795,
      "step": 7788,
      "time_per_iteration": 2.4943387508392334
    },
    {
      "auxiliary_loss_clip": 0.0113694,
      "auxiliary_loss_mlp": 0.01022434,
      "balance_loss_clip": 1.04392338,
      "balance_loss_mlp": 1.01568115,
      "epoch": 0.9365718751878795,
      "flos": 20703758895360.0,
      "grad_norm": 1.776814523996822,
      "language_loss": 0.80460912,
      "learning_rate": 4.1983556273309293e-08,
      "loss": 0.82620281,
      "num_input_tokens_seen": 168367545,
      "step": 7789,
      "time_per_iteration": 2.529265880584717
    },
    {
      "auxiliary_loss_clip": 0.01166028,
      "auxiliary_loss_mlp": 0.01025241,
      "balance_loss_clip": 1.048877,
      "balance_loss_mlp": 1.01757574,
      "epoch": 0.9366921180785186,
      "flos": 18655270260480.0,
      "grad_norm": 3.1830713768131917,
      "language_loss": 0.68943739,
      "learning_rate": 4.182493420022526e-08,
      "loss": 0.71135008,
      "num_input_tokens_seen": 168383215,
      "step": 7790,
      "time_per_iteration": 2.416956901550293
    },
    {
      "auxiliary_loss_clip": 0.01115976,
      "auxiliary_loss_mlp": 0.01021112,
      "balance_loss_clip": 1.0414474,
      "balance_loss_mlp": 1.01491308,
      "epoch": 0.9368123609691577,
      "flos": 25774955815680.0,
      "grad_norm": 1.7148593750000773,
      "language_loss": 0.78417742,
      "learning_rate": 4.166660918074139e-08,
      "loss": 0.80554831,
      "num_input_tokens_seen": 168403120,
      "step": 7791,
      "time_per_iteration": 2.5783751010894775
    },
    {
      "auxiliary_loss_clip": 0.01116481,
      "auxiliary_loss_mlp": 0.01021417,
      "balance_loss_clip": 1.04184055,
      "balance_loss_mlp": 1.01449752,
      "epoch": 0.9369326038597968,
      "flos": 25553386771200.0,
      "grad_norm": 1.493905696052169,
      "language_loss": 0.73331589,
      "learning_rate": 4.15085812388758e-08,
      "loss": 0.75469494,
      "num_input_tokens_seen": 168425340,
      "step": 7792,
      "time_per_iteration": 2.5835413932800293
    },
    {
      "auxiliary_loss_clip": 0.01136269,
      "auxiliary_loss_mlp": 0.01026674,
      "balance_loss_clip": 1.04496646,
      "balance_loss_mlp": 1.01994455,
      "epoch": 0.9370528467504359,
      "flos": 23220019370880.0,
      "grad_norm": 1.6023713129318065,
      "language_loss": 0.78220081,
      "learning_rate": 4.135085039860153e-08,
      "loss": 0.80383027,
      "num_input_tokens_seen": 168444740,
      "step": 7793,
      "time_per_iteration": 3.37681245803833
    },
    {
      "auxiliary_loss_clip": 0.01134294,
      "auxiliary_loss_mlp": 0.01024299,
      "balance_loss_clip": 1.04663372,
      "balance_loss_mlp": 1.01717675,
      "epoch": 0.9371730896410749,
      "flos": 24967468120320.0,
      "grad_norm": 2.309138611134046,
      "language_loss": 0.78335238,
      "learning_rate": 4.1193416683845906e-08,
      "loss": 0.80493832,
      "num_input_tokens_seen": 168463670,
      "step": 7794,
      "time_per_iteration": 2.6191189289093018
    },
    {
      "auxiliary_loss_clip": 0.01126321,
      "auxiliary_loss_mlp": 0.01023852,
      "balance_loss_clip": 1.04601216,
      "balance_loss_mlp": 1.01756036,
      "epoch": 0.9372933325317141,
      "flos": 15553091134080.0,
      "grad_norm": 2.0939335935191274,
      "language_loss": 0.82850063,
      "learning_rate": 4.103628011849136e-08,
      "loss": 0.85000235,
      "num_input_tokens_seen": 168479030,
      "step": 7795,
      "time_per_iteration": 2.5286953449249268
    },
    {
      "auxiliary_loss_clip": 0.01138557,
      "auxiliary_loss_mlp": 0.01028461,
      "balance_loss_clip": 1.04751682,
      "balance_loss_mlp": 1.02120686,
      "epoch": 0.9374135754223532,
      "flos": 21871861182720.0,
      "grad_norm": 1.8988982234540235,
      "language_loss": 0.75823998,
      "learning_rate": 4.0879440726375506e-08,
      "loss": 0.77991021,
      "num_input_tokens_seen": 168496815,
      "step": 7796,
      "time_per_iteration": 2.5289549827575684
    },
    {
      "auxiliary_loss_clip": 0.01130055,
      "auxiliary_loss_mlp": 0.0102666,
      "balance_loss_clip": 1.03962898,
      "balance_loss_mlp": 1.0192275,
      "epoch": 0.9375338183129922,
      "flos": 22631048064000.0,
      "grad_norm": 2.2711319160270365,
      "language_loss": 0.55539215,
      "learning_rate": 4.0722898531291074e-08,
      "loss": 0.57695937,
      "num_input_tokens_seen": 168514055,
      "step": 7797,
      "time_per_iteration": 2.5229363441467285
    },
    {
      "auxiliary_loss_clip": 0.01139704,
      "auxiliary_loss_mlp": 0.0102228,
      "balance_loss_clip": 1.04481351,
      "balance_loss_mlp": 1.01534486,
      "epoch": 0.9376540612036314,
      "flos": 26104292640000.0,
      "grad_norm": 1.7805401920416644,
      "language_loss": 0.76766115,
      "learning_rate": 4.0566653556985295e-08,
      "loss": 0.78928101,
      "num_input_tokens_seen": 168534600,
      "step": 7798,
      "time_per_iteration": 2.556483268737793
    },
    {
      "auxiliary_loss_clip": 0.01084069,
      "auxiliary_loss_mlp": 0.01027195,
      "balance_loss_clip": 1.04017854,
      "balance_loss_mlp": 1.01968527,
      "epoch": 0.9377743040942704,
      "flos": 19717580016000.0,
      "grad_norm": 2.109982636322407,
      "language_loss": 0.81316531,
      "learning_rate": 4.0410705827159886e-08,
      "loss": 0.83427799,
      "num_input_tokens_seen": 168551895,
      "step": 7799,
      "time_per_iteration": 2.646326780319214
    },
    {
      "auxiliary_loss_clip": 0.01129781,
      "auxiliary_loss_mlp": 0.01024468,
      "balance_loss_clip": 1.04100442,
      "balance_loss_mlp": 1.01778305,
      "epoch": 0.9378945469849095,
      "flos": 15267530010240.0,
      "grad_norm": 2.1479696902140804,
      "language_loss": 0.70819843,
      "learning_rate": 4.0255055365472356e-08,
      "loss": 0.72974092,
      "num_input_tokens_seen": 168569990,
      "step": 7800,
      "time_per_iteration": 2.498863458633423
    },
    {
      "auxiliary_loss_clip": 0.01094414,
      "auxiliary_loss_mlp": 0.01025358,
      "balance_loss_clip": 1.03764534,
      "balance_loss_mlp": 1.01878345,
      "epoch": 0.9380147898755486,
      "flos": 20591394174720.0,
      "grad_norm": 2.093777265018202,
      "language_loss": 0.74845409,
      "learning_rate": 4.009970219553471e-08,
      "loss": 0.76965183,
      "num_input_tokens_seen": 168586940,
      "step": 7801,
      "time_per_iteration": 2.622087240219116
    },
    {
      "auxiliary_loss_clip": 0.01152881,
      "auxiliary_loss_mlp": 0.01028458,
      "balance_loss_clip": 1.0451833,
      "balance_loss_mlp": 1.02059937,
      "epoch": 0.9381350327661877,
      "flos": 26281116316800.0,
      "grad_norm": 2.387110436471645,
      "language_loss": 0.76505542,
      "learning_rate": 3.99446463409141e-08,
      "loss": 0.78686881,
      "num_input_tokens_seen": 168604795,
      "step": 7802,
      "time_per_iteration": 2.524853467941284
    },
    {
      "auxiliary_loss_clip": 0.01151843,
      "auxiliary_loss_mlp": 0.01026966,
      "balance_loss_clip": 1.04328144,
      "balance_loss_mlp": 1.0196408,
      "epoch": 0.9382552756568268,
      "flos": 23586344225280.0,
      "grad_norm": 2.225644310535053,
      "language_loss": 0.6921078,
      "learning_rate": 3.978988782513215e-08,
      "loss": 0.71389592,
      "num_input_tokens_seen": 168622290,
      "step": 7803,
      "time_per_iteration": 2.5081052780151367
    },
    {
      "auxiliary_loss_clip": 0.0115173,
      "auxiliary_loss_mlp": 0.01027586,
      "balance_loss_clip": 1.04629731,
      "balance_loss_mlp": 1.02048111,
      "epoch": 0.9383755185474659,
      "flos": 28438809275520.0,
      "grad_norm": 1.554936386355212,
      "language_loss": 0.76177895,
      "learning_rate": 3.963542667166586e-08,
      "loss": 0.78357214,
      "num_input_tokens_seen": 168642395,
      "step": 7804,
      "time_per_iteration": 3.3359320163726807
    },
    {
      "auxiliary_loss_clip": 0.01118788,
      "auxiliary_loss_mlp": 0.01026958,
      "balance_loss_clip": 1.04516661,
      "balance_loss_mlp": 1.02004123,
      "epoch": 0.938495761438105,
      "flos": 20449583280000.0,
      "grad_norm": 1.6070979509014245,
      "language_loss": 0.68308115,
      "learning_rate": 3.9481262903946486e-08,
      "loss": 0.70453864,
      "num_input_tokens_seen": 168661840,
      "step": 7805,
      "time_per_iteration": 2.556002616882324
    },
    {
      "auxiliary_loss_clip": 0.01019924,
      "auxiliary_loss_mlp": 0.01002138,
      "balance_loss_clip": 1.00959826,
      "balance_loss_mlp": 1.00095165,
      "epoch": 0.938616004328744,
      "flos": 69302711658240.0,
      "grad_norm": 0.7677515167106046,
      "language_loss": 0.54510033,
      "learning_rate": 3.932739654536066e-08,
      "loss": 0.56532097,
      "num_input_tokens_seen": 168724540,
      "step": 7806,
      "time_per_iteration": 3.9419124126434326
    },
    {
      "auxiliary_loss_clip": 0.01150038,
      "auxiliary_loss_mlp": 0.01028722,
      "balance_loss_clip": 1.04773951,
      "balance_loss_mlp": 1.02224946,
      "epoch": 0.9387362472193832,
      "flos": 18911636605440.0,
      "grad_norm": 2.3552810100565873,
      "language_loss": 0.74123824,
      "learning_rate": 3.917382761925014e-08,
      "loss": 0.76302576,
      "num_input_tokens_seen": 168740375,
      "step": 7807,
      "time_per_iteration": 3.168048858642578
    },
    {
      "auxiliary_loss_clip": 0.01146683,
      "auxiliary_loss_mlp": 0.01022821,
      "balance_loss_clip": 1.04670393,
      "balance_loss_mlp": 1.01608908,
      "epoch": 0.9388564901100223,
      "flos": 26501967089280.0,
      "grad_norm": 1.7801962176963786,
      "language_loss": 0.7928862,
      "learning_rate": 3.9020556148910754e-08,
      "loss": 0.81458127,
      "num_input_tokens_seen": 168759730,
      "step": 7808,
      "time_per_iteration": 2.5362441539764404
    },
    {
      "auxiliary_loss_clip": 0.01037689,
      "auxiliary_loss_mlp": 0.01002015,
      "balance_loss_clip": 1.008672,
      "balance_loss_mlp": 1.00088811,
      "epoch": 0.9389767330006613,
      "flos": 58941083157120.0,
      "grad_norm": 0.7112825180991138,
      "language_loss": 0.56715786,
      "learning_rate": 3.8867582157593895e-08,
      "loss": 0.58755481,
      "num_input_tokens_seen": 168813935,
      "step": 7809,
      "time_per_iteration": 2.9470691680908203
    },
    {
      "auxiliary_loss_clip": 0.01149256,
      "auxiliary_loss_mlp": 0.01029238,
      "balance_loss_clip": 1.0487318,
      "balance_loss_mlp": 1.02258277,
      "epoch": 0.9390969758913005,
      "flos": 31102554994560.0,
      "grad_norm": 1.56524852068423,
      "language_loss": 0.76501733,
      "learning_rate": 3.871490566850544e-08,
      "loss": 0.78680229,
      "num_input_tokens_seen": 168838145,
      "step": 7810,
      "time_per_iteration": 2.5800228118896484
    },
    {
      "auxiliary_loss_clip": 0.0112941,
      "auxiliary_loss_mlp": 0.01024016,
      "balance_loss_clip": 1.04333377,
      "balance_loss_mlp": 1.0173521,
      "epoch": 0.9392172187819395,
      "flos": 22419391173120.0,
      "grad_norm": 1.5443948146054773,
      "language_loss": 0.70739436,
      "learning_rate": 3.856252670480642e-08,
      "loss": 0.72892869,
      "num_input_tokens_seen": 168856805,
      "step": 7811,
      "time_per_iteration": 2.5255258083343506
    },
    {
      "auxiliary_loss_clip": 0.01129976,
      "auxiliary_loss_mlp": 0.01024129,
      "balance_loss_clip": 1.03986776,
      "balance_loss_mlp": 1.01770663,
      "epoch": 0.9393374616725786,
      "flos": 19719483436800.0,
      "grad_norm": 1.716715487317215,
      "language_loss": 0.81214583,
      "learning_rate": 3.841044528961279e-08,
      "loss": 0.83368689,
      "num_input_tokens_seen": 168874600,
      "step": 7812,
      "time_per_iteration": 2.5202105045318604
    },
    {
      "auxiliary_loss_clip": 0.01162719,
      "auxiliary_loss_mlp": 0.0102417,
      "balance_loss_clip": 1.04586053,
      "balance_loss_mlp": 1.01724422,
      "epoch": 0.9394577045632178,
      "flos": 24170215800960.0,
      "grad_norm": 1.7598535922979848,
      "language_loss": 0.78586698,
      "learning_rate": 3.825866144599477e-08,
      "loss": 0.80773586,
      "num_input_tokens_seen": 168893655,
      "step": 7813,
      "time_per_iteration": 2.484039545059204
    },
    {
      "auxiliary_loss_clip": 0.01132334,
      "auxiliary_loss_mlp": 0.01028508,
      "balance_loss_clip": 1.04174113,
      "balance_loss_mlp": 1.02148092,
      "epoch": 0.9395779474538568,
      "flos": 19023929498880.0,
      "grad_norm": 1.858198884327698,
      "language_loss": 0.75050503,
      "learning_rate": 3.8107175196978145e-08,
      "loss": 0.77211344,
      "num_input_tokens_seen": 168909960,
      "step": 7814,
      "time_per_iteration": 2.4888949394226074
    },
    {
      "auxiliary_loss_clip": 0.01119674,
      "auxiliary_loss_mlp": 0.01025018,
      "balance_loss_clip": 1.04362273,
      "balance_loss_mlp": 1.01836336,
      "epoch": 0.9396981903444959,
      "flos": 14319129260160.0,
      "grad_norm": 1.8088293081396876,
      "language_loss": 0.77055323,
      "learning_rate": 3.7955986565542996e-08,
      "loss": 0.79200017,
      "num_input_tokens_seen": 168928040,
      "step": 7815,
      "time_per_iteration": 2.5284740924835205
    },
    {
      "auxiliary_loss_clip": 0.01118838,
      "auxiliary_loss_mlp": 0.01022281,
      "balance_loss_clip": 1.04135907,
      "balance_loss_mlp": 1.01540899,
      "epoch": 0.9398184332351349,
      "flos": 34787564202240.0,
      "grad_norm": 2.0045543764491462,
      "language_loss": 0.68187559,
      "learning_rate": 3.780509557462497e-08,
      "loss": 0.70328677,
      "num_input_tokens_seen": 168948240,
      "step": 7816,
      "time_per_iteration": 2.6729462146759033
    },
    {
      "auxiliary_loss_clip": 0.01129724,
      "auxiliary_loss_mlp": 0.01026222,
      "balance_loss_clip": 1.04095507,
      "balance_loss_mlp": 1.01927173,
      "epoch": 0.9399386761257741,
      "flos": 25372253462400.0,
      "grad_norm": 1.5917918748338227,
      "language_loss": 0.75544906,
      "learning_rate": 3.765450224711375e-08,
      "loss": 0.77700853,
      "num_input_tokens_seen": 168968745,
      "step": 7817,
      "time_per_iteration": 2.5744264125823975
    },
    {
      "auxiliary_loss_clip": 0.01127533,
      "auxiliary_loss_mlp": 0.01025978,
      "balance_loss_clip": 1.04395938,
      "balance_loss_mlp": 1.01873863,
      "epoch": 0.9400589190164131,
      "flos": 27304965584640.0,
      "grad_norm": 1.5992913075300974,
      "language_loss": 0.79398316,
      "learning_rate": 3.750420660585396e-08,
      "loss": 0.81551826,
      "num_input_tokens_seen": 168990685,
      "step": 7818,
      "time_per_iteration": 2.562063694000244
    },
    {
      "auxiliary_loss_clip": 0.01162659,
      "auxiliary_loss_mlp": 0.01025244,
      "balance_loss_clip": 1.04797268,
      "balance_loss_mlp": 1.01876771,
      "epoch": 0.9401791619070522,
      "flos": 23399859790080.0,
      "grad_norm": 1.6110346793902586,
      "language_loss": 0.79321873,
      "learning_rate": 3.735420867364603e-08,
      "loss": 0.81509781,
      "num_input_tokens_seen": 169011665,
      "step": 7819,
      "time_per_iteration": 3.1933391094207764
    },
    {
      "auxiliary_loss_clip": 0.01084481,
      "auxiliary_loss_mlp": 0.01020883,
      "balance_loss_clip": 1.03594887,
      "balance_loss_mlp": 1.01472259,
      "epoch": 0.9402994047976914,
      "flos": 35881403120640.0,
      "grad_norm": 1.7754371540725589,
      "language_loss": 0.61849916,
      "learning_rate": 3.7204508473244186e-08,
      "loss": 0.63955283,
      "num_input_tokens_seen": 169035290,
      "step": 7820,
      "time_per_iteration": 2.745845317840576
    },
    {
      "auxiliary_loss_clip": 0.01072868,
      "auxiliary_loss_mlp": 0.01023915,
      "balance_loss_clip": 1.03962922,
      "balance_loss_mlp": 1.01719773,
      "epoch": 0.9404196476883304,
      "flos": 22236821320320.0,
      "grad_norm": 1.5925048918621816,
      "language_loss": 0.69172627,
      "learning_rate": 3.7055106027357395e-08,
      "loss": 0.71269417,
      "num_input_tokens_seen": 169055155,
      "step": 7821,
      "time_per_iteration": 2.7217726707458496
    },
    {
      "auxiliary_loss_clip": 0.01143672,
      "auxiliary_loss_mlp": 0.01024978,
      "balance_loss_clip": 1.04392219,
      "balance_loss_mlp": 1.01817143,
      "epoch": 0.9405398905789695,
      "flos": 18915802583040.0,
      "grad_norm": 2.260502111371493,
      "language_loss": 0.71852183,
      "learning_rate": 3.690600135865063e-08,
      "loss": 0.74020839,
      "num_input_tokens_seen": 169072080,
      "step": 7822,
      "time_per_iteration": 2.5672714710235596
    },
    {
      "auxiliary_loss_clip": 0.01010678,
      "auxiliary_loss_mlp": 0.00999595,
      "balance_loss_clip": 1.0077529,
      "balance_loss_mlp": 0.99851036,
      "epoch": 0.9406601334696086,
      "flos": 70274130048000.0,
      "grad_norm": 0.7933200997460268,
      "language_loss": 0.58128268,
      "learning_rate": 3.675719448974246e-08,
      "loss": 0.60138541,
      "num_input_tokens_seen": 169137170,
      "step": 7823,
      "time_per_iteration": 3.2249016761779785
    },
    {
      "auxiliary_loss_clip": 0.01102067,
      "auxiliary_loss_mlp": 0.00760849,
      "balance_loss_clip": 1.04041195,
      "balance_loss_mlp": 1.00031614,
      "epoch": 0.9407803763602477,
      "flos": 22165071903360.0,
      "grad_norm": 2.198367804024523,
      "language_loss": 0.59603286,
      "learning_rate": 3.6608685443207054e-08,
      "loss": 0.61466205,
      "num_input_tokens_seen": 169156320,
      "step": 7824,
      "time_per_iteration": 2.614950656890869
    },
    {
      "auxiliary_loss_clip": 0.01121774,
      "auxiliary_loss_mlp": 0.01026739,
      "balance_loss_clip": 1.04263377,
      "balance_loss_mlp": 1.02048326,
      "epoch": 0.9409006192508867,
      "flos": 18879496911360.0,
      "grad_norm": 2.332072710913994,
      "language_loss": 0.66374218,
      "learning_rate": 3.646047424157306e-08,
      "loss": 0.68522727,
      "num_input_tokens_seen": 169173295,
      "step": 7825,
      "time_per_iteration": 2.5535805225372314
    },
    {
      "auxiliary_loss_clip": 0.01137539,
      "auxiliary_loss_mlp": 0.01028693,
      "balance_loss_clip": 1.04681182,
      "balance_loss_mlp": 1.02114105,
      "epoch": 0.9410208621415259,
      "flos": 23368258800000.0,
      "grad_norm": 2.483121316754697,
      "language_loss": 0.68977094,
      "learning_rate": 3.631256090732382e-08,
      "loss": 0.71143329,
      "num_input_tokens_seen": 169193755,
      "step": 7826,
      "time_per_iteration": 2.533738851547241
    },
    {
      "auxiliary_loss_clip": 0.01121921,
      "auxiliary_loss_mlp": 0.01028158,
      "balance_loss_clip": 1.0443145,
      "balance_loss_mlp": 1.02128601,
      "epoch": 0.941141105032165,
      "flos": 22742227635840.0,
      "grad_norm": 1.7813442178698875,
      "language_loss": 0.82422233,
      "learning_rate": 3.6164945462897833e-08,
      "loss": 0.84572303,
      "num_input_tokens_seen": 169213045,
      "step": 7827,
      "time_per_iteration": 2.589723587036133
    },
    {
      "auxiliary_loss_clip": 0.01147385,
      "auxiliary_loss_mlp": 0.00760618,
      "balance_loss_clip": 1.04708707,
      "balance_loss_mlp": 1.00034666,
      "epoch": 0.941261347922804,
      "flos": 20704908130560.0,
      "grad_norm": 1.6799163954114396,
      "language_loss": 0.75916636,
      "learning_rate": 3.6017627930687856e-08,
      "loss": 0.7782464,
      "num_input_tokens_seen": 169232870,
      "step": 7828,
      "time_per_iteration": 2.4951610565185547
    },
    {
      "auxiliary_loss_clip": 0.01100999,
      "auxiliary_loss_mlp": 0.01024301,
      "balance_loss_clip": 1.03777122,
      "balance_loss_mlp": 1.01744914,
      "epoch": 0.9413815908134432,
      "flos": 19421998997760.0,
      "grad_norm": 1.9121151126637648,
      "language_loss": 0.77189875,
      "learning_rate": 3.587060833304267e-08,
      "loss": 0.79315174,
      "num_input_tokens_seen": 169251060,
      "step": 7829,
      "time_per_iteration": 2.597127676010132
    },
    {
      "auxiliary_loss_clip": 0.01153732,
      "auxiliary_loss_mlp": 0.01027121,
      "balance_loss_clip": 1.04806757,
      "balance_loss_mlp": 1.02020097,
      "epoch": 0.9415018337040822,
      "flos": 17493452853120.0,
      "grad_norm": 2.3592500473333713,
      "language_loss": 0.64244378,
      "learning_rate": 3.5723886692264225e-08,
      "loss": 0.66425234,
      "num_input_tokens_seen": 169268600,
      "step": 7830,
      "time_per_iteration": 3.2147905826568604
    },
    {
      "auxiliary_loss_clip": 0.0113429,
      "auxiliary_loss_mlp": 0.01025042,
      "balance_loss_clip": 1.04296708,
      "balance_loss_mlp": 1.01850986,
      "epoch": 0.9416220765947213,
      "flos": 31831613343360.0,
      "grad_norm": 9.858099993651397,
      "language_loss": 0.62082922,
      "learning_rate": 3.557746303061071e-08,
      "loss": 0.64242256,
      "num_input_tokens_seen": 169290355,
      "step": 7831,
      "time_per_iteration": 2.6179165840148926
    },
    {
      "auxiliary_loss_clip": 0.01132442,
      "auxiliary_loss_mlp": 0.01029814,
      "balance_loss_clip": 1.04235828,
      "balance_loss_mlp": 1.02281618,
      "epoch": 0.9417423194853605,
      "flos": 23511973115520.0,
      "grad_norm": 1.6024473452230574,
      "language_loss": 0.72426558,
      "learning_rate": 3.543133737029391e-08,
      "loss": 0.74588811,
      "num_input_tokens_seen": 169310865,
      "step": 7832,
      "time_per_iteration": 3.357239007949829
    },
    {
      "auxiliary_loss_clip": 0.01152462,
      "auxiliary_loss_mlp": 0.01027765,
      "balance_loss_clip": 1.04619014,
      "balance_loss_mlp": 1.02049661,
      "epoch": 0.9418625623759995,
      "flos": 23915106432000.0,
      "grad_norm": 1.6996353717873292,
      "language_loss": 0.69079733,
      "learning_rate": 3.5285509733481214e-08,
      "loss": 0.71259964,
      "num_input_tokens_seen": 169330590,
      "step": 7833,
      "time_per_iteration": 3.253981113433838
    },
    {
      "auxiliary_loss_clip": 0.01146853,
      "auxiliary_loss_mlp": 0.01027435,
      "balance_loss_clip": 1.04561555,
      "balance_loss_mlp": 1.02029443,
      "epoch": 0.9419828052666386,
      "flos": 18076965292800.0,
      "grad_norm": 1.7293718737511643,
      "language_loss": 0.76270008,
      "learning_rate": 3.513998014229469e-08,
      "loss": 0.7844429,
      "num_input_tokens_seen": 169349540,
      "step": 7834,
      "time_per_iteration": 2.4833109378814697
    },
    {
      "auxiliary_loss_clip": 0.01133056,
      "auxiliary_loss_mlp": 0.01027558,
      "balance_loss_clip": 1.04426491,
      "balance_loss_mlp": 1.02051544,
      "epoch": 0.9421030481572777,
      "flos": 17712328377600.0,
      "grad_norm": 2.088423072095952,
      "language_loss": 0.86368865,
      "learning_rate": 3.499474861881069e-08,
      "loss": 0.8852948,
      "num_input_tokens_seen": 169366765,
      "step": 7835,
      "time_per_iteration": 2.4863979816436768
    },
    {
      "auxiliary_loss_clip": 0.01091146,
      "auxiliary_loss_mlp": 0.01026594,
      "balance_loss_clip": 1.03993261,
      "balance_loss_mlp": 1.01995659,
      "epoch": 0.9422232910479168,
      "flos": 20194114775040.0,
      "grad_norm": 1.807713037251671,
      "language_loss": 0.68167871,
      "learning_rate": 3.4849815185061136e-08,
      "loss": 0.70285606,
      "num_input_tokens_seen": 169386655,
      "step": 7836,
      "time_per_iteration": 2.636234760284424
    },
    {
      "auxiliary_loss_clip": 0.01147066,
      "auxiliary_loss_mlp": 0.01025486,
      "balance_loss_clip": 1.04354715,
      "balance_loss_mlp": 1.01947236,
      "epoch": 0.9423435339385559,
      "flos": 18442571875200.0,
      "grad_norm": 2.239326485236803,
      "language_loss": 0.75779176,
      "learning_rate": 3.470517986303223e-08,
      "loss": 0.77951729,
      "num_input_tokens_seen": 169405640,
      "step": 7837,
      "time_per_iteration": 2.4704041481018066
    },
    {
      "auxiliary_loss_clip": 0.01121805,
      "auxiliary_loss_mlp": 0.01032637,
      "balance_loss_clip": 1.0446887,
      "balance_loss_mlp": 1.02550197,
      "epoch": 0.942463776829195,
      "flos": 20080636732800.0,
      "grad_norm": 1.6292308370657256,
      "language_loss": 0.78878438,
      "learning_rate": 3.4560842674664856e-08,
      "loss": 0.81032884,
      "num_input_tokens_seen": 169424155,
      "step": 7838,
      "time_per_iteration": 2.543653726577759
    },
    {
      "auxiliary_loss_clip": 0.01148672,
      "auxiliary_loss_mlp": 0.01025582,
      "balance_loss_clip": 1.0433445,
      "balance_loss_mlp": 1.01882291,
      "epoch": 0.9425840197198341,
      "flos": 22636255536000.0,
      "grad_norm": 1.7144492391606259,
      "language_loss": 0.75177646,
      "learning_rate": 3.441680364185506e-08,
      "loss": 0.77351904,
      "num_input_tokens_seen": 169444025,
      "step": 7839,
      "time_per_iteration": 2.5238759517669678
    },
    {
      "auxiliary_loss_clip": 0.0113727,
      "auxiliary_loss_mlp": 0.01032255,
      "balance_loss_clip": 1.04598498,
      "balance_loss_mlp": 1.02473855,
      "epoch": 0.9427042626104731,
      "flos": 19937892084480.0,
      "grad_norm": 2.5039899843402975,
      "language_loss": 0.7452696,
      "learning_rate": 3.427306278645314e-08,
      "loss": 0.76696479,
      "num_input_tokens_seen": 169462480,
      "step": 7840,
      "time_per_iteration": 2.5204057693481445
    },
    {
      "auxiliary_loss_clip": 0.01106898,
      "auxiliary_loss_mlp": 0.01024238,
      "balance_loss_clip": 1.04308295,
      "balance_loss_mlp": 1.01671553,
      "epoch": 0.9428245055011123,
      "flos": 22856998567680.0,
      "grad_norm": 4.664057847075759,
      "language_loss": 0.72968,
      "learning_rate": 3.4129620130264767e-08,
      "loss": 0.7509914,
      "num_input_tokens_seen": 169480840,
      "step": 7841,
      "time_per_iteration": 2.5876123905181885
    },
    {
      "auxiliary_loss_clip": 0.0113664,
      "auxiliary_loss_mlp": 0.00760306,
      "balance_loss_clip": 1.04321957,
      "balance_loss_mlp": 1.00029588,
      "epoch": 0.9429447483917514,
      "flos": 20951757371520.0,
      "grad_norm": 2.052812771655024,
      "language_loss": 0.77630752,
      "learning_rate": 3.398647569505009e-08,
      "loss": 0.795277,
      "num_input_tokens_seen": 169498265,
      "step": 7842,
      "time_per_iteration": 2.516140937805176
    },
    {
      "auxiliary_loss_clip": 0.01126465,
      "auxiliary_loss_mlp": 0.01026059,
      "balance_loss_clip": 1.04456389,
      "balance_loss_mlp": 1.01883483,
      "epoch": 0.9430649912823904,
      "flos": 18843658116480.0,
      "grad_norm": 2.2415034742228874,
      "language_loss": 0.75060242,
      "learning_rate": 3.384362950252373e-08,
      "loss": 0.77212763,
      "num_input_tokens_seen": 169515235,
      "step": 7843,
      "time_per_iteration": 2.5558602809906006
    },
    {
      "auxiliary_loss_clip": 0.01130706,
      "auxiliary_loss_mlp": 0.01026647,
      "balance_loss_clip": 1.04146719,
      "balance_loss_mlp": 1.01991749,
      "epoch": 0.9431852341730296,
      "flos": 32556038837760.0,
      "grad_norm": 2.4069058063997915,
      "language_loss": 0.57203591,
      "learning_rate": 3.3701081574355473e-08,
      "loss": 0.59360945,
      "num_input_tokens_seen": 169537195,
      "step": 7844,
      "time_per_iteration": 3.445439577102661
    },
    {
      "auxiliary_loss_clip": 0.01033911,
      "auxiliary_loss_mlp": 0.01003373,
      "balance_loss_clip": 1.0097481,
      "balance_loss_mlp": 1.00224018,
      "epoch": 0.9433054770636686,
      "flos": 66904490252160.0,
      "grad_norm": 0.6389987282097837,
      "language_loss": 0.51684117,
      "learning_rate": 3.3558831932169796e-08,
      "loss": 0.53721404,
      "num_input_tokens_seen": 169605865,
      "step": 7845,
      "time_per_iteration": 3.189997434616089
    },
    {
      "auxiliary_loss_clip": 0.01147425,
      "auxiliary_loss_mlp": 0.01022977,
      "balance_loss_clip": 1.04477513,
      "balance_loss_mlp": 1.01616096,
      "epoch": 0.9434257199543077,
      "flos": 26140346916480.0,
      "grad_norm": 2.1358149765852743,
      "language_loss": 0.88235438,
      "learning_rate": 3.341688059754588e-08,
      "loss": 0.90405846,
      "num_input_tokens_seen": 169621520,
      "step": 7846,
      "time_per_iteration": 2.542449712753296
    },
    {
      "auxiliary_loss_clip": 0.01125119,
      "auxiliary_loss_mlp": 0.00760281,
      "balance_loss_clip": 1.04115915,
      "balance_loss_mlp": 1.00030899,
      "epoch": 0.9435459628449467,
      "flos": 25003486483200.0,
      "grad_norm": 3.133036843827062,
      "language_loss": 0.78208011,
      "learning_rate": 3.327522759201762e-08,
      "loss": 0.80093408,
      "num_input_tokens_seen": 169641390,
      "step": 7847,
      "time_per_iteration": 2.6135764122009277
    },
    {
      "auxiliary_loss_clip": 0.01118445,
      "auxiliary_loss_mlp": 0.01028386,
      "balance_loss_clip": 1.04257226,
      "balance_loss_mlp": 1.02132905,
      "epoch": 0.9436662057355859,
      "flos": 22163240309760.0,
      "grad_norm": 2.0289915753863257,
      "language_loss": 0.66874892,
      "learning_rate": 3.313387293707359e-08,
      "loss": 0.6902172,
      "num_input_tokens_seen": 169660095,
      "step": 7848,
      "time_per_iteration": 2.5888869762420654
    },
    {
      "auxiliary_loss_clip": 0.01116694,
      "auxiliary_loss_mlp": 0.01025715,
      "balance_loss_clip": 1.04335523,
      "balance_loss_mlp": 1.01841009,
      "epoch": 0.943786448626225,
      "flos": 20118522602880.0,
      "grad_norm": 4.414891000974369,
      "language_loss": 0.68271494,
      "learning_rate": 3.29928166541571e-08,
      "loss": 0.70413899,
      "num_input_tokens_seen": 169679050,
      "step": 7849,
      "time_per_iteration": 2.554841995239258
    },
    {
      "auxiliary_loss_clip": 0.01125903,
      "auxiliary_loss_mlp": 0.01025057,
      "balance_loss_clip": 1.04370999,
      "balance_loss_mlp": 1.01821113,
      "epoch": 0.943906691516864,
      "flos": 22090808534400.0,
      "grad_norm": 1.8408814622775624,
      "language_loss": 0.80458856,
      "learning_rate": 3.2852058764666346e-08,
      "loss": 0.82609808,
      "num_input_tokens_seen": 169698150,
      "step": 7850,
      "time_per_iteration": 2.5464818477630615
    },
    {
      "auxiliary_loss_clip": 0.0111204,
      "auxiliary_loss_mlp": 0.01025774,
      "balance_loss_clip": 1.04529166,
      "balance_loss_mlp": 1.0195483,
      "epoch": 0.9440269344075032,
      "flos": 35298501212160.0,
      "grad_norm": 2.1687854063073244,
      "language_loss": 0.68360722,
      "learning_rate": 3.2711599289954264e-08,
      "loss": 0.70498538,
      "num_input_tokens_seen": 169722185,
      "step": 7851,
      "time_per_iteration": 2.6829347610473633
    },
    {
      "auxiliary_loss_clip": 0.01094991,
      "auxiliary_loss_mlp": 0.01027016,
      "balance_loss_clip": 1.04180741,
      "balance_loss_mlp": 1.02002478,
      "epoch": 0.9441471772981422,
      "flos": 19238136255360.0,
      "grad_norm": 1.6917853199192072,
      "language_loss": 0.77689737,
      "learning_rate": 3.257143825132847e-08,
      "loss": 0.79811746,
      "num_input_tokens_seen": 169740355,
      "step": 7852,
      "time_per_iteration": 2.6205673217773438
    },
    {
      "auxiliary_loss_clip": 0.01134252,
      "auxiliary_loss_mlp": 0.0102459,
      "balance_loss_clip": 1.04340935,
      "balance_loss_mlp": 1.01831412,
      "epoch": 0.9442674201887813,
      "flos": 25739799379200.0,
      "grad_norm": 1.8382499479510535,
      "language_loss": 0.75969118,
      "learning_rate": 3.243157567005106e-08,
      "loss": 0.78127962,
      "num_input_tokens_seen": 169758535,
      "step": 7853,
      "time_per_iteration": 2.5508577823638916
    },
    {
      "auxiliary_loss_clip": 0.01169893,
      "auxiliary_loss_mlp": 0.01032921,
      "balance_loss_clip": 1.0530479,
      "balance_loss_mlp": 1.0257417,
      "epoch": 0.9443876630794205,
      "flos": 15523321737600.0,
      "grad_norm": 2.0899974014911034,
      "language_loss": 0.63829231,
      "learning_rate": 3.2292011567339296e-08,
      "loss": 0.6603204,
      "num_input_tokens_seen": 169776340,
      "step": 7854,
      "time_per_iteration": 2.4419314861297607
    },
    {
      "auxiliary_loss_clip": 0.01148638,
      "auxiliary_loss_mlp": 0.00760342,
      "balance_loss_clip": 1.04454553,
      "balance_loss_mlp": 1.00033689,
      "epoch": 0.9445079059700595,
      "flos": 13400821128960.0,
      "grad_norm": 2.1562052841905106,
      "language_loss": 0.55630159,
      "learning_rate": 3.21527459643649e-08,
      "loss": 0.57539141,
      "num_input_tokens_seen": 169793225,
      "step": 7855,
      "time_per_iteration": 3.3025741577148438
    },
    {
      "auxiliary_loss_clip": 0.01151573,
      "auxiliary_loss_mlp": 0.01025829,
      "balance_loss_clip": 1.04673398,
      "balance_loss_mlp": 1.01954961,
      "epoch": 0.9446281488606986,
      "flos": 23659242877440.0,
      "grad_norm": 1.742933173610097,
      "language_loss": 0.74262267,
      "learning_rate": 3.2013778882254536e-08,
      "loss": 0.76439673,
      "num_input_tokens_seen": 169812020,
      "step": 7856,
      "time_per_iteration": 2.516793727874756
    },
    {
      "auxiliary_loss_clip": 0.0113971,
      "auxiliary_loss_mlp": 0.01030192,
      "balance_loss_clip": 1.04354227,
      "balance_loss_mlp": 1.02341247,
      "epoch": 0.9447483917513377,
      "flos": 25557337267200.0,
      "grad_norm": 1.7741789508996997,
      "language_loss": 0.76081747,
      "learning_rate": 3.1875110342088676e-08,
      "loss": 0.78251648,
      "num_input_tokens_seen": 169833470,
      "step": 7857,
      "time_per_iteration": 3.3104164600372314
    },
    {
      "auxiliary_loss_clip": 0.01129899,
      "auxiliary_loss_mlp": 0.01023114,
      "balance_loss_clip": 1.04460156,
      "balance_loss_mlp": 1.01666522,
      "epoch": 0.9448686346419768,
      "flos": 24535463247360.0,
      "grad_norm": 1.699922696872583,
      "language_loss": 0.65175974,
      "learning_rate": 3.1736740364904035e-08,
      "loss": 0.6732899,
      "num_input_tokens_seen": 169854000,
      "step": 7858,
      "time_per_iteration": 2.5478928089141846
    },
    {
      "auxiliary_loss_clip": 0.01104605,
      "auxiliary_loss_mlp": 0.00760612,
      "balance_loss_clip": 1.04174995,
      "balance_loss_mlp": 1.00031769,
      "epoch": 0.9449888775326158,
      "flos": 14721256995840.0,
      "grad_norm": 2.263865333191504,
      "language_loss": 0.77223802,
      "learning_rate": 3.159866897169094e-08,
      "loss": 0.79089022,
      "num_input_tokens_seen": 169872200,
      "step": 7859,
      "time_per_iteration": 2.5768275260925293
    },
    {
      "auxiliary_loss_clip": 0.01125341,
      "auxiliary_loss_mlp": 0.01030128,
      "balance_loss_clip": 1.04384315,
      "balance_loss_mlp": 1.02251673,
      "epoch": 0.945109120423255,
      "flos": 15447873219840.0,
      "grad_norm": 1.8162390693322277,
      "language_loss": 0.75274599,
      "learning_rate": 3.146089618339487e-08,
      "loss": 0.77430069,
      "num_input_tokens_seen": 169889055,
      "step": 7860,
      "time_per_iteration": 3.286191701889038
    },
    {
      "auxiliary_loss_clip": 0.01121783,
      "auxiliary_loss_mlp": 0.01024994,
      "balance_loss_clip": 1.0420121,
      "balance_loss_mlp": 1.01776361,
      "epoch": 0.9452293633138941,
      "flos": 25448097029760.0,
      "grad_norm": 1.7580521866105308,
      "language_loss": 0.67982447,
      "learning_rate": 3.132342202091554e-08,
      "loss": 0.70129228,
      "num_input_tokens_seen": 169909280,
      "step": 7861,
      "time_per_iteration": 2.604379177093506
    },
    {
      "auxiliary_loss_clip": 0.01163072,
      "auxiliary_loss_mlp": 0.01023795,
      "balance_loss_clip": 1.04587913,
      "balance_loss_mlp": 1.01689577,
      "epoch": 0.9453496062045331,
      "flos": 21215342350080.0,
      "grad_norm": 2.3163921762467674,
      "language_loss": 0.68567705,
      "learning_rate": 3.1186246505107595e-08,
      "loss": 0.7075457,
      "num_input_tokens_seen": 169928420,
      "step": 7862,
      "time_per_iteration": 2.465827465057373
    },
    {
      "auxiliary_loss_clip": 0.01151191,
      "auxiliary_loss_mlp": 0.01023407,
      "balance_loss_clip": 1.04845119,
      "balance_loss_mlp": 1.01624215,
      "epoch": 0.9454698490951723,
      "flos": 20010898477440.0,
      "grad_norm": 1.8715530446747268,
      "language_loss": 0.83447945,
      "learning_rate": 3.104936965678084e-08,
      "loss": 0.85622549,
      "num_input_tokens_seen": 169946750,
      "step": 7863,
      "time_per_iteration": 2.476027250289917
    },
    {
      "auxiliary_loss_clip": 0.0114921,
      "auxiliary_loss_mlp": 0.01025575,
      "balance_loss_clip": 1.04488873,
      "balance_loss_mlp": 1.01803815,
      "epoch": 0.9455900919858113,
      "flos": 21069652786560.0,
      "grad_norm": 1.7879559228512736,
      "language_loss": 0.81843674,
      "learning_rate": 3.091279149669956e-08,
      "loss": 0.84018463,
      "num_input_tokens_seen": 169965540,
      "step": 7864,
      "time_per_iteration": 2.500020742416382
    },
    {
      "auxiliary_loss_clip": 0.01147356,
      "auxiliary_loss_mlp": 0.00760474,
      "balance_loss_clip": 1.04412127,
      "balance_loss_mlp": 1.00026524,
      "epoch": 0.9457103348764504,
      "flos": 20740854666240.0,
      "grad_norm": 1.9315327949157024,
      "language_loss": 0.73934829,
      "learning_rate": 3.0776512045581624e-08,
      "loss": 0.75842655,
      "num_input_tokens_seen": 169984330,
      "step": 7865,
      "time_per_iteration": 2.505927562713623
    },
    {
      "auxiliary_loss_clip": 0.01128865,
      "auxiliary_loss_mlp": 0.01025206,
      "balance_loss_clip": 1.04384089,
      "balance_loss_mlp": 1.01733851,
      "epoch": 0.9458305777670896,
      "flos": 21428363957760.0,
      "grad_norm": 1.8808828668393796,
      "language_loss": 0.77449566,
      "learning_rate": 3.0640531324101384e-08,
      "loss": 0.79603636,
      "num_input_tokens_seen": 170002095,
      "step": 7866,
      "time_per_iteration": 2.5965847969055176
    },
    {
      "auxiliary_loss_clip": 0.01153507,
      "auxiliary_loss_mlp": 0.01025691,
      "balance_loss_clip": 1.04923129,
      "balance_loss_mlp": 1.01853824,
      "epoch": 0.9459508206577286,
      "flos": 20011185786240.0,
      "grad_norm": 1.530999888074661,
      "language_loss": 0.76049536,
      "learning_rate": 3.0504849352886554e-08,
      "loss": 0.78228736,
      "num_input_tokens_seen": 170020240,
      "step": 7867,
      "time_per_iteration": 2.492549180984497
    },
    {
      "auxiliary_loss_clip": 0.01149181,
      "auxiliary_loss_mlp": 0.01025617,
      "balance_loss_clip": 1.04730916,
      "balance_loss_mlp": 1.01850605,
      "epoch": 0.9460710635483677,
      "flos": 12166428291840.0,
      "grad_norm": 2.473424405570833,
      "language_loss": 0.7173934,
      "learning_rate": 3.036946615252023e-08,
      "loss": 0.73914135,
      "num_input_tokens_seen": 170035770,
      "step": 7868,
      "time_per_iteration": 2.4889395236968994
    },
    {
      "auxiliary_loss_clip": 0.01136824,
      "auxiliary_loss_mlp": 0.01033417,
      "balance_loss_clip": 1.04531407,
      "balance_loss_mlp": 1.02638626,
      "epoch": 0.9461913064390068,
      "flos": 34276196229120.0,
      "grad_norm": 3.7906350228915198,
      "language_loss": 0.67041469,
      "learning_rate": 3.0234381743539984e-08,
      "loss": 0.69211709,
      "num_input_tokens_seen": 170053385,
      "step": 7869,
      "time_per_iteration": 2.616807699203491
    },
    {
      "auxiliary_loss_clip": 0.01140782,
      "auxiliary_loss_mlp": 0.01027509,
      "balance_loss_clip": 1.0441072,
      "balance_loss_mlp": 1.02090776,
      "epoch": 0.9463115493296459,
      "flos": 19463763536640.0,
      "grad_norm": 2.0142482925192153,
      "language_loss": 0.79791063,
      "learning_rate": 3.0099596146437863e-08,
      "loss": 0.81959355,
      "num_input_tokens_seen": 170070490,
      "step": 7870,
      "time_per_iteration": 3.2165918350219727
    },
    {
      "auxiliary_loss_clip": 0.0105463,
      "auxiliary_loss_mlp": 0.01003958,
      "balance_loss_clip": 1.00776398,
      "balance_loss_mlp": 1.00284922,
      "epoch": 0.946431792220285,
      "flos": 70570824387840.0,
      "grad_norm": 0.7697381012311698,
      "language_loss": 0.60093939,
      "learning_rate": 2.996510938166086e-08,
      "loss": 0.62152529,
      "num_input_tokens_seen": 170133465,
      "step": 7871,
      "time_per_iteration": 3.132199287414551
    },
    {
      "auxiliary_loss_clip": 0.01147081,
      "auxiliary_loss_mlp": 0.01034082,
      "balance_loss_clip": 1.04651105,
      "balance_loss_mlp": 1.02726912,
      "epoch": 0.9465520351109241,
      "flos": 18947906363520.0,
      "grad_norm": 1.8024686988521683,
      "language_loss": 0.73806316,
      "learning_rate": 2.983092146960997e-08,
      "loss": 0.75987482,
      "num_input_tokens_seen": 170150810,
      "step": 7872,
      "time_per_iteration": 2.486224889755249
    },
    {
      "auxiliary_loss_clip": 0.01135064,
      "auxiliary_loss_mlp": 0.01027304,
      "balance_loss_clip": 1.04281354,
      "balance_loss_mlp": 1.02004457,
      "epoch": 0.9466722780015632,
      "flos": 19135647774720.0,
      "grad_norm": 2.084563397428503,
      "language_loss": 0.80198222,
      "learning_rate": 2.9697032430642256e-08,
      "loss": 0.8236059,
      "num_input_tokens_seen": 170169025,
      "step": 7873,
      "time_per_iteration": 2.5300962924957275
    },
    {
      "auxiliary_loss_clip": 0.01157573,
      "auxiliary_loss_mlp": 0.01025345,
      "balance_loss_clip": 1.04597378,
      "balance_loss_mlp": 1.01848495,
      "epoch": 0.9467925208922022,
      "flos": 17237912520960.0,
      "grad_norm": 2.0821897676764998,
      "language_loss": 0.73474622,
      "learning_rate": 2.9563442285067906e-08,
      "loss": 0.75657535,
      "num_input_tokens_seen": 170186070,
      "step": 7874,
      "time_per_iteration": 2.42889404296875
    },
    {
      "auxiliary_loss_clip": 0.0115291,
      "auxiliary_loss_mlp": 0.01024495,
      "balance_loss_clip": 1.04913282,
      "balance_loss_mlp": 1.01691651,
      "epoch": 0.9469127637828414,
      "flos": 29169016859520.0,
      "grad_norm": 1.78074238174984,
      "language_loss": 0.79711008,
      "learning_rate": 2.943015105315294e-08,
      "loss": 0.81888413,
      "num_input_tokens_seen": 170206265,
      "step": 7875,
      "time_per_iteration": 2.5479679107666016
    },
    {
      "auxiliary_loss_clip": 0.01109656,
      "auxiliary_loss_mlp": 0.01031744,
      "balance_loss_clip": 1.0415678,
      "balance_loss_mlp": 1.02467251,
      "epoch": 0.9470330066734804,
      "flos": 26030460234240.0,
      "grad_norm": 2.2084693325018487,
      "language_loss": 0.66389579,
      "learning_rate": 2.929715875511718e-08,
      "loss": 0.68530977,
      "num_input_tokens_seen": 170225300,
      "step": 7876,
      "time_per_iteration": 2.6138758659362793
    },
    {
      "auxiliary_loss_clip": 0.01148309,
      "auxiliary_loss_mlp": 0.01031402,
      "balance_loss_clip": 1.04210925,
      "balance_loss_mlp": 1.0243361,
      "epoch": 0.9471532495641195,
      "flos": 23440906056960.0,
      "grad_norm": 1.7670058736927796,
      "language_loss": 0.6969589,
      "learning_rate": 2.9164465411135375e-08,
      "loss": 0.71875596,
      "num_input_tokens_seen": 170245070,
      "step": 7877,
      "time_per_iteration": 2.505213975906372
    },
    {
      "auxiliary_loss_clip": 0.01151732,
      "auxiliary_loss_mlp": 0.01029083,
      "balance_loss_clip": 1.04932225,
      "balance_loss_mlp": 1.02222288,
      "epoch": 0.9472734924547586,
      "flos": 15815850099840.0,
      "grad_norm": 1.8234716944687261,
      "language_loss": 0.80500925,
      "learning_rate": 2.9032071041337426e-08,
      "loss": 0.82681739,
      "num_input_tokens_seen": 170263305,
      "step": 7878,
      "time_per_iteration": 2.4810664653778076
    },
    {
      "auxiliary_loss_clip": 0.01128224,
      "auxiliary_loss_mlp": 0.01024633,
      "balance_loss_clip": 1.04427028,
      "balance_loss_mlp": 1.01792192,
      "epoch": 0.9473937353453977,
      "flos": 11181793697280.0,
      "grad_norm": 1.5489906720755573,
      "language_loss": 0.72767842,
      "learning_rate": 2.889997566580704e-08,
      "loss": 0.74920702,
      "num_input_tokens_seen": 170281460,
      "step": 7879,
      "time_per_iteration": 2.505730628967285
    },
    {
      "auxiliary_loss_clip": 0.01165359,
      "auxiliary_loss_mlp": 0.01028827,
      "balance_loss_clip": 1.04741037,
      "balance_loss_mlp": 1.02155268,
      "epoch": 0.9475139782360368,
      "flos": 25775530433280.0,
      "grad_norm": 2.2171724764294285,
      "language_loss": 0.70444131,
      "learning_rate": 2.8768179304583086e-08,
      "loss": 0.72638321,
      "num_input_tokens_seen": 170303515,
      "step": 7880,
      "time_per_iteration": 2.4875340461730957
    },
    {
      "auxiliary_loss_clip": 0.01123962,
      "auxiliary_loss_mlp": 0.0102615,
      "balance_loss_clip": 1.04714239,
      "balance_loss_mlp": 1.01964974,
      "epoch": 0.9476342211266758,
      "flos": 22820046451200.0,
      "grad_norm": 1.528788223455236,
      "language_loss": 0.73573351,
      "learning_rate": 2.8636681977659117e-08,
      "loss": 0.75723463,
      "num_input_tokens_seen": 170323165,
      "step": 7881,
      "time_per_iteration": 2.5657408237457275
    },
    {
      "auxiliary_loss_clip": 0.01104558,
      "auxiliary_loss_mlp": 0.01031958,
      "balance_loss_clip": 1.04321361,
      "balance_loss_mlp": 1.02509737,
      "epoch": 0.947754464017315,
      "flos": 20193611984640.0,
      "grad_norm": 1.9706916207313943,
      "language_loss": 0.78007978,
      "learning_rate": 2.850548370498318e-08,
      "loss": 0.80144489,
      "num_input_tokens_seen": 170341005,
      "step": 7882,
      "time_per_iteration": 3.3997416496276855
    },
    {
      "auxiliary_loss_clip": 0.01145991,
      "auxiliary_loss_mlp": 0.01023069,
      "balance_loss_clip": 1.04284871,
      "balance_loss_mlp": 1.01674843,
      "epoch": 0.9478747069079541,
      "flos": 24717925359360.0,
      "grad_norm": 1.6178602827214181,
      "language_loss": 0.71245843,
      "learning_rate": 2.8374584506457798e-08,
      "loss": 0.73414904,
      "num_input_tokens_seen": 170362280,
      "step": 7883,
      "time_per_iteration": 3.321948528289795
    },
    {
      "auxiliary_loss_clip": 0.01135068,
      "auxiliary_loss_mlp": 0.01027144,
      "balance_loss_clip": 1.04606175,
      "balance_loss_mlp": 1.01997328,
      "epoch": 0.9479949497985931,
      "flos": 21361355136000.0,
      "grad_norm": 2.068504587678788,
      "language_loss": 0.67222464,
      "learning_rate": 2.824398440193998e-08,
      "loss": 0.69384676,
      "num_input_tokens_seen": 170381080,
      "step": 7884,
      "time_per_iteration": 2.53171706199646
    },
    {
      "auxiliary_loss_clip": 0.01099065,
      "auxiliary_loss_mlp": 0.01024462,
      "balance_loss_clip": 1.03951693,
      "balance_loss_mlp": 1.0173986,
      "epoch": 0.9481151926892323,
      "flos": 18148606968960.0,
      "grad_norm": 1.9560648701428318,
      "language_loss": 0.71251249,
      "learning_rate": 2.811368341124232e-08,
      "loss": 0.73374772,
      "num_input_tokens_seen": 170400150,
      "step": 7885,
      "time_per_iteration": 3.342376470565796
    },
    {
      "auxiliary_loss_clip": 0.01134411,
      "auxiliary_loss_mlp": 0.01022721,
      "balance_loss_clip": 1.04384851,
      "balance_loss_mlp": 1.01624227,
      "epoch": 0.9482354355798713,
      "flos": 22128012046080.0,
      "grad_norm": 2.0136728404733315,
      "language_loss": 0.67911029,
      "learning_rate": 2.7983681554131222e-08,
      "loss": 0.70068157,
      "num_input_tokens_seen": 170420410,
      "step": 7886,
      "time_per_iteration": 2.5555262565612793
    },
    {
      "auxiliary_loss_clip": 0.01131899,
      "auxiliary_loss_mlp": 0.01027411,
      "balance_loss_clip": 1.04265809,
      "balance_loss_mlp": 1.02030635,
      "epoch": 0.9483556784705104,
      "flos": 19063072344960.0,
      "grad_norm": 2.0583865246900537,
      "language_loss": 0.7023648,
      "learning_rate": 2.7853978850327365e-08,
      "loss": 0.72395796,
      "num_input_tokens_seen": 170439580,
      "step": 7887,
      "time_per_iteration": 2.5009889602661133
    },
    {
      "auxiliary_loss_clip": 0.01120222,
      "auxiliary_loss_mlp": 0.01027582,
      "balance_loss_clip": 1.04580867,
      "balance_loss_mlp": 1.02137113,
      "epoch": 0.9484759213611496,
      "flos": 25777110631680.0,
      "grad_norm": 1.836767995519403,
      "language_loss": 0.87437701,
      "learning_rate": 2.7724575319507225e-08,
      "loss": 0.89585507,
      "num_input_tokens_seen": 170459290,
      "step": 7888,
      "time_per_iteration": 2.6226418018341064
    },
    {
      "auxiliary_loss_clip": 0.01148595,
      "auxiliary_loss_mlp": 0.01028096,
      "balance_loss_clip": 1.04393983,
      "balance_loss_mlp": 1.02165556,
      "epoch": 0.9485961642517886,
      "flos": 20667740532480.0,
      "grad_norm": 1.7593854987719415,
      "language_loss": 0.76858532,
      "learning_rate": 2.759547098130044e-08,
      "loss": 0.79035223,
      "num_input_tokens_seen": 170478020,
      "step": 7889,
      "time_per_iteration": 2.5076053142547607
    },
    {
      "auxiliary_loss_clip": 0.01161402,
      "auxiliary_loss_mlp": 0.01026061,
      "balance_loss_clip": 1.04713941,
      "balance_loss_mlp": 1.01995742,
      "epoch": 0.9487164071424277,
      "flos": 22674069578880.0,
      "grad_norm": 1.920853285690736,
      "language_loss": 0.76929057,
      "learning_rate": 2.746666585529267e-08,
      "loss": 0.79116517,
      "num_input_tokens_seen": 170498295,
      "step": 7890,
      "time_per_iteration": 2.4890506267547607
    },
    {
      "auxiliary_loss_clip": 0.0114209,
      "auxiliary_loss_mlp": 0.01022091,
      "balance_loss_clip": 1.04433906,
      "balance_loss_mlp": 1.01498318,
      "epoch": 0.9488366500330668,
      "flos": 38726461716480.0,
      "grad_norm": 2.074887690392293,
      "language_loss": 0.73870969,
      "learning_rate": 2.73381599610234e-08,
      "loss": 0.76035148,
      "num_input_tokens_seen": 170518695,
      "step": 7891,
      "time_per_iteration": 2.639042854309082
    },
    {
      "auxiliary_loss_clip": 0.01143165,
      "auxiliary_loss_mlp": 0.01024072,
      "balance_loss_clip": 1.04269743,
      "balance_loss_mlp": 1.01741409,
      "epoch": 0.9489568929237059,
      "flos": 27890920149120.0,
      "grad_norm": 1.9831387733387673,
      "language_loss": 0.71669292,
      "learning_rate": 2.7209953317987033e-08,
      "loss": 0.73836529,
      "num_input_tokens_seen": 170539735,
      "step": 7892,
      "time_per_iteration": 2.5367581844329834
    },
    {
      "auxiliary_loss_clip": 0.01149749,
      "auxiliary_loss_mlp": 0.01020216,
      "balance_loss_clip": 1.04625344,
      "balance_loss_mlp": 1.01309597,
      "epoch": 0.9490771358143449,
      "flos": 33580642291200.0,
      "grad_norm": 1.8415623170070479,
      "language_loss": 0.78201765,
      "learning_rate": 2.7082045945631793e-08,
      "loss": 0.80371726,
      "num_input_tokens_seen": 170561950,
      "step": 7893,
      "time_per_iteration": 2.5902230739593506
    },
    {
      "auxiliary_loss_clip": 0.01115048,
      "auxiliary_loss_mlp": 0.01026976,
      "balance_loss_clip": 1.04270387,
      "balance_loss_mlp": 1.02004993,
      "epoch": 0.9491973787049841,
      "flos": 14793796512000.0,
      "grad_norm": 2.0322578596917427,
      "language_loss": 0.69408488,
      "learning_rate": 2.6954437863361712e-08,
      "loss": 0.71550512,
      "num_input_tokens_seen": 170579865,
      "step": 7894,
      "time_per_iteration": 2.533811092376709
    },
    {
      "auxiliary_loss_clip": 0.01097011,
      "auxiliary_loss_mlp": 0.01021375,
      "balance_loss_clip": 1.04017842,
      "balance_loss_mlp": 1.01506305,
      "epoch": 0.9493176215956232,
      "flos": 25332535998720.0,
      "grad_norm": 2.0920236436505624,
      "language_loss": 0.7055819,
      "learning_rate": 2.6827129090534862e-08,
      "loss": 0.72676575,
      "num_input_tokens_seen": 170600165,
      "step": 7895,
      "time_per_iteration": 2.6666183471679688
    },
    {
      "auxiliary_loss_clip": 0.01130675,
      "auxiliary_loss_mlp": 0.01022661,
      "balance_loss_clip": 1.04381776,
      "balance_loss_mlp": 1.0161252,
      "epoch": 0.9494378644862622,
      "flos": 21029971236480.0,
      "grad_norm": 2.024898480946682,
      "language_loss": 0.77630615,
      "learning_rate": 2.670011964646335e-08,
      "loss": 0.79783952,
      "num_input_tokens_seen": 170618845,
      "step": 7896,
      "time_per_iteration": 3.335475206375122
    },
    {
      "auxiliary_loss_clip": 0.01082419,
      "auxiliary_loss_mlp": 0.01025014,
      "balance_loss_clip": 1.03480339,
      "balance_loss_mlp": 1.01818609,
      "epoch": 0.9495581073769014,
      "flos": 15195134148480.0,
      "grad_norm": 1.8714646218424649,
      "language_loss": 0.67868996,
      "learning_rate": 2.657340955041487e-08,
      "loss": 0.69976425,
      "num_input_tokens_seen": 170637620,
      "step": 7897,
      "time_per_iteration": 2.6449859142303467
    },
    {
      "auxiliary_loss_clip": 0.01131055,
      "auxiliary_loss_mlp": 0.01029606,
      "balance_loss_clip": 1.04510069,
      "balance_loss_mlp": 1.02200317,
      "epoch": 0.9496783502675404,
      "flos": 28616566705920.0,
      "grad_norm": 1.8926893590468887,
      "language_loss": 0.71491122,
      "learning_rate": 2.6446998821611167e-08,
      "loss": 0.73651785,
      "num_input_tokens_seen": 170657815,
      "step": 7898,
      "time_per_iteration": 2.5684080123901367
    },
    {
      "auxiliary_loss_clip": 0.0110877,
      "auxiliary_loss_mlp": 0.01029062,
      "balance_loss_clip": 1.04281676,
      "balance_loss_mlp": 1.02230239,
      "epoch": 0.9497985931581795,
      "flos": 14866874732160.0,
      "grad_norm": 2.4701581353210753,
      "language_loss": 0.7157073,
      "learning_rate": 2.6320887479228228e-08,
      "loss": 0.73708564,
      "num_input_tokens_seen": 170674415,
      "step": 7899,
      "time_per_iteration": 2.547853469848633
    },
    {
      "auxiliary_loss_clip": 0.01138222,
      "auxiliary_loss_mlp": 0.01022155,
      "balance_loss_clip": 1.04493666,
      "balance_loss_mlp": 1.01541114,
      "epoch": 0.9499188360488187,
      "flos": 27193319136000.0,
      "grad_norm": 2.295392077443322,
      "language_loss": 0.72826374,
      "learning_rate": 2.619507554239786e-08,
      "loss": 0.74986744,
      "num_input_tokens_seen": 170692975,
      "step": 7900,
      "time_per_iteration": 2.570667266845703
    },
    {
      "auxiliary_loss_clip": 0.01132942,
      "auxiliary_loss_mlp": 0.01024651,
      "balance_loss_clip": 1.04412162,
      "balance_loss_mlp": 1.01779652,
      "epoch": 0.9500390789394577,
      "flos": 24316479982080.0,
      "grad_norm": 1.545141823646415,
      "language_loss": 0.69540977,
      "learning_rate": 2.606956303020502e-08,
      "loss": 0.7169857,
      "num_input_tokens_seen": 170713780,
      "step": 7901,
      "time_per_iteration": 2.5605649948120117
    },
    {
      "auxiliary_loss_clip": 0.01150465,
      "auxiliary_loss_mlp": 0.0102552,
      "balance_loss_clip": 1.04760814,
      "balance_loss_mlp": 1.01857591,
      "epoch": 0.9501593218300968,
      "flos": 14354752573440.0,
      "grad_norm": 1.8068983799193632,
      "language_loss": 0.84060031,
      "learning_rate": 2.5944349961690036e-08,
      "loss": 0.86236018,
      "num_input_tokens_seen": 170730800,
      "step": 7902,
      "time_per_iteration": 2.4686028957366943
    },
    {
      "auxiliary_loss_clip": 0.01121914,
      "auxiliary_loss_mlp": 0.01023832,
      "balance_loss_clip": 1.04303932,
      "balance_loss_mlp": 1.01698303,
      "epoch": 0.9502795647207359,
      "flos": 38728113742080.0,
      "grad_norm": 1.7650757846732053,
      "language_loss": 0.72842133,
      "learning_rate": 2.581943635584749e-08,
      "loss": 0.74987876,
      "num_input_tokens_seen": 170753630,
      "step": 7903,
      "time_per_iteration": 2.7076985836029053
    },
    {
      "auxiliary_loss_clip": 0.01127932,
      "auxiliary_loss_mlp": 0.01031542,
      "balance_loss_clip": 1.04428411,
      "balance_loss_mlp": 1.02475643,
      "epoch": 0.950399807611375,
      "flos": 40808023799040.0,
      "grad_norm": 1.4812359540192912,
      "language_loss": 0.653153,
      "learning_rate": 2.569482223162689e-08,
      "loss": 0.67474771,
      "num_input_tokens_seen": 170777605,
      "step": 7904,
      "time_per_iteration": 2.68632173538208
    },
    {
      "auxiliary_loss_clip": 0.01149057,
      "auxiliary_loss_mlp": 0.01025605,
      "balance_loss_clip": 1.04560804,
      "balance_loss_mlp": 1.01855659,
      "epoch": 0.950520050502014,
      "flos": 23440403266560.0,
      "grad_norm": 1.6685831155075352,
      "language_loss": 0.72111511,
      "learning_rate": 2.5570507607932e-08,
      "loss": 0.74286175,
      "num_input_tokens_seen": 170797520,
      "step": 7905,
      "time_per_iteration": 2.5133769512176514
    },
    {
      "auxiliary_loss_clip": 0.01151071,
      "auxiliary_loss_mlp": 0.01026718,
      "balance_loss_clip": 1.04508543,
      "balance_loss_mlp": 1.01955962,
      "epoch": 0.9506402933926532,
      "flos": 17783718658560.0,
      "grad_norm": 3.7973890661965215,
      "language_loss": 0.64131904,
      "learning_rate": 2.54464925036213e-08,
      "loss": 0.6630969,
      "num_input_tokens_seen": 170814810,
      "step": 7906,
      "time_per_iteration": 2.4693918228149414
    },
    {
      "auxiliary_loss_clip": 0.01148256,
      "auxiliary_loss_mlp": 0.0102466,
      "balance_loss_clip": 1.04694366,
      "balance_loss_mlp": 1.01777303,
      "epoch": 0.9507605362832923,
      "flos": 32561928668160.0,
      "grad_norm": 1.761703395028214,
      "language_loss": 0.60960472,
      "learning_rate": 2.532277693750773e-08,
      "loss": 0.63133383,
      "num_input_tokens_seen": 170835735,
      "step": 7907,
      "time_per_iteration": 2.561464309692383
    },
    {
      "auxiliary_loss_clip": 0.01103562,
      "auxiliary_loss_mlp": 0.01025205,
      "balance_loss_clip": 1.04317939,
      "balance_loss_mlp": 1.01838017,
      "epoch": 0.9508807791739313,
      "flos": 19602054898560.0,
      "grad_norm": 2.0120881366925656,
      "language_loss": 0.75609457,
      "learning_rate": 2.5199360928358948e-08,
      "loss": 0.77738225,
      "num_input_tokens_seen": 170852970,
      "step": 7908,
      "time_per_iteration": 3.3993828296661377
    },
    {
      "auxiliary_loss_clip": 0.0113914,
      "auxiliary_loss_mlp": 0.00760052,
      "balance_loss_clip": 1.04289854,
      "balance_loss_mlp": 1.0002929,
      "epoch": 0.9510010220645704,
      "flos": 21471852349440.0,
      "grad_norm": 1.5820022184686973,
      "language_loss": 0.86942559,
      "learning_rate": 2.507624449489665e-08,
      "loss": 0.8884176,
      "num_input_tokens_seen": 170871600,
      "step": 7909,
      "time_per_iteration": 3.3025944232940674
    },
    {
      "auxiliary_loss_clip": 0.01134441,
      "auxiliary_loss_mlp": 0.01024182,
      "balance_loss_clip": 1.04499483,
      "balance_loss_mlp": 1.01739287,
      "epoch": 0.9511212649552095,
      "flos": 18879999701760.0,
      "grad_norm": 2.1207602743605447,
      "language_loss": 0.65253234,
      "learning_rate": 2.495342765579811e-08,
      "loss": 0.67411858,
      "num_input_tokens_seen": 170890260,
      "step": 7910,
      "time_per_iteration": 2.513471841812134
    },
    {
      "auxiliary_loss_clip": 0.011021,
      "auxiliary_loss_mlp": 0.01025021,
      "balance_loss_clip": 1.04042482,
      "balance_loss_mlp": 1.0178324,
      "epoch": 0.9512415078458486,
      "flos": 20810521094400.0,
      "grad_norm": 1.8478054926460676,
      "language_loss": 0.70757937,
      "learning_rate": 2.4830910429693984e-08,
      "loss": 0.7288506,
      "num_input_tokens_seen": 170910220,
      "step": 7911,
      "time_per_iteration": 3.3309640884399414
    },
    {
      "auxiliary_loss_clip": 0.01162709,
      "auxiliary_loss_mlp": 0.01025653,
      "balance_loss_clip": 1.04718328,
      "balance_loss_mlp": 1.01855707,
      "epoch": 0.9513617507364877,
      "flos": 18369565482240.0,
      "grad_norm": 1.801422876623645,
      "language_loss": 0.7967304,
      "learning_rate": 2.470869283517052e-08,
      "loss": 0.81861401,
      "num_input_tokens_seen": 170928255,
      "step": 7912,
      "time_per_iteration": 2.443518877029419
    },
    {
      "auxiliary_loss_clip": 0.01144583,
      "auxiliary_loss_mlp": 0.01027009,
      "balance_loss_clip": 1.04535592,
      "balance_loss_mlp": 1.02015197,
      "epoch": 0.9514819936271268,
      "flos": 25010166412800.0,
      "grad_norm": 1.6857098925852978,
      "language_loss": 0.77014732,
      "learning_rate": 2.458677489076777e-08,
      "loss": 0.79186326,
      "num_input_tokens_seen": 170949265,
      "step": 7913,
      "time_per_iteration": 2.524535894393921
    },
    {
      "auxiliary_loss_clip": 0.01141022,
      "auxiliary_loss_mlp": 0.01028338,
      "balance_loss_clip": 1.04459929,
      "balance_loss_mlp": 1.02217209,
      "epoch": 0.9516022365177659,
      "flos": 18662129758080.0,
      "grad_norm": 1.649405978975886,
      "language_loss": 0.82723534,
      "learning_rate": 2.446515661498072e-08,
      "loss": 0.84892893,
      "num_input_tokens_seen": 170968595,
      "step": 7914,
      "time_per_iteration": 2.497105121612549
    },
    {
      "auxiliary_loss_clip": 0.01091358,
      "auxiliary_loss_mlp": 0.01020692,
      "balance_loss_clip": 1.03982162,
      "balance_loss_mlp": 1.01407909,
      "epoch": 0.9517224794084049,
      "flos": 25372109808000.0,
      "grad_norm": 1.9501716516712877,
      "language_loss": 0.73938251,
      "learning_rate": 2.434383802625861e-08,
      "loss": 0.76050305,
      "num_input_tokens_seen": 170987550,
      "step": 7915,
      "time_per_iteration": 2.6496002674102783
    },
    {
      "auxiliary_loss_clip": 0.01120976,
      "auxiliary_loss_mlp": 0.01022001,
      "balance_loss_clip": 1.04071927,
      "balance_loss_mlp": 1.01514053,
      "epoch": 0.9518427222990441,
      "flos": 21470918595840.0,
      "grad_norm": 1.9855222520895315,
      "language_loss": 0.73785037,
      "learning_rate": 2.4222819143005168e-08,
      "loss": 0.7592802,
      "num_input_tokens_seen": 171007145,
      "step": 7916,
      "time_per_iteration": 2.5636677742004395
    },
    {
      "auxiliary_loss_clip": 0.01161615,
      "auxiliary_loss_mlp": 0.01022693,
      "balance_loss_clip": 1.04880071,
      "balance_loss_mlp": 1.0162015,
      "epoch": 0.9519629651896832,
      "flos": 21033634423680.0,
      "grad_norm": 1.646995455386403,
      "language_loss": 0.80710578,
      "learning_rate": 2.4102099983579706e-08,
      "loss": 0.82894886,
      "num_input_tokens_seen": 171026295,
      "step": 7917,
      "time_per_iteration": 2.4575631618499756
    },
    {
      "auxiliary_loss_clip": 0.01148762,
      "auxiliary_loss_mlp": 0.01022499,
      "balance_loss_clip": 1.04490614,
      "balance_loss_mlp": 1.01525426,
      "epoch": 0.9520832080803222,
      "flos": 21689219502720.0,
      "grad_norm": 1.6859170259011533,
      "language_loss": 0.77212524,
      "learning_rate": 2.3981680566294236e-08,
      "loss": 0.79383785,
      "num_input_tokens_seen": 171045895,
      "step": 7918,
      "time_per_iteration": 2.511439800262451
    },
    {
      "auxiliary_loss_clip": 0.01161373,
      "auxiliary_loss_mlp": 0.01024876,
      "balance_loss_clip": 1.04808199,
      "balance_loss_mlp": 1.01836765,
      "epoch": 0.9522034509709614,
      "flos": 23145289125120.0,
      "grad_norm": 1.8569559298984726,
      "language_loss": 0.73472846,
      "learning_rate": 2.3861560909416822e-08,
      "loss": 0.7565909,
      "num_input_tokens_seen": 171065445,
      "step": 7919,
      "time_per_iteration": 2.4702367782592773
    },
    {
      "auxiliary_loss_clip": 0.011024,
      "auxiliary_loss_mlp": 0.01030643,
      "balance_loss_clip": 1.04343975,
      "balance_loss_mlp": 1.02346063,
      "epoch": 0.9523236938616004,
      "flos": 24679428958080.0,
      "grad_norm": 1.7530768167979855,
      "language_loss": 0.82661229,
      "learning_rate": 2.3741741031169325e-08,
      "loss": 0.84794271,
      "num_input_tokens_seen": 171085015,
      "step": 7920,
      "time_per_iteration": 2.621312141418457
    },
    {
      "auxiliary_loss_clip": 0.01102955,
      "auxiliary_loss_mlp": 0.01024986,
      "balance_loss_clip": 1.04062366,
      "balance_loss_mlp": 1.01813126,
      "epoch": 0.9524439367522395,
      "flos": 22672309812480.0,
      "grad_norm": 1.7758829844082649,
      "language_loss": 0.71437663,
      "learning_rate": 2.3622220949728544e-08,
      "loss": 0.73565614,
      "num_input_tokens_seen": 171103900,
      "step": 7921,
      "time_per_iteration": 2.590489625930786
    },
    {
      "auxiliary_loss_clip": 0.01141133,
      "auxiliary_loss_mlp": 0.01023753,
      "balance_loss_clip": 1.043926,
      "balance_loss_mlp": 1.0164361,
      "epoch": 0.9525641796428787,
      "flos": 34055525024640.0,
      "grad_norm": 2.9703364674233748,
      "language_loss": 0.60959595,
      "learning_rate": 2.3503000683225526e-08,
      "loss": 0.63124478,
      "num_input_tokens_seen": 171121615,
      "step": 7922,
      "time_per_iteration": 3.3196325302124023
    },
    {
      "auxiliary_loss_clip": 0.01163396,
      "auxiliary_loss_mlp": 0.01028028,
      "balance_loss_clip": 1.0465095,
      "balance_loss_mlp": 1.02083373,
      "epoch": 0.9526844225335177,
      "flos": 16727083251840.0,
      "grad_norm": 2.1846749666102925,
      "language_loss": 0.84483755,
      "learning_rate": 2.3384080249745585e-08,
      "loss": 0.86675179,
      "num_input_tokens_seen": 171139505,
      "step": 7923,
      "time_per_iteration": 2.460360050201416
    },
    {
      "auxiliary_loss_clip": 0.01111341,
      "auxiliary_loss_mlp": 0.01024827,
      "balance_loss_clip": 1.04194558,
      "balance_loss_mlp": 1.01844299,
      "epoch": 0.9528046654241568,
      "flos": 36939367330560.0,
      "grad_norm": 2.527129576715102,
      "language_loss": 0.82441032,
      "learning_rate": 2.3265459667329178e-08,
      "loss": 0.84577197,
      "num_input_tokens_seen": 171158995,
      "step": 7924,
      "time_per_iteration": 2.7342448234558105
    },
    {
      "auxiliary_loss_clip": 0.01140465,
      "auxiliary_loss_mlp": 0.01024054,
      "balance_loss_clip": 1.04779387,
      "balance_loss_mlp": 1.01744342,
      "epoch": 0.9529249083147959,
      "flos": 18255010032000.0,
      "grad_norm": 1.9396786700388284,
      "language_loss": 0.85966545,
      "learning_rate": 2.31471389539708e-08,
      "loss": 0.8813107,
      "num_input_tokens_seen": 171176120,
      "step": 7925,
      "time_per_iteration": 2.5428967475891113
    },
    {
      "auxiliary_loss_clip": 0.01148997,
      "auxiliary_loss_mlp": 0.00760101,
      "balance_loss_clip": 1.04583526,
      "balance_loss_mlp": 1.00029182,
      "epoch": 0.953045151205435,
      "flos": 28658438985600.0,
      "grad_norm": 1.8628895356412594,
      "language_loss": 0.72673297,
      "learning_rate": 2.3029118127619872e-08,
      "loss": 0.74582398,
      "num_input_tokens_seen": 171195835,
      "step": 7926,
      "time_per_iteration": 2.575881004333496
    },
    {
      "auxiliary_loss_clip": 0.0112444,
      "auxiliary_loss_mlp": 0.01025374,
      "balance_loss_clip": 1.04117024,
      "balance_loss_mlp": 1.01872182,
      "epoch": 0.953165394096074,
      "flos": 21835232288640.0,
      "grad_norm": 2.150307555912813,
      "language_loss": 0.87056887,
      "learning_rate": 2.2911397206179628e-08,
      "loss": 0.89206707,
      "num_input_tokens_seen": 171212585,
      "step": 7927,
      "time_per_iteration": 2.5083611011505127
    },
    {
      "auxiliary_loss_clip": 0.01163636,
      "auxiliary_loss_mlp": 0.01026226,
      "balance_loss_clip": 1.04881251,
      "balance_loss_mlp": 1.01959467,
      "epoch": 0.9532856369867132,
      "flos": 19975059682560.0,
      "grad_norm": 1.7182645805307202,
      "language_loss": 0.62782133,
      "learning_rate": 2.279397620750845e-08,
      "loss": 0.64971995,
      "num_input_tokens_seen": 171231630,
      "step": 7928,
      "time_per_iteration": 2.4710488319396973
    },
    {
      "auxiliary_loss_clip": 0.01131594,
      "auxiliary_loss_mlp": 0.01030024,
      "balance_loss_clip": 1.04262137,
      "balance_loss_mlp": 1.02338147,
      "epoch": 0.9534058798773523,
      "flos": 15049588239360.0,
      "grad_norm": 1.8755278718336674,
      "language_loss": 0.7872082,
      "learning_rate": 2.2676855149419195e-08,
      "loss": 0.80882436,
      "num_input_tokens_seen": 171248800,
      "step": 7929,
      "time_per_iteration": 2.5160133838653564
    },
    {
      "auxiliary_loss_clip": 0.01130449,
      "auxiliary_loss_mlp": 0.0102796,
      "balance_loss_clip": 1.04719353,
      "balance_loss_mlp": 1.02031851,
      "epoch": 0.9535261227679913,
      "flos": 17602800831360.0,
      "grad_norm": 2.360955527824072,
      "language_loss": 0.75064731,
      "learning_rate": 2.2560034049678988e-08,
      "loss": 0.77223134,
      "num_input_tokens_seen": 171263150,
      "step": 7930,
      "time_per_iteration": 2.497328042984009
    },
    {
      "auxiliary_loss_clip": 0.01166309,
      "auxiliary_loss_mlp": 0.01026048,
      "balance_loss_clip": 1.04893017,
      "balance_loss_mlp": 1.01872241,
      "epoch": 0.9536463656586305,
      "flos": 23142954741120.0,
      "grad_norm": 1.6816326985458678,
      "language_loss": 0.75341535,
      "learning_rate": 2.2443512926008988e-08,
      "loss": 0.77533895,
      "num_input_tokens_seen": 171282480,
      "step": 7931,
      "time_per_iteration": 2.472205638885498
    },
    {
      "auxiliary_loss_clip": 0.01124612,
      "auxiliary_loss_mlp": 0.01026275,
      "balance_loss_clip": 1.04432535,
      "balance_loss_mlp": 1.01939392,
      "epoch": 0.9537666085492695,
      "flos": 18625033987200.0,
      "grad_norm": 2.32691376050846,
      "language_loss": 0.69540119,
      "learning_rate": 2.2327291796085946e-08,
      "loss": 0.71691006,
      "num_input_tokens_seen": 171300840,
      "step": 7932,
      "time_per_iteration": 2.5359432697296143
    },
    {
      "auxiliary_loss_clip": 0.01161918,
      "auxiliary_loss_mlp": 0.01023155,
      "balance_loss_clip": 1.04678321,
      "balance_loss_mlp": 1.015836,
      "epoch": 0.9538868514399086,
      "flos": 18989347680000.0,
      "grad_norm": 2.5417419410564697,
      "language_loss": 0.77222002,
      "learning_rate": 2.2211370677540197e-08,
      "loss": 0.79407078,
      "num_input_tokens_seen": 171317365,
      "step": 7933,
      "time_per_iteration": 2.4328887462615967
    },
    {
      "auxiliary_loss_clip": 0.01166106,
      "auxiliary_loss_mlp": 0.01032247,
      "balance_loss_clip": 1.04951131,
      "balance_loss_mlp": 1.02584839,
      "epoch": 0.9540070943305478,
      "flos": 16800556521600.0,
      "grad_norm": 2.5873844454302573,
      "language_loss": 0.78030205,
      "learning_rate": 2.2095749587957012e-08,
      "loss": 0.80228555,
      "num_input_tokens_seen": 171335270,
      "step": 7934,
      "time_per_iteration": 3.261431932449341
    },
    {
      "auxiliary_loss_clip": 0.01130144,
      "auxiliary_loss_mlp": 0.01023978,
      "balance_loss_clip": 1.04031932,
      "balance_loss_mlp": 1.01695979,
      "epoch": 0.9541273372211868,
      "flos": 20156911263360.0,
      "grad_norm": 1.810720897448241,
      "language_loss": 0.69742072,
      "learning_rate": 2.1980428544876138e-08,
      "loss": 0.71896195,
      "num_input_tokens_seen": 171353910,
      "step": 7935,
      "time_per_iteration": 3.3143246173858643
    },
    {
      "auxiliary_loss_clip": 0.01098271,
      "auxiliary_loss_mlp": 0.01021245,
      "balance_loss_clip": 1.03606629,
      "balance_loss_mlp": 1.01450026,
      "epoch": 0.9542475801118259,
      "flos": 26725511381760.0,
      "grad_norm": 1.6655063642925354,
      "language_loss": 0.73815668,
      "learning_rate": 2.1865407565791584e-08,
      "loss": 0.75935179,
      "num_input_tokens_seen": 171375480,
      "step": 7936,
      "time_per_iteration": 2.639575719833374
    },
    {
      "auxiliary_loss_clip": 0.01134288,
      "auxiliary_loss_mlp": 0.01021714,
      "balance_loss_clip": 1.04254246,
      "balance_loss_mlp": 1.0151515,
      "epoch": 0.954367823002465,
      "flos": 23330911633920.0,
      "grad_norm": 1.9395849067260493,
      "language_loss": 0.77577543,
      "learning_rate": 2.175068666815183e-08,
      "loss": 0.79733551,
      "num_input_tokens_seen": 171396320,
      "step": 7937,
      "time_per_iteration": 3.2276721000671387
    },
    {
      "auxiliary_loss_clip": 0.01122672,
      "auxiliary_loss_mlp": 0.01035291,
      "balance_loss_clip": 1.04391956,
      "balance_loss_mlp": 1.0277276,
      "epoch": 0.9544880658931041,
      "flos": 14902713527040.0,
      "grad_norm": 2.075506215072328,
      "language_loss": 0.79049915,
      "learning_rate": 2.163626586935985e-08,
      "loss": 0.81207883,
      "num_input_tokens_seen": 171412860,
      "step": 7938,
      "time_per_iteration": 2.5457711219787598
    },
    {
      "auxiliary_loss_clip": 0.01144343,
      "auxiliary_loss_mlp": 0.01028278,
      "balance_loss_clip": 1.04336894,
      "balance_loss_mlp": 1.02100885,
      "epoch": 0.9546083087837431,
      "flos": 29095902725760.0,
      "grad_norm": 1.9382700284821877,
      "language_loss": 0.62922144,
      "learning_rate": 2.1522145186773755e-08,
      "loss": 0.65094763,
      "num_input_tokens_seen": 171431780,
      "step": 7939,
      "time_per_iteration": 2.539872646331787
    },
    {
      "auxiliary_loss_clip": 0.01134788,
      "auxiliary_loss_mlp": 0.0102525,
      "balance_loss_clip": 1.04499173,
      "balance_loss_mlp": 1.01827621,
      "epoch": 0.9547285516743822,
      "flos": 21142335957120.0,
      "grad_norm": 1.9304987341021955,
      "language_loss": 0.85650831,
      "learning_rate": 2.140832463770481e-08,
      "loss": 0.87810868,
      "num_input_tokens_seen": 171450975,
      "step": 7940,
      "time_per_iteration": 2.5629146099090576
    },
    {
      "auxiliary_loss_clip": 0.011364,
      "auxiliary_loss_mlp": 0.01026522,
      "balance_loss_clip": 1.04242373,
      "balance_loss_mlp": 1.01913393,
      "epoch": 0.9548487945650214,
      "flos": 27490157130240.0,
      "grad_norm": 2.2815317655781615,
      "language_loss": 0.76083231,
      "learning_rate": 2.129480423941987e-08,
      "loss": 0.78246152,
      "num_input_tokens_seen": 171467645,
      "step": 7941,
      "time_per_iteration": 2.5610032081604004
    },
    {
      "auxiliary_loss_clip": 0.01134823,
      "auxiliary_loss_mlp": 0.01022295,
      "balance_loss_clip": 1.04528093,
      "balance_loss_mlp": 1.01579475,
      "epoch": 0.9549690374556604,
      "flos": 22273198819200.0,
      "grad_norm": 1.6283804033895795,
      "language_loss": 0.80229771,
      "learning_rate": 2.1181584009140052e-08,
      "loss": 0.82386887,
      "num_input_tokens_seen": 171487185,
      "step": 7942,
      "time_per_iteration": 2.530864953994751
    },
    {
      "auxiliary_loss_clip": 0.01127447,
      "auxiliary_loss_mlp": 0.01021017,
      "balance_loss_clip": 1.0432303,
      "balance_loss_mlp": 1.01471364,
      "epoch": 0.9550892803462995,
      "flos": 17595294888960.0,
      "grad_norm": 2.203500013182331,
      "language_loss": 0.83454543,
      "learning_rate": 2.10686639640405e-08,
      "loss": 0.85603005,
      "num_input_tokens_seen": 171501275,
      "step": 7943,
      "time_per_iteration": 2.5469717979431152
    },
    {
      "auxiliary_loss_clip": 0.01145655,
      "auxiliary_loss_mlp": 0.01024409,
      "balance_loss_clip": 1.04357362,
      "balance_loss_mlp": 1.01711917,
      "epoch": 0.9552095232369386,
      "flos": 24353144789760.0,
      "grad_norm": 1.5649408122038833,
      "language_loss": 0.81138825,
      "learning_rate": 2.0956044121251294e-08,
      "loss": 0.83308887,
      "num_input_tokens_seen": 171520060,
      "step": 7944,
      "time_per_iteration": 2.5175137519836426
    },
    {
      "auxiliary_loss_clip": 0.01119496,
      "auxiliary_loss_mlp": 0.01023771,
      "balance_loss_clip": 1.04524159,
      "balance_loss_mlp": 1.01602817,
      "epoch": 0.9553297661275777,
      "flos": 22746860490240.0,
      "grad_norm": 1.7815938619649618,
      "language_loss": 0.81238055,
      "learning_rate": 2.084372449785654e-08,
      "loss": 0.83381325,
      "num_input_tokens_seen": 171539895,
      "step": 7945,
      "time_per_iteration": 2.5679187774658203
    },
    {
      "auxiliary_loss_clip": 0.01129985,
      "auxiliary_loss_mlp": 0.01024386,
      "balance_loss_clip": 1.04196763,
      "balance_loss_mlp": 1.01792216,
      "epoch": 0.9554500090182168,
      "flos": 15413866018560.0,
      "grad_norm": 1.6902451338785822,
      "language_loss": 0.68593669,
      "learning_rate": 2.0731705110895282e-08,
      "loss": 0.70748043,
      "num_input_tokens_seen": 171557385,
      "step": 7946,
      "time_per_iteration": 2.51924991607666
    },
    {
      "auxiliary_loss_clip": 0.01151677,
      "auxiliary_loss_mlp": 0.01029235,
      "balance_loss_clip": 1.04832244,
      "balance_loss_mlp": 1.02245831,
      "epoch": 0.9555702519088559,
      "flos": 23513517400320.0,
      "grad_norm": 2.1378180441024575,
      "language_loss": 0.86633778,
      "learning_rate": 2.0619985977360587e-08,
      "loss": 0.88814688,
      "num_input_tokens_seen": 171575705,
      "step": 7947,
      "time_per_iteration": 2.516890048980713
    },
    {
      "auxiliary_loss_clip": 0.01107319,
      "auxiliary_loss_mlp": 0.01026027,
      "balance_loss_clip": 1.03996718,
      "balance_loss_mlp": 1.01985538,
      "epoch": 0.955690494799495,
      "flos": 22962072827520.0,
      "grad_norm": 1.806933176924031,
      "language_loss": 0.7659893,
      "learning_rate": 2.0508567114200237e-08,
      "loss": 0.78732276,
      "num_input_tokens_seen": 171595620,
      "step": 7948,
      "time_per_iteration": 3.4240827560424805
    },
    {
      "auxiliary_loss_clip": 0.01138483,
      "auxiliary_loss_mlp": 0.01023638,
      "balance_loss_clip": 1.04499078,
      "balance_loss_mlp": 1.01735592,
      "epoch": 0.955810737690134,
      "flos": 26031250333440.0,
      "grad_norm": 1.9613280032332647,
      "language_loss": 0.78864652,
      "learning_rate": 2.0397448538316485e-08,
      "loss": 0.81026775,
      "num_input_tokens_seen": 171616660,
      "step": 7949,
      "time_per_iteration": 2.5868725776672363
    },
    {
      "auxiliary_loss_clip": 0.01114372,
      "auxiliary_loss_mlp": 0.01027177,
      "balance_loss_clip": 1.04110646,
      "balance_loss_mlp": 1.02106452,
      "epoch": 0.9559309805807732,
      "flos": 20849951249280.0,
      "grad_norm": 1.874655764486858,
      "language_loss": 0.66738731,
      "learning_rate": 2.028663026656563e-08,
      "loss": 0.68880272,
      "num_input_tokens_seen": 171635515,
      "step": 7950,
      "time_per_iteration": 2.564220428466797
    },
    {
      "auxiliary_loss_clip": 0.01161102,
      "auxiliary_loss_mlp": 0.00760581,
      "balance_loss_clip": 1.04664993,
      "balance_loss_mlp": 1.00033307,
      "epoch": 0.9560512234714122,
      "flos": 21578219498880.0,
      "grad_norm": 1.9809210709014204,
      "language_loss": 0.71698362,
      "learning_rate": 2.0176112315758885e-08,
      "loss": 0.73620045,
      "num_input_tokens_seen": 171653305,
      "step": 7951,
      "time_per_iteration": 2.451366662979126
    },
    {
      "auxiliary_loss_clip": 0.01114575,
      "auxiliary_loss_mlp": 0.0103035,
      "balance_loss_clip": 1.04311395,
      "balance_loss_mlp": 1.02289367,
      "epoch": 0.9561714663620513,
      "flos": 17450144029440.0,
      "grad_norm": 2.726791228028147,
      "language_loss": 0.69659364,
      "learning_rate": 2.0065894702661957e-08,
      "loss": 0.71804285,
      "num_input_tokens_seen": 171669980,
      "step": 7952,
      "time_per_iteration": 2.5818333625793457
    },
    {
      "auxiliary_loss_clip": 0.0111176,
      "auxiliary_loss_mlp": 0.00761138,
      "balance_loss_clip": 1.0395751,
      "balance_loss_mlp": 1.00029814,
      "epoch": 0.9562917092526905,
      "flos": 26098510550400.0,
      "grad_norm": 1.9120014640203198,
      "language_loss": 0.77680653,
      "learning_rate": 1.9955977443994577e-08,
      "loss": 0.7955355,
      "num_input_tokens_seen": 171689970,
      "step": 7953,
      "time_per_iteration": 2.5979108810424805
    },
    {
      "auxiliary_loss_clip": 0.01132851,
      "auxiliary_loss_mlp": 0.01025973,
      "balance_loss_clip": 1.04314661,
      "balance_loss_mlp": 1.0179739,
      "epoch": 0.9564119521433295,
      "flos": 24096742531200.0,
      "grad_norm": 2.60970077115704,
      "language_loss": 0.62206787,
      "learning_rate": 1.9846360556430965e-08,
      "loss": 0.64365613,
      "num_input_tokens_seen": 171708270,
      "step": 7954,
      "time_per_iteration": 2.5528271198272705
    },
    {
      "auxiliary_loss_clip": 0.01161423,
      "auxiliary_loss_mlp": 0.01023838,
      "balance_loss_clip": 1.04621172,
      "balance_loss_mlp": 1.01683736,
      "epoch": 0.9565321950339686,
      "flos": 32008903896960.0,
      "grad_norm": 2.1395829503083306,
      "language_loss": 0.61498952,
      "learning_rate": 1.973704405660004e-08,
      "loss": 0.63684213,
      "num_input_tokens_seen": 171729385,
      "step": 7955,
      "time_per_iteration": 2.5460877418518066
    },
    {
      "auxiliary_loss_clip": 0.01090174,
      "auxiliary_loss_mlp": 0.01024362,
      "balance_loss_clip": 1.03893149,
      "balance_loss_mlp": 1.01797509,
      "epoch": 0.9566524379246077,
      "flos": 23588642695680.0,
      "grad_norm": 1.4375879785027537,
      "language_loss": 0.77867413,
      "learning_rate": 1.9628027961085203e-08,
      "loss": 0.79981953,
      "num_input_tokens_seen": 171752615,
      "step": 7956,
      "time_per_iteration": 2.6747639179229736
    },
    {
      "auxiliary_loss_clip": 0.01109253,
      "auxiliary_loss_mlp": 0.0102387,
      "balance_loss_clip": 1.038486,
      "balance_loss_mlp": 1.01716483,
      "epoch": 0.9567726808152468,
      "flos": 38067716240640.0,
      "grad_norm": 1.8186485631267544,
      "language_loss": 0.84116441,
      "learning_rate": 1.9519312286423894e-08,
      "loss": 0.86249566,
      "num_input_tokens_seen": 171775810,
      "step": 7957,
      "time_per_iteration": 2.7127623558044434
    },
    {
      "auxiliary_loss_clip": 0.01147158,
      "auxiliary_loss_mlp": 0.01028298,
      "balance_loss_clip": 1.04654002,
      "balance_loss_mlp": 1.02106833,
      "epoch": 0.9568929237058859,
      "flos": 22744059229440.0,
      "grad_norm": 1.4927151193798343,
      "language_loss": 0.77671731,
      "learning_rate": 1.9410897049108255e-08,
      "loss": 0.79847193,
      "num_input_tokens_seen": 171795090,
      "step": 7958,
      "time_per_iteration": 2.509469747543335
    },
    {
      "auxiliary_loss_clip": 0.01169109,
      "auxiliary_loss_mlp": 0.01030606,
      "balance_loss_clip": 1.05068994,
      "balance_loss_mlp": 1.02333117,
      "epoch": 0.957013166596525,
      "flos": 23841633162240.0,
      "grad_norm": 1.7084350605418128,
      "language_loss": 0.91182852,
      "learning_rate": 1.9302782265584905e-08,
      "loss": 0.93382573,
      "num_input_tokens_seen": 171815755,
      "step": 7959,
      "time_per_iteration": 2.494845151901245
    },
    {
      "auxiliary_loss_clip": 0.01094004,
      "auxiliary_loss_mlp": 0.01027163,
      "balance_loss_clip": 1.04119349,
      "balance_loss_mlp": 1.02073741,
      "epoch": 0.9571334094871641,
      "flos": 17639286071040.0,
      "grad_norm": 2.1382424909288744,
      "language_loss": 0.86973548,
      "learning_rate": 1.9194967952254282e-08,
      "loss": 0.8909471,
      "num_input_tokens_seen": 171834330,
      "step": 7960,
      "time_per_iteration": 3.380476713180542
    },
    {
      "auxiliary_loss_clip": 0.01150887,
      "auxiliary_loss_mlp": 0.01026989,
      "balance_loss_clip": 1.04972529,
      "balance_loss_mlp": 1.02016461,
      "epoch": 0.9572536523778031,
      "flos": 15369623441280.0,
      "grad_norm": 2.4840838323193397,
      "language_loss": 0.80781662,
      "learning_rate": 1.9087454125472635e-08,
      "loss": 0.82959533,
      "num_input_tokens_seen": 171848805,
      "step": 7961,
      "time_per_iteration": 3.310548782348633
    },
    {
      "auxiliary_loss_clip": 0.01164894,
      "auxiliary_loss_mlp": 0.01025772,
      "balance_loss_clip": 1.04920328,
      "balance_loss_mlp": 1.01817274,
      "epoch": 0.9573738952684423,
      "flos": 24969838417920.0,
      "grad_norm": 1.8096127494770102,
      "language_loss": 0.7824899,
      "learning_rate": 1.8980240801548696e-08,
      "loss": 0.80439651,
      "num_input_tokens_seen": 171867995,
      "step": 7962,
      "time_per_iteration": 2.479408025741577
    },
    {
      "auxiliary_loss_clip": 0.01135357,
      "auxiliary_loss_mlp": 0.01026083,
      "balance_loss_clip": 1.04787564,
      "balance_loss_mlp": 1.01969695,
      "epoch": 0.9574941381590814,
      "flos": 25769461034880.0,
      "grad_norm": 1.6374709454689123,
      "language_loss": 0.74479234,
      "learning_rate": 1.8873327996747458e-08,
      "loss": 0.76640677,
      "num_input_tokens_seen": 171886495,
      "step": 7963,
      "time_per_iteration": 3.2566113471984863
    },
    {
      "auxiliary_loss_clip": 0.0115175,
      "auxiliary_loss_mlp": 0.01023247,
      "balance_loss_clip": 1.04513717,
      "balance_loss_mlp": 1.01659214,
      "epoch": 0.9576143810497204,
      "flos": 32307178435200.0,
      "grad_norm": 2.5505086838690447,
      "language_loss": 0.65816391,
      "learning_rate": 1.8766715727287053e-08,
      "loss": 0.67991382,
      "num_input_tokens_seen": 171908200,
      "step": 7964,
      "time_per_iteration": 2.5714282989501953
    },
    {
      "auxiliary_loss_clip": 0.01152868,
      "auxiliary_loss_mlp": 0.00761068,
      "balance_loss_clip": 1.04611385,
      "balance_loss_mlp": 1.00034475,
      "epoch": 0.9577346239403596,
      "flos": 27745733376000.0,
      "grad_norm": 1.688603296093393,
      "language_loss": 0.79251325,
      "learning_rate": 1.8660404009340546e-08,
      "loss": 0.81165266,
      "num_input_tokens_seen": 171928650,
      "step": 7965,
      "time_per_iteration": 2.526607036590576
    },
    {
      "auxiliary_loss_clip": 0.01045076,
      "auxiliary_loss_mlp": 0.01001168,
      "balance_loss_clip": 1.0073595,
      "balance_loss_mlp": 1.00007689,
      "epoch": 0.9578548668309986,
      "flos": 57468313710720.0,
      "grad_norm": 0.8684609969920518,
      "language_loss": 0.59510231,
      "learning_rate": 1.8554392859035485e-08,
      "loss": 0.61556476,
      "num_input_tokens_seen": 171986400,
      "step": 7966,
      "time_per_iteration": 3.0949604511260986
    },
    {
      "auxiliary_loss_clip": 0.0108195,
      "auxiliary_loss_mlp": 0.01024565,
      "balance_loss_clip": 1.03858936,
      "balance_loss_mlp": 1.01751423,
      "epoch": 0.9579751097216377,
      "flos": 19756040503680.0,
      "grad_norm": 1.6473451490826385,
      "language_loss": 0.78981209,
      "learning_rate": 1.8448682292453444e-08,
      "loss": 0.8108772,
      "num_input_tokens_seen": 172005475,
      "step": 7967,
      "time_per_iteration": 2.6576881408691406
    },
    {
      "auxiliary_loss_clip": 0.01163706,
      "auxiliary_loss_mlp": 0.01026308,
      "balance_loss_clip": 1.04761469,
      "balance_loss_mlp": 1.01947129,
      "epoch": 0.9580953526122769,
      "flos": 18041270152320.0,
      "grad_norm": 1.6350014146219816,
      "language_loss": 0.65824908,
      "learning_rate": 1.8343272325631154e-08,
      "loss": 0.6801492,
      "num_input_tokens_seen": 172024420,
      "step": 7968,
      "time_per_iteration": 2.4339351654052734
    },
    {
      "auxiliary_loss_clip": 0.01084061,
      "auxiliary_loss_mlp": 0.00760742,
      "balance_loss_clip": 1.03899169,
      "balance_loss_mlp": 1.00028443,
      "epoch": 0.9582155955029159,
      "flos": 24270154416000.0,
      "grad_norm": 2.246627474268514,
      "language_loss": 0.77989918,
      "learning_rate": 1.8238162974558492e-08,
      "loss": 0.79834723,
      "num_input_tokens_seen": 172038350,
      "step": 7969,
      "time_per_iteration": 2.6724510192871094
    },
    {
      "auxiliary_loss_clip": 0.01134468,
      "auxiliary_loss_mlp": 0.010283,
      "balance_loss_clip": 1.04542005,
      "balance_loss_mlp": 1.02136171,
      "epoch": 0.958335838393555,
      "flos": 22783309816320.0,
      "grad_norm": 1.875044826283786,
      "language_loss": 0.74728358,
      "learning_rate": 1.8133354255181144e-08,
      "loss": 0.76891124,
      "num_input_tokens_seen": 172058665,
      "step": 7970,
      "time_per_iteration": 2.5501389503479004
    },
    {
      "auxiliary_loss_clip": 0.01143606,
      "auxiliary_loss_mlp": 0.01022564,
      "balance_loss_clip": 1.04420853,
      "balance_loss_mlp": 1.01590061,
      "epoch": 0.958456081284194,
      "flos": 16911484698240.0,
      "grad_norm": 1.6525956815057354,
      "language_loss": 0.74900681,
      "learning_rate": 1.802884618339795e-08,
      "loss": 0.77066845,
      "num_input_tokens_seen": 172077470,
      "step": 7971,
      "time_per_iteration": 2.5046470165252686
    },
    {
      "auxiliary_loss_clip": 0.01153006,
      "auxiliary_loss_mlp": 0.01029262,
      "balance_loss_clip": 1.04756451,
      "balance_loss_mlp": 1.02180207,
      "epoch": 0.9585763241748332,
      "flos": 19974951941760.0,
      "grad_norm": 1.9007492823122745,
      "language_loss": 0.80918318,
      "learning_rate": 1.7924638775062894e-08,
      "loss": 0.83100581,
      "num_input_tokens_seen": 172096590,
      "step": 7972,
      "time_per_iteration": 2.4831409454345703
    },
    {
      "auxiliary_loss_clip": 0.01118412,
      "auxiliary_loss_mlp": 0.01024822,
      "balance_loss_clip": 1.04557991,
      "balance_loss_mlp": 1.01789021,
      "epoch": 0.9586965670654722,
      "flos": 21395649646080.0,
      "grad_norm": 1.9689571904760679,
      "language_loss": 0.81484783,
      "learning_rate": 1.7820732045984444e-08,
      "loss": 0.83628023,
      "num_input_tokens_seen": 172116735,
      "step": 7973,
      "time_per_iteration": 2.592085361480713
    },
    {
      "auxiliary_loss_clip": 0.01144725,
      "auxiliary_loss_mlp": 0.01030861,
      "balance_loss_clip": 1.04435754,
      "balance_loss_mlp": 1.02311254,
      "epoch": 0.9588168099561113,
      "flos": 21435115714560.0,
      "grad_norm": 1.7689515913873297,
      "language_loss": 0.73986429,
      "learning_rate": 1.7717126011924655e-08,
      "loss": 0.76162016,
      "num_input_tokens_seen": 172138320,
      "step": 7974,
      "time_per_iteration": 3.2654218673706055
    },
    {
      "auxiliary_loss_clip": 0.01099422,
      "auxiliary_loss_mlp": 0.01029322,
      "balance_loss_clip": 1.03684878,
      "balance_loss_mlp": 1.0220356,
      "epoch": 0.9589370528467505,
      "flos": 11763761852160.0,
      "grad_norm": 2.5049492259722537,
      "language_loss": 0.7601344,
      "learning_rate": 1.7613820688600957e-08,
      "loss": 0.78142184,
      "num_input_tokens_seen": 172154225,
      "step": 7975,
      "time_per_iteration": 2.604381561279297
    },
    {
      "auxiliary_loss_clip": 0.01137727,
      "auxiliary_loss_mlp": 0.01021631,
      "balance_loss_clip": 1.04508519,
      "balance_loss_mlp": 1.01509833,
      "epoch": 0.9590572957373895,
      "flos": 23441516588160.0,
      "grad_norm": 1.8710593928655854,
      "language_loss": 0.78712595,
      "learning_rate": 1.7510816091684588e-08,
      "loss": 0.80871952,
      "num_input_tokens_seen": 172174150,
      "step": 7976,
      "time_per_iteration": 2.5552797317504883
    },
    {
      "auxiliary_loss_clip": 0.01136931,
      "auxiliary_loss_mlp": 0.01028834,
      "balance_loss_clip": 1.04613435,
      "balance_loss_mlp": 1.02098393,
      "epoch": 0.9591775386280286,
      "flos": 22528272274560.0,
      "grad_norm": 3.3617539517850727,
      "language_loss": 0.78402609,
      "learning_rate": 1.740811223680083e-08,
      "loss": 0.80568373,
      "num_input_tokens_seen": 172191005,
      "step": 7977,
      "time_per_iteration": 2.539546012878418
    },
    {
      "auxiliary_loss_clip": 0.01163127,
      "auxiliary_loss_mlp": 0.01026906,
      "balance_loss_clip": 1.04817355,
      "balance_loss_mlp": 1.01989102,
      "epoch": 0.9592977815186677,
      "flos": 18186959715840.0,
      "grad_norm": 3.902698516837599,
      "language_loss": 0.74071181,
      "learning_rate": 1.7305709139530334e-08,
      "loss": 0.7626121,
      "num_input_tokens_seen": 172209785,
      "step": 7978,
      "time_per_iteration": 2.4497592449188232
    },
    {
      "auxiliary_loss_clip": 0.01141815,
      "auxiliary_loss_mlp": 0.01021717,
      "balance_loss_clip": 1.04331708,
      "balance_loss_mlp": 1.01474333,
      "epoch": 0.9594180244093068,
      "flos": 16537797555840.0,
      "grad_norm": 2.2719415736336988,
      "language_loss": 0.74410778,
      "learning_rate": 1.7203606815407334e-08,
      "loss": 0.76574308,
      "num_input_tokens_seen": 172224380,
      "step": 7979,
      "time_per_iteration": 2.45404314994812
    },
    {
      "auxiliary_loss_clip": 0.01139608,
      "auxiliary_loss_mlp": 0.01028009,
      "balance_loss_clip": 1.04702401,
      "balance_loss_mlp": 1.02060664,
      "epoch": 0.9595382672999458,
      "flos": 20554334317440.0,
      "grad_norm": 1.674285512132708,
      "language_loss": 0.79451597,
      "learning_rate": 1.7101805279920557e-08,
      "loss": 0.81619209,
      "num_input_tokens_seen": 172242540,
      "step": 7980,
      "time_per_iteration": 2.523052215576172
    },
    {
      "auxiliary_loss_clip": 0.01163699,
      "auxiliary_loss_mlp": 0.01027359,
      "balance_loss_clip": 1.04839087,
      "balance_loss_mlp": 1.02066231,
      "epoch": 0.959658510190585,
      "flos": 22638266697600.0,
      "grad_norm": 1.9170229606506757,
      "language_loss": 0.80864215,
      "learning_rate": 1.7000304548513643e-08,
      "loss": 0.8305527,
      "num_input_tokens_seen": 172262645,
      "step": 7981,
      "time_per_iteration": 2.4750542640686035
    },
    {
      "auxiliary_loss_clip": 0.01117557,
      "auxiliary_loss_mlp": 0.01029802,
      "balance_loss_clip": 1.04205489,
      "balance_loss_mlp": 1.02269757,
      "epoch": 0.9597787530812241,
      "flos": 19135252725120.0,
      "grad_norm": 1.9069935938248344,
      "language_loss": 0.82814127,
      "learning_rate": 1.6899104636583394e-08,
      "loss": 0.84961486,
      "num_input_tokens_seen": 172280695,
      "step": 7982,
      "time_per_iteration": 2.5450358390808105
    },
    {
      "auxiliary_loss_clip": 0.01045617,
      "auxiliary_loss_mlp": 0.01002766,
      "balance_loss_clip": 1.00732327,
      "balance_loss_mlp": 1.00166321,
      "epoch": 0.9598989959718631,
      "flos": 60098124055680.0,
      "grad_norm": 0.7226842863766808,
      "language_loss": 0.61933124,
      "learning_rate": 1.6798205559482638e-08,
      "loss": 0.63981509,
      "num_input_tokens_seen": 172343075,
      "step": 7983,
      "time_per_iteration": 3.2594106197357178
    },
    {
      "auxiliary_loss_clip": 0.01120318,
      "auxiliary_loss_mlp": 0.01022001,
      "balance_loss_clip": 1.04369044,
      "balance_loss_mlp": 1.01565909,
      "epoch": 0.9600192388625023,
      "flos": 20886795624960.0,
      "grad_norm": 1.718562349535918,
      "language_loss": 0.76443392,
      "learning_rate": 1.669760733251713e-08,
      "loss": 0.78585708,
      "num_input_tokens_seen": 172361950,
      "step": 7984,
      "time_per_iteration": 2.5551419258117676
    },
    {
      "auxiliary_loss_clip": 0.01098306,
      "auxiliary_loss_mlp": 0.010256,
      "balance_loss_clip": 1.04177082,
      "balance_loss_mlp": 1.0195713,
      "epoch": 0.9601394817531413,
      "flos": 20445740524800.0,
      "grad_norm": 1.936057030903933,
      "language_loss": 0.82363009,
      "learning_rate": 1.659730997094755e-08,
      "loss": 0.8448692,
      "num_input_tokens_seen": 172380440,
      "step": 7985,
      "time_per_iteration": 2.6322832107543945
    },
    {
      "auxiliary_loss_clip": 0.011418,
      "auxiliary_loss_mlp": 0.01025123,
      "balance_loss_clip": 1.04342496,
      "balance_loss_mlp": 1.01866817,
      "epoch": 0.9602597246437804,
      "flos": 21507152440320.0,
      "grad_norm": 1.8587172726778574,
      "language_loss": 0.61934876,
      "learning_rate": 1.6497313489989283e-08,
      "loss": 0.64101803,
      "num_input_tokens_seen": 172400265,
      "step": 7986,
      "time_per_iteration": 3.2295594215393066
    },
    {
      "auxiliary_loss_clip": 0.0110375,
      "auxiliary_loss_mlp": 0.01023529,
      "balance_loss_clip": 1.03555393,
      "balance_loss_mlp": 1.01674032,
      "epoch": 0.9603799675344196,
      "flos": 29935099152000.0,
      "grad_norm": 2.992291760722505,
      "language_loss": 0.70147073,
      "learning_rate": 1.639761790481131e-08,
      "loss": 0.72274351,
      "num_input_tokens_seen": 172421145,
      "step": 7987,
      "time_per_iteration": 3.438467025756836
    },
    {
      "auxiliary_loss_clip": 0.01148115,
      "auxiliary_loss_mlp": 0.01027148,
      "balance_loss_clip": 1.04554462,
      "balance_loss_mlp": 1.02030492,
      "epoch": 0.9605002104250586,
      "flos": 28001525103360.0,
      "grad_norm": 1.9602602835153848,
      "language_loss": 0.79494596,
      "learning_rate": 1.6298223230537754e-08,
      "loss": 0.81669861,
      "num_input_tokens_seen": 172438945,
      "step": 7988,
      "time_per_iteration": 2.5490968227386475
    },
    {
      "auxiliary_loss_clip": 0.01134268,
      "auxiliary_loss_mlp": 0.00761482,
      "balance_loss_clip": 1.04463458,
      "balance_loss_mlp": 1.0003047,
      "epoch": 0.9606204533156977,
      "flos": 35590490870400.0,
      "grad_norm": 1.8592926522215996,
      "language_loss": 0.69862187,
      "learning_rate": 1.619912948224611e-08,
      "loss": 0.71757936,
      "num_input_tokens_seen": 172460150,
      "step": 7989,
      "time_per_iteration": 3.392298460006714
    },
    {
      "auxiliary_loss_clip": 0.01116401,
      "auxiliary_loss_mlp": 0.01027179,
      "balance_loss_clip": 1.0429635,
      "balance_loss_mlp": 1.02043796,
      "epoch": 0.9607406962063368,
      "flos": 26574614346240.0,
      "grad_norm": 2.104105829381176,
      "language_loss": 0.61350882,
      "learning_rate": 1.6100336674969682e-08,
      "loss": 0.63494462,
      "num_input_tokens_seen": 172478990,
      "step": 7990,
      "time_per_iteration": 2.598083019256592
    },
    {
      "auxiliary_loss_clip": 0.01105148,
      "auxiliary_loss_mlp": 0.01029362,
      "balance_loss_clip": 1.03935313,
      "balance_loss_mlp": 1.02231371,
      "epoch": 0.9608609390969759,
      "flos": 25331781813120.0,
      "grad_norm": 1.6998164256453825,
      "language_loss": 0.76649106,
      "learning_rate": 1.600184482369449e-08,
      "loss": 0.78783613,
      "num_input_tokens_seen": 172498905,
      "step": 7991,
      "time_per_iteration": 2.6181211471557617
    },
    {
      "auxiliary_loss_clip": 0.01122706,
      "auxiliary_loss_mlp": 0.01026539,
      "balance_loss_clip": 1.04163432,
      "balance_loss_mlp": 1.01827741,
      "epoch": 0.960981181987615,
      "flos": 21069114082560.0,
      "grad_norm": 2.1901811447142285,
      "language_loss": 0.89137876,
      "learning_rate": 1.5903653943362126e-08,
      "loss": 0.91287124,
      "num_input_tokens_seen": 172517900,
      "step": 7992,
      "time_per_iteration": 2.555037498474121
    },
    {
      "auxiliary_loss_clip": 0.01137653,
      "auxiliary_loss_mlp": 0.01031259,
      "balance_loss_clip": 1.04616225,
      "balance_loss_mlp": 1.02453279,
      "epoch": 0.9611014248782541,
      "flos": 17823256554240.0,
      "grad_norm": 1.836182927726961,
      "language_loss": 0.77256715,
      "learning_rate": 1.580576404886802e-08,
      "loss": 0.79425633,
      "num_input_tokens_seen": 172536430,
      "step": 7993,
      "time_per_iteration": 2.538790225982666
    },
    {
      "auxiliary_loss_clip": 0.01148479,
      "auxiliary_loss_mlp": 0.01023767,
      "balance_loss_clip": 1.04520273,
      "balance_loss_mlp": 1.01762521,
      "epoch": 0.9612216677688932,
      "flos": 19354631040000.0,
      "grad_norm": 1.8825170543333891,
      "language_loss": 0.79666519,
      "learning_rate": 1.570817515506162e-08,
      "loss": 0.81838763,
      "num_input_tokens_seen": 172555120,
      "step": 7994,
      "time_per_iteration": 2.499115467071533
    },
    {
      "auxiliary_loss_clip": 0.01161538,
      "auxiliary_loss_mlp": 0.01027221,
      "balance_loss_clip": 1.04777336,
      "balance_loss_mlp": 1.02045608,
      "epoch": 0.9613419106595322,
      "flos": 15808739207040.0,
      "grad_norm": 2.057360885495906,
      "language_loss": 0.81423044,
      "learning_rate": 1.561088727674753e-08,
      "loss": 0.83611798,
      "num_input_tokens_seen": 172569330,
      "step": 7995,
      "time_per_iteration": 2.42051100730896
    },
    {
      "auxiliary_loss_clip": 0.01119659,
      "auxiliary_loss_mlp": 0.01030684,
      "balance_loss_clip": 1.04361892,
      "balance_loss_mlp": 1.02332008,
      "epoch": 0.9614621535501714,
      "flos": 25702488126720.0,
      "grad_norm": 2.2726108391107225,
      "language_loss": 0.71202362,
      "learning_rate": 1.551390042868417e-08,
      "loss": 0.73352695,
      "num_input_tokens_seen": 172591100,
      "step": 7996,
      "time_per_iteration": 2.6651251316070557
    },
    {
      "auxiliary_loss_clip": 0.01149182,
      "auxiliary_loss_mlp": 0.01029836,
      "balance_loss_clip": 1.04587507,
      "balance_loss_mlp": 1.02305591,
      "epoch": 0.9615823964408104,
      "flos": 17819054663040.0,
      "grad_norm": 1.6851700242856786,
      "language_loss": 0.70589644,
      "learning_rate": 1.5417214625584207e-08,
      "loss": 0.72768658,
      "num_input_tokens_seen": 172608755,
      "step": 7997,
      "time_per_iteration": 2.4734814167022705
    },
    {
      "auxiliary_loss_clip": 0.01143264,
      "auxiliary_loss_mlp": 0.01022389,
      "balance_loss_clip": 1.0436101,
      "balance_loss_mlp": 1.01531148,
      "epoch": 0.9617026393314495,
      "flos": 20190020624640.0,
      "grad_norm": 1.7566921800583892,
      "language_loss": 0.85435522,
      "learning_rate": 1.5320829882114806e-08,
      "loss": 0.87601173,
      "num_input_tokens_seen": 172626830,
      "step": 7998,
      "time_per_iteration": 2.4975128173828125
    },
    {
      "auxiliary_loss_clip": 0.01161409,
      "auxiliary_loss_mlp": 0.01026055,
      "balance_loss_clip": 1.04583633,
      "balance_loss_mlp": 1.01881349,
      "epoch": 0.9618228822220887,
      "flos": 20267013427200.0,
      "grad_norm": 1.849774283957806,
      "language_loss": 0.78620625,
      "learning_rate": 1.5224746212897378e-08,
      "loss": 0.80808085,
      "num_input_tokens_seen": 172646125,
      "step": 7999,
      "time_per_iteration": 2.464493989944458
    },
    {
      "auxiliary_loss_clip": 0.01158189,
      "auxiliary_loss_mlp": 0.01026117,
      "balance_loss_clip": 1.04587173,
      "balance_loss_mlp": 1.01898265,
      "epoch": 0.9619431251127277,
      "flos": 21031300039680.0,
      "grad_norm": 1.6863043696975948,
      "language_loss": 0.77466083,
      "learning_rate": 1.512896363250804e-08,
      "loss": 0.7965039,
      "num_input_tokens_seen": 172666235,
      "step": 8000,
      "time_per_iteration": 2.4748361110687256
    },
    {
      "auxiliary_loss_clip": 0.011511,
      "auxiliary_loss_mlp": 0.01026543,
      "balance_loss_clip": 1.0450511,
      "balance_loss_mlp": 1.01952446,
      "epoch": 0.9620633680033668,
      "flos": 22382654538240.0,
      "grad_norm": 2.0285535344219956,
      "language_loss": 0.75599778,
      "learning_rate": 1.503348215547673e-08,
      "loss": 0.77777421,
      "num_input_tokens_seen": 172687325,
      "step": 8001,
      "time_per_iteration": 3.2231593132019043
    },
    {
      "auxiliary_loss_clip": 0.01132075,
      "auxiliary_loss_mlp": 0.01024739,
      "balance_loss_clip": 1.04357624,
      "balance_loss_mlp": 1.01775897,
      "epoch": 0.962183610894006,
      "flos": 18471730740480.0,
      "grad_norm": 1.7545911779760777,
      "language_loss": 0.80729938,
      "learning_rate": 1.4938301796288078e-08,
      "loss": 0.82886755,
      "num_input_tokens_seen": 172703895,
      "step": 8002,
      "time_per_iteration": 2.502425193786621
    },
    {
      "auxiliary_loss_clip": 0.01163318,
      "auxiliary_loss_mlp": 0.01032466,
      "balance_loss_clip": 1.04760408,
      "balance_loss_mlp": 1.02527165,
      "epoch": 0.962303853784645,
      "flos": 18435245500800.0,
      "grad_norm": 2.137984722156509,
      "language_loss": 0.81459463,
      "learning_rate": 1.4843422569380537e-08,
      "loss": 0.8365525,
      "num_input_tokens_seen": 172720650,
      "step": 8003,
      "time_per_iteration": 2.41196870803833
    },
    {
      "auxiliary_loss_clip": 0.01101641,
      "auxiliary_loss_mlp": 0.01026637,
      "balance_loss_clip": 1.03995061,
      "balance_loss_mlp": 1.01965094,
      "epoch": 0.9624240966752841,
      "flos": 26391074826240.0,
      "grad_norm": 1.6015325716235473,
      "language_loss": 0.83028871,
      "learning_rate": 1.4748844489147483e-08,
      "loss": 0.85157144,
      "num_input_tokens_seen": 172737640,
      "step": 8004,
      "time_per_iteration": 2.5983951091766357
    },
    {
      "auxiliary_loss_clip": 0.01134274,
      "auxiliary_loss_mlp": 0.01024603,
      "balance_loss_clip": 1.04218674,
      "balance_loss_mlp": 1.0183773,
      "epoch": 0.9625443395659231,
      "flos": 14647675985280.0,
      "grad_norm": 1.877892659116549,
      "language_loss": 0.70602912,
      "learning_rate": 1.4654567569936326e-08,
      "loss": 0.72761786,
      "num_input_tokens_seen": 172755215,
      "step": 8005,
      "time_per_iteration": 2.4951412677764893
    },
    {
      "auxiliary_loss_clip": 0.01103083,
      "auxiliary_loss_mlp": 0.01026352,
      "balance_loss_clip": 1.04124856,
      "balance_loss_mlp": 1.01883626,
      "epoch": 0.9626645824565623,
      "flos": 18367626147840.0,
      "grad_norm": 2.005413088382172,
      "language_loss": 0.83309728,
      "learning_rate": 1.456059182604874e-08,
      "loss": 0.85439163,
      "num_input_tokens_seen": 172774020,
      "step": 8006,
      "time_per_iteration": 2.5613787174224854
    },
    {
      "auxiliary_loss_clip": 0.01164235,
      "auxiliary_loss_mlp": 0.01027482,
      "balance_loss_clip": 1.0485431,
      "balance_loss_mlp": 1.02023065,
      "epoch": 0.9627848253472013,
      "flos": 16580424021120.0,
      "grad_norm": 1.7753137676770518,
      "language_loss": 0.76191568,
      "learning_rate": 1.4466917271740653e-08,
      "loss": 0.78383291,
      "num_input_tokens_seen": 172792220,
      "step": 8007,
      "time_per_iteration": 2.4156441688537598
    },
    {
      "auxiliary_loss_clip": 0.01130965,
      "auxiliary_loss_mlp": 0.01029711,
      "balance_loss_clip": 1.04409993,
      "balance_loss_mlp": 1.02215874,
      "epoch": 0.9629050682378404,
      "flos": 20886867452160.0,
      "grad_norm": 1.9766583494069903,
      "language_loss": 0.67520797,
      "learning_rate": 1.4373543921222697e-08,
      "loss": 0.69681472,
      "num_input_tokens_seen": 172811805,
      "step": 8008,
      "time_per_iteration": 2.5161733627319336
    },
    {
      "auxiliary_loss_clip": 0.01131438,
      "auxiliary_loss_mlp": 0.0101967,
      "balance_loss_clip": 1.04498756,
      "balance_loss_mlp": 1.01262498,
      "epoch": 0.9630253111284796,
      "flos": 17019252478080.0,
      "grad_norm": 1.9634002622344724,
      "language_loss": 0.77854943,
      "learning_rate": 1.428047178865932e-08,
      "loss": 0.80006051,
      "num_input_tokens_seen": 172828595,
      "step": 8009,
      "time_per_iteration": 2.486952781677246
    },
    {
      "auxiliary_loss_clip": 0.01130944,
      "auxiliary_loss_mlp": 0.01023636,
      "balance_loss_clip": 1.04096889,
      "balance_loss_mlp": 1.01648903,
      "epoch": 0.9631455540191186,
      "flos": 20338942412160.0,
      "grad_norm": 1.5905516074286141,
      "language_loss": 0.74504638,
      "learning_rate": 1.4187700888169451e-08,
      "loss": 0.76659214,
      "num_input_tokens_seen": 172847770,
      "step": 8010,
      "time_per_iteration": 2.5184552669525146
    },
    {
      "auxiliary_loss_clip": 0.01044979,
      "auxiliary_loss_mlp": 0.01001636,
      "balance_loss_clip": 1.00886941,
      "balance_loss_mlp": 1.00052691,
      "epoch": 0.9632657969097577,
      "flos": 65956700033280.0,
      "grad_norm": 0.7794178827847943,
      "language_loss": 0.57004815,
      "learning_rate": 1.40952312338265e-08,
      "loss": 0.5905143,
      "num_input_tokens_seen": 172912415,
      "step": 8011,
      "time_per_iteration": 3.1478281021118164
    },
    {
      "auxiliary_loss_clip": 0.01122401,
      "auxiliary_loss_mlp": 0.01025662,
      "balance_loss_clip": 1.04165483,
      "balance_loss_mlp": 1.0190165,
      "epoch": 0.9633860398003968,
      "flos": 44419523823360.0,
      "grad_norm": 1.724001864901442,
      "language_loss": 0.6837216,
      "learning_rate": 1.4003062839657909e-08,
      "loss": 0.70520228,
      "num_input_tokens_seen": 172934895,
      "step": 8012,
      "time_per_iteration": 3.5965487957000732
    },
    {
      "auxiliary_loss_clip": 0.01120899,
      "auxiliary_loss_mlp": 0.01024161,
      "balance_loss_clip": 1.04278278,
      "balance_loss_mlp": 1.01685667,
      "epoch": 0.9635062826910359,
      "flos": 24827704300800.0,
      "grad_norm": 1.6030984427089152,
      "language_loss": 0.80081511,
      "learning_rate": 1.391119571964583e-08,
      "loss": 0.82226562,
      "num_input_tokens_seen": 172955835,
      "step": 8013,
      "time_per_iteration": 3.373582363128662
    },
    {
      "auxiliary_loss_clip": 0.01145904,
      "auxiliary_loss_mlp": 0.0102656,
      "balance_loss_clip": 1.04472685,
      "balance_loss_mlp": 1.02010202,
      "epoch": 0.9636265255816749,
      "flos": 15961360095360.0,
      "grad_norm": 1.8235726575009459,
      "language_loss": 0.72715318,
      "learning_rate": 1.3819629887726225e-08,
      "loss": 0.74887782,
      "num_input_tokens_seen": 172973925,
      "step": 8014,
      "time_per_iteration": 2.4722163677215576
    },
    {
      "auxiliary_loss_clip": 0.01141567,
      "auxiliary_loss_mlp": 0.01023205,
      "balance_loss_clip": 1.04645801,
      "balance_loss_mlp": 1.01639795,
      "epoch": 0.9637467684723141,
      "flos": 22601781457920.0,
      "grad_norm": 1.7471794693052107,
      "language_loss": 0.76093793,
      "learning_rate": 1.3728365357789317e-08,
      "loss": 0.78258562,
      "num_input_tokens_seen": 172993290,
      "step": 8015,
      "time_per_iteration": 3.3034112453460693
    },
    {
      "auxiliary_loss_clip": 0.01081123,
      "auxiliary_loss_mlp": 0.01030377,
      "balance_loss_clip": 1.03944802,
      "balance_loss_mlp": 1.02334976,
      "epoch": 0.9638670113629532,
      "flos": 17565812801280.0,
      "grad_norm": 4.740638599253653,
      "language_loss": 0.76497662,
      "learning_rate": 1.3637402143680254e-08,
      "loss": 0.78609163,
      "num_input_tokens_seen": 173008190,
      "step": 8016,
      "time_per_iteration": 2.6376500129699707
    },
    {
      "auxiliary_loss_clip": 0.0101512,
      "auxiliary_loss_mlp": 0.01001466,
      "balance_loss_clip": 1.00894451,
      "balance_loss_mlp": 1.00045824,
      "epoch": 0.9639872542535922,
      "flos": 55072139379840.0,
      "grad_norm": 0.720713846831545,
      "language_loss": 0.55068421,
      "learning_rate": 1.3546740259197998e-08,
      "loss": 0.57085013,
      "num_input_tokens_seen": 173061000,
      "step": 8017,
      "time_per_iteration": 3.090174913406372
    },
    {
      "auxiliary_loss_clip": 0.01134884,
      "auxiliary_loss_mlp": 0.01031663,
      "balance_loss_clip": 1.04525805,
      "balance_loss_mlp": 1.02403975,
      "epoch": 0.9641074971442314,
      "flos": 24134484746880.0,
      "grad_norm": 1.8184694991409707,
      "language_loss": 0.69451487,
      "learning_rate": 1.3456379718095989e-08,
      "loss": 0.71618032,
      "num_input_tokens_seen": 173081415,
      "step": 8018,
      "time_per_iteration": 2.5589301586151123
    },
    {
      "auxiliary_loss_clip": 0.01033026,
      "auxiliary_loss_mlp": 0.01002611,
      "balance_loss_clip": 1.00700903,
      "balance_loss_mlp": 1.00152004,
      "epoch": 0.9642277400348704,
      "flos": 66747416077440.0,
      "grad_norm": 0.8369845924738059,
      "language_loss": 0.62050074,
      "learning_rate": 1.3366320534081487e-08,
      "loss": 0.6408571,
      "num_input_tokens_seen": 173144095,
      "step": 8019,
      "time_per_iteration": 3.133159875869751
    },
    {
      "auxiliary_loss_clip": 0.01148231,
      "auxiliary_loss_mlp": 0.01025795,
      "balance_loss_clip": 1.04605758,
      "balance_loss_mlp": 1.0185585,
      "epoch": 0.9643479829255095,
      "flos": 30920272450560.0,
      "grad_norm": 2.0691412256184565,
      "language_loss": 0.75809002,
      "learning_rate": 1.3276562720816675e-08,
      "loss": 0.77983034,
      "num_input_tokens_seen": 173165605,
      "step": 8020,
      "time_per_iteration": 2.556793451309204
    },
    {
      "auxiliary_loss_clip": 0.0116206,
      "auxiliary_loss_mlp": 0.01022247,
      "balance_loss_clip": 1.04591751,
      "balance_loss_mlp": 1.01497817,
      "epoch": 0.9644682258161487,
      "flos": 20048245643520.0,
      "grad_norm": 1.9585675962535611,
      "language_loss": 0.82268924,
      "learning_rate": 1.3187106291917549e-08,
      "loss": 0.84453225,
      "num_input_tokens_seen": 173182595,
      "step": 8021,
      "time_per_iteration": 2.4417073726654053
    },
    {
      "auxiliary_loss_clip": 0.01144539,
      "auxiliary_loss_mlp": 0.01022393,
      "balance_loss_clip": 1.04507875,
      "balance_loss_mlp": 1.01549625,
      "epoch": 0.9645884687067877,
      "flos": 21178713456000.0,
      "grad_norm": 1.8680588291869193,
      "language_loss": 0.70478874,
      "learning_rate": 1.309795126095503e-08,
      "loss": 0.72645807,
      "num_input_tokens_seen": 173200895,
      "step": 8022,
      "time_per_iteration": 2.473043441772461
    },
    {
      "auxiliary_loss_clip": 0.01073435,
      "auxiliary_loss_mlp": 0.01023473,
      "balance_loss_clip": 1.03568912,
      "balance_loss_mlp": 1.01645446,
      "epoch": 0.9647087115974268,
      "flos": 18945967029120.0,
      "grad_norm": 2.0155131711993794,
      "language_loss": 0.80428147,
      "learning_rate": 1.3009097641453192e-08,
      "loss": 0.82525051,
      "num_input_tokens_seen": 173218745,
      "step": 8023,
      "time_per_iteration": 2.636964797973633
    },
    {
      "auxiliary_loss_clip": 0.01131064,
      "auxiliary_loss_mlp": 0.01021894,
      "balance_loss_clip": 1.04320383,
      "balance_loss_mlp": 1.01537037,
      "epoch": 0.9648289544880659,
      "flos": 16545088016640.0,
      "grad_norm": 1.7013841529184617,
      "language_loss": 0.75984609,
      "learning_rate": 1.2920545446891474e-08,
      "loss": 0.78137571,
      "num_input_tokens_seen": 173235465,
      "step": 8024,
      "time_per_iteration": 2.499638557434082
    },
    {
      "auxiliary_loss_clip": 0.01139012,
      "auxiliary_loss_mlp": 0.01031999,
      "balance_loss_clip": 1.04730082,
      "balance_loss_mlp": 1.02530241,
      "epoch": 0.964949197378705,
      "flos": 24057527857920.0,
      "grad_norm": 1.6335595392790765,
      "language_loss": 0.70644033,
      "learning_rate": 1.2832294690703127e-08,
      "loss": 0.72815037,
      "num_input_tokens_seen": 173254440,
      "step": 8025,
      "time_per_iteration": 2.5447640419006348
    },
    {
      "auxiliary_loss_clip": 0.01148508,
      "auxiliary_loss_mlp": 0.01026559,
      "balance_loss_clip": 1.04738665,
      "balance_loss_mlp": 1.01968956,
      "epoch": 0.965069440269344,
      "flos": 23365565280000.0,
      "grad_norm": 1.8873550800801469,
      "language_loss": 0.77633178,
      "learning_rate": 1.2744345386275668e-08,
      "loss": 0.79808241,
      "num_input_tokens_seen": 173273980,
      "step": 8026,
      "time_per_iteration": 3.336887836456299
    },
    {
      "auxiliary_loss_clip": 0.01138753,
      "auxiliary_loss_mlp": 0.0102854,
      "balance_loss_clip": 1.04812884,
      "balance_loss_mlp": 1.02171838,
      "epoch": 0.9651896831599832,
      "flos": 25374875155200.0,
      "grad_norm": 2.591240529162104,
      "language_loss": 0.78804624,
      "learning_rate": 1.265669754695109e-08,
      "loss": 0.80971915,
      "num_input_tokens_seen": 173293550,
      "step": 8027,
      "time_per_iteration": 2.5648958683013916
    },
    {
      "auxiliary_loss_clip": 0.01092723,
      "auxiliary_loss_mlp": 0.01028962,
      "balance_loss_clip": 1.03666782,
      "balance_loss_mlp": 1.02228951,
      "epoch": 0.9653099260506223,
      "flos": 22272875596800.0,
      "grad_norm": 2.003720901965906,
      "language_loss": 0.82313472,
      "learning_rate": 1.2569351186025201e-08,
      "loss": 0.84435153,
      "num_input_tokens_seen": 173312005,
      "step": 8028,
      "time_per_iteration": 2.60941481590271
    },
    {
      "auxiliary_loss_clip": 0.0111078,
      "auxiliary_loss_mlp": 0.01027008,
      "balance_loss_clip": 1.04040575,
      "balance_loss_mlp": 1.02040684,
      "epoch": 0.9654301689412613,
      "flos": 26760847386240.0,
      "grad_norm": 1.4928113826956364,
      "language_loss": 0.75279081,
      "learning_rate": 1.2482306316748737e-08,
      "loss": 0.77416867,
      "num_input_tokens_seen": 173332450,
      "step": 8029,
      "time_per_iteration": 2.60005521774292
    },
    {
      "auxiliary_loss_clip": 0.01154075,
      "auxiliary_loss_mlp": 0.01026278,
      "balance_loss_clip": 1.0455364,
      "balance_loss_mlp": 1.01919127,
      "epoch": 0.9655504118319005,
      "flos": 17412689122560.0,
      "grad_norm": 2.009953060543939,
      "language_loss": 0.78113323,
      "learning_rate": 1.2395562952326021e-08,
      "loss": 0.80293679,
      "num_input_tokens_seen": 173349610,
      "step": 8030,
      "time_per_iteration": 2.5015058517456055
    },
    {
      "auxiliary_loss_clip": 0.01145365,
      "auxiliary_loss_mlp": 0.01032734,
      "balance_loss_clip": 1.04739654,
      "balance_loss_mlp": 1.02528977,
      "epoch": 0.9656706547225395,
      "flos": 22126970551680.0,
      "grad_norm": 2.0495203606173953,
      "language_loss": 0.81058371,
      "learning_rate": 1.2309121105916309e-08,
      "loss": 0.83236474,
      "num_input_tokens_seen": 173367900,
      "step": 8031,
      "time_per_iteration": 2.540872573852539
    },
    {
      "auxiliary_loss_clip": 0.01152545,
      "auxiliary_loss_mlp": 0.0102345,
      "balance_loss_clip": 1.04607797,
      "balance_loss_mlp": 1.01616597,
      "epoch": 0.9657908976131786,
      "flos": 37049289926400.0,
      "grad_norm": 1.9882603532194607,
      "language_loss": 0.69072533,
      "learning_rate": 1.222298079063222e-08,
      "loss": 0.71248519,
      "num_input_tokens_seen": 173389040,
      "step": 8032,
      "time_per_iteration": 2.624593734741211
    },
    {
      "auxiliary_loss_clip": 0.01147167,
      "auxiliary_loss_mlp": 0.01029301,
      "balance_loss_clip": 1.04459548,
      "balance_loss_mlp": 1.02285147,
      "epoch": 0.9659111405038178,
      "flos": 24389809597440.0,
      "grad_norm": 2.0659332924208464,
      "language_loss": 0.72624552,
      "learning_rate": 1.2137142019541524e-08,
      "loss": 0.74801022,
      "num_input_tokens_seen": 173407595,
      "step": 8033,
      "time_per_iteration": 2.5246968269348145
    },
    {
      "auxiliary_loss_clip": 0.01138128,
      "auxiliary_loss_mlp": 0.01024463,
      "balance_loss_clip": 1.04474676,
      "balance_loss_mlp": 1.01778173,
      "epoch": 0.9660313833944568,
      "flos": 25009412227200.0,
      "grad_norm": 1.8572177958528888,
      "language_loss": 0.73451936,
      "learning_rate": 1.2051604805666027e-08,
      "loss": 0.7561453,
      "num_input_tokens_seen": 173424720,
      "step": 8034,
      "time_per_iteration": 2.5371508598327637
    },
    {
      "auxiliary_loss_clip": 0.01162816,
      "auxiliary_loss_mlp": 0.00760801,
      "balance_loss_clip": 1.04788685,
      "balance_loss_mlp": 1.00030971,
      "epoch": 0.9661516262850959,
      "flos": 11801575895040.0,
      "grad_norm": 2.0617299561210074,
      "language_loss": 0.78705388,
      "learning_rate": 1.196636916198135e-08,
      "loss": 0.80629009,
      "num_input_tokens_seen": 173442260,
      "step": 8035,
      "time_per_iteration": 2.4587206840515137
    },
    {
      "auxiliary_loss_clip": 0.01165419,
      "auxiliary_loss_mlp": 0.01019921,
      "balance_loss_clip": 1.04791093,
      "balance_loss_mlp": 1.01353776,
      "epoch": 0.9662718691757349,
      "flos": 20047778766720.0,
      "grad_norm": 2.0605757614902607,
      "language_loss": 0.77120352,
      "learning_rate": 1.1881435101418036e-08,
      "loss": 0.79305691,
      "num_input_tokens_seen": 173461675,
      "step": 8036,
      "time_per_iteration": 2.4489939212799072
    },
    {
      "auxiliary_loss_clip": 0.01036253,
      "auxiliary_loss_mlp": 0.01004656,
      "balance_loss_clip": 1.00958371,
      "balance_loss_mlp": 1.00346982,
      "epoch": 0.9663921120663741,
      "flos": 68027703517440.0,
      "grad_norm": 0.726068738057863,
      "language_loss": 0.65608633,
      "learning_rate": 1.1796802636860003e-08,
      "loss": 0.67649543,
      "num_input_tokens_seen": 173530205,
      "step": 8037,
      "time_per_iteration": 3.165302276611328
    },
    {
      "auxiliary_loss_clip": 0.01164414,
      "auxiliary_loss_mlp": 0.01023534,
      "balance_loss_clip": 1.04721463,
      "balance_loss_mlp": 1.01630056,
      "epoch": 0.9665123549570132,
      "flos": 26322916769280.0,
      "grad_norm": 1.8186478840542408,
      "language_loss": 0.73872185,
      "learning_rate": 1.1712471781146316e-08,
      "loss": 0.76060128,
      "num_input_tokens_seen": 173549540,
      "step": 8038,
      "time_per_iteration": 3.3481268882751465
    },
    {
      "auxiliary_loss_clip": 0.01161295,
      "auxiliary_loss_mlp": 0.01022796,
      "balance_loss_clip": 1.04602647,
      "balance_loss_mlp": 1.0158577,
      "epoch": 0.9666325978476522,
      "flos": 43941121557120.0,
      "grad_norm": 1.8243842175417146,
      "language_loss": 0.66707993,
      "learning_rate": 1.1628442547069628e-08,
      "loss": 0.68892074,
      "num_input_tokens_seen": 173571740,
      "step": 8039,
      "time_per_iteration": 3.4186246395111084
    },
    {
      "auxiliary_loss_clip": 0.01151956,
      "auxiliary_loss_mlp": 0.00760898,
      "balance_loss_clip": 1.04544163,
      "balance_loss_mlp": 1.00032139,
      "epoch": 0.9667528407382914,
      "flos": 21543422198400.0,
      "grad_norm": 2.223766572856778,
      "language_loss": 0.77345681,
      "learning_rate": 1.1544714947377521e-08,
      "loss": 0.79258537,
      "num_input_tokens_seen": 173589425,
      "step": 8040,
      "time_per_iteration": 2.502368927001953
    },
    {
      "auxiliary_loss_clip": 0.01165358,
      "auxiliary_loss_mlp": 0.01021985,
      "balance_loss_clip": 1.04885459,
      "balance_loss_mlp": 1.01499605,
      "epoch": 0.9668730836289304,
      "flos": 23878585278720.0,
      "grad_norm": 2.116444685873814,
      "language_loss": 0.69865924,
      "learning_rate": 1.1461288994770945e-08,
      "loss": 0.72053266,
      "num_input_tokens_seen": 173608500,
      "step": 8041,
      "time_per_iteration": 3.158148765563965
    },
    {
      "auxiliary_loss_clip": 0.01164267,
      "auxiliary_loss_mlp": 0.01032041,
      "balance_loss_clip": 1.04609609,
      "balance_loss_mlp": 1.0239706,
      "epoch": 0.9669933265195695,
      "flos": 28293011971200.0,
      "grad_norm": 1.592504663399579,
      "language_loss": 0.77075821,
      "learning_rate": 1.1378164701906002e-08,
      "loss": 0.79272127,
      "num_input_tokens_seen": 173630265,
      "step": 8042,
      "time_per_iteration": 2.507467746734619
    },
    {
      "auxiliary_loss_clip": 0.01164516,
      "auxiliary_loss_mlp": 0.01027153,
      "balance_loss_clip": 1.04721248,
      "balance_loss_mlp": 1.02015853,
      "epoch": 0.9671135694102087,
      "flos": 22454763091200.0,
      "grad_norm": 1.851034624061108,
      "language_loss": 0.6688627,
      "learning_rate": 1.1295342081392156e-08,
      "loss": 0.69077933,
      "num_input_tokens_seen": 173649625,
      "step": 8043,
      "time_per_iteration": 2.441541910171509
    },
    {
      "auxiliary_loss_clip": 0.01134592,
      "auxiliary_loss_mlp": 0.01024117,
      "balance_loss_clip": 1.04525661,
      "balance_loss_mlp": 1.01770914,
      "epoch": 0.9672338123008477,
      "flos": 20155941596160.0,
      "grad_norm": 1.7795042378469135,
      "language_loss": 0.69441521,
      "learning_rate": 1.1212821145793804e-08,
      "loss": 0.71600223,
      "num_input_tokens_seen": 173669240,
      "step": 8044,
      "time_per_iteration": 2.5213499069213867
    },
    {
      "auxiliary_loss_clip": 0.01134966,
      "auxiliary_loss_mlp": 0.01025128,
      "balance_loss_clip": 1.0425694,
      "balance_loss_mlp": 1.01855052,
      "epoch": 0.9673540551914868,
      "flos": 16977487939200.0,
      "grad_norm": 1.971050004116924,
      "language_loss": 0.78719628,
      "learning_rate": 1.1130601907629156e-08,
      "loss": 0.80879724,
      "num_input_tokens_seen": 173686970,
      "step": 8045,
      "time_per_iteration": 2.492058277130127
    },
    {
      "auxiliary_loss_clip": 0.01045608,
      "auxiliary_loss_mlp": 0.01001667,
      "balance_loss_clip": 1.00739837,
      "balance_loss_mlp": 1.00052834,
      "epoch": 0.9674742980821259,
      "flos": 61892903952000.0,
      "grad_norm": 0.8126656294730421,
      "language_loss": 0.64934099,
      "learning_rate": 1.1048684379370899e-08,
      "loss": 0.66981375,
      "num_input_tokens_seen": 173747655,
      "step": 8046,
      "time_per_iteration": 3.052504301071167
    },
    {
      "auxiliary_loss_clip": 0.01124525,
      "auxiliary_loss_mlp": 0.01030192,
      "balance_loss_clip": 1.04240799,
      "balance_loss_mlp": 1.02377534,
      "epoch": 0.967594540972765,
      "flos": 18697824898560.0,
      "grad_norm": 2.3702098241923855,
      "language_loss": 0.74418485,
      "learning_rate": 1.0967068573445759e-08,
      "loss": 0.76573205,
      "num_input_tokens_seen": 173765140,
      "step": 8047,
      "time_per_iteration": 2.4811084270477295
    },
    {
      "auxiliary_loss_clip": 0.0113343,
      "auxiliary_loss_mlp": 0.01025286,
      "balance_loss_clip": 1.04336452,
      "balance_loss_mlp": 1.01824093,
      "epoch": 0.967714783863404,
      "flos": 20777411733120.0,
      "grad_norm": 2.0145606640807414,
      "language_loss": 0.65232885,
      "learning_rate": 1.0885754502234945e-08,
      "loss": 0.67391598,
      "num_input_tokens_seen": 173784800,
      "step": 8048,
      "time_per_iteration": 2.526123046875
    },
    {
      "auxiliary_loss_clip": 0.0111399,
      "auxiliary_loss_mlp": 0.01023044,
      "balance_loss_clip": 1.04245353,
      "balance_loss_mlp": 1.01636231,
      "epoch": 0.9678350267540432,
      "flos": 23185473465600.0,
      "grad_norm": 1.9384649418964794,
      "language_loss": 0.7771014,
      "learning_rate": 1.08047421780737e-08,
      "loss": 0.79847175,
      "num_input_tokens_seen": 173803990,
      "step": 8049,
      "time_per_iteration": 2.5408973693847656
    },
    {
      "auxiliary_loss_clip": 0.01142995,
      "auxiliary_loss_mlp": 0.00760694,
      "balance_loss_clip": 1.0464282,
      "balance_loss_mlp": 1.00026476,
      "epoch": 0.9679552696446823,
      "flos": 21726063878400.0,
      "grad_norm": 1.9868533916047577,
      "language_loss": 0.73874098,
      "learning_rate": 1.0724031613251305e-08,
      "loss": 0.75777787,
      "num_input_tokens_seen": 173821890,
      "step": 8050,
      "time_per_iteration": 2.527113914489746
    },
    {
      "auxiliary_loss_clip": 0.01154689,
      "auxiliary_loss_mlp": 0.01029758,
      "balance_loss_clip": 1.04585457,
      "balance_loss_mlp": 1.02231908,
      "epoch": 0.9680755125353213,
      "flos": 26869046129280.0,
      "grad_norm": 2.193039141365398,
      "language_loss": 0.66236264,
      "learning_rate": 1.0643622820011744e-08,
      "loss": 0.68420708,
      "num_input_tokens_seen": 173842945,
      "step": 8051,
      "time_per_iteration": 2.526540994644165
    },
    {
      "auxiliary_loss_clip": 0.01165673,
      "auxiliary_loss_mlp": 0.01024026,
      "balance_loss_clip": 1.04705775,
      "balance_loss_mlp": 1.01684105,
      "epoch": 0.9681957554259605,
      "flos": 28325008010880.0,
      "grad_norm": 2.220781664766999,
      "language_loss": 0.68001628,
      "learning_rate": 1.0563515810552814e-08,
      "loss": 0.70191324,
      "num_input_tokens_seen": 173859915,
      "step": 8052,
      "time_per_iteration": 3.1951167583465576
    },
    {
      "auxiliary_loss_clip": 0.01163723,
      "auxiliary_loss_mlp": 0.01025297,
      "balance_loss_clip": 1.04853463,
      "balance_loss_mlp": 1.01877046,
      "epoch": 0.9683159983165995,
      "flos": 20557674282240.0,
      "grad_norm": 1.485311455285863,
      "language_loss": 0.73130739,
      "learning_rate": 1.0483710597026795e-08,
      "loss": 0.75319755,
      "num_input_tokens_seen": 173879775,
      "step": 8053,
      "time_per_iteration": 2.4701015949249268
    },
    {
      "auxiliary_loss_clip": 0.01121117,
      "auxiliary_loss_mlp": 0.01023842,
      "balance_loss_clip": 1.04428351,
      "balance_loss_mlp": 1.01716948,
      "epoch": 0.9684362412072386,
      "flos": 24207958016640.0,
      "grad_norm": 2.0197127879419994,
      "language_loss": 0.73988187,
      "learning_rate": 1.0404207191540227e-08,
      "loss": 0.76133144,
      "num_input_tokens_seen": 173900230,
      "step": 8054,
      "time_per_iteration": 2.5984811782836914
    },
    {
      "auxiliary_loss_clip": 0.0116225,
      "auxiliary_loss_mlp": 0.01025857,
      "balance_loss_clip": 1.04755437,
      "balance_loss_mlp": 1.01926231,
      "epoch": 0.9685564840978778,
      "flos": 22346241125760.0,
      "grad_norm": 1.7973320348293138,
      "language_loss": 0.74580777,
      "learning_rate": 1.0325005606153236e-08,
      "loss": 0.76768887,
      "num_input_tokens_seen": 173919690,
      "step": 8055,
      "time_per_iteration": 2.5618155002593994
    },
    {
      "auxiliary_loss_clip": 0.01105739,
      "auxiliary_loss_mlp": 0.01026144,
      "balance_loss_clip": 1.04054284,
      "balance_loss_mlp": 1.0193069,
      "epoch": 0.9686767269885168,
      "flos": 14386389477120.0,
      "grad_norm": 2.461169327159431,
      "language_loss": 0.78769219,
      "learning_rate": 1.0246105852881104e-08,
      "loss": 0.80901098,
      "num_input_tokens_seen": 173934790,
      "step": 8056,
      "time_per_iteration": 2.6134443283081055
    },
    {
      "auxiliary_loss_clip": 0.01164731,
      "auxiliary_loss_mlp": 0.01024256,
      "balance_loss_clip": 1.04690683,
      "balance_loss_mlp": 1.01717782,
      "epoch": 0.9687969698791559,
      "flos": 21287630471040.0,
      "grad_norm": 1.7879268187574222,
      "language_loss": 0.78713799,
      "learning_rate": 1.0167507943692476e-08,
      "loss": 0.80902779,
      "num_input_tokens_seen": 173953875,
      "step": 8057,
      "time_per_iteration": 2.5377798080444336
    },
    {
      "auxiliary_loss_clip": 0.01148445,
      "auxiliary_loss_mlp": 0.01028977,
      "balance_loss_clip": 1.04938304,
      "balance_loss_mlp": 1.02199447,
      "epoch": 0.968917212769795,
      "flos": 19828328624640.0,
      "grad_norm": 2.1446800513721884,
      "language_loss": 0.71085203,
      "learning_rate": 1.008921189051093e-08,
      "loss": 0.73262626,
      "num_input_tokens_seen": 173971220,
      "step": 8058,
      "time_per_iteration": 2.527008295059204
    },
    {
      "auxiliary_loss_clip": 0.01165992,
      "auxiliary_loss_mlp": 0.01026387,
      "balance_loss_clip": 1.04906547,
      "balance_loss_mlp": 1.01870668,
      "epoch": 0.9690374556604341,
      "flos": 21681749473920.0,
      "grad_norm": 1.919580542145979,
      "language_loss": 0.7686522,
      "learning_rate": 1.0011217705213848e-08,
      "loss": 0.79057604,
      "num_input_tokens_seen": 173989095,
      "step": 8059,
      "time_per_iteration": 2.5254733562469482
    },
    {
      "auxiliary_loss_clip": 0.01146268,
      "auxiliary_loss_mlp": 0.01022523,
      "balance_loss_clip": 1.04599035,
      "balance_loss_mlp": 1.01645541,
      "epoch": 0.9691576985510731,
      "flos": 32635437851520.0,
      "grad_norm": 1.7550523547618049,
      "language_loss": 0.74199909,
      "learning_rate": 9.933525399632658e-09,
      "loss": 0.76368701,
      "num_input_tokens_seen": 174007330,
      "step": 8060,
      "time_per_iteration": 2.610304832458496
    },
    {
      "auxiliary_loss_clip": 0.01130959,
      "auxiliary_loss_mlp": 0.01026309,
      "balance_loss_clip": 1.04256654,
      "balance_loss_mlp": 1.01895034,
      "epoch": 0.9692779414417123,
      "flos": 35663174040960.0,
      "grad_norm": 1.7486629397755933,
      "language_loss": 0.64891952,
      "learning_rate": 9.856134985553488e-09,
      "loss": 0.67049217,
      "num_input_tokens_seen": 174027055,
      "step": 8061,
      "time_per_iteration": 2.6411588191986084
    },
    {
      "auxiliary_loss_clip": 0.01163483,
      "auxiliary_loss_mlp": 0.01026599,
      "balance_loss_clip": 1.04761577,
      "balance_loss_mlp": 1.01916599,
      "epoch": 0.9693981843323514,
      "flos": 28366952117760.0,
      "grad_norm": 1.538402884096897,
      "language_loss": 0.73476315,
      "learning_rate": 9.77904647471628e-09,
      "loss": 0.75666398,
      "num_input_tokens_seen": 174050235,
      "step": 8062,
      "time_per_iteration": 2.5111989974975586
    },
    {
      "auxiliary_loss_clip": 0.01097147,
      "auxiliary_loss_mlp": 0.01026132,
      "balance_loss_clip": 1.04008579,
      "balance_loss_mlp": 1.0190661,
      "epoch": 0.9695184272229904,
      "flos": 23622865378560.0,
      "grad_norm": 1.4112846098866993,
      "language_loss": 0.73961675,
      "learning_rate": 9.702259878815454e-09,
      "loss": 0.76084948,
      "num_input_tokens_seen": 174070560,
      "step": 8063,
      "time_per_iteration": 2.590095043182373
    },
    {
      "auxiliary_loss_clip": 0.01152205,
      "auxiliary_loss_mlp": 0.01039069,
      "balance_loss_clip": 1.04749322,
      "balance_loss_mlp": 1.03074849,
      "epoch": 0.9696386701136296,
      "flos": 23294677789440.0,
      "grad_norm": 2.0668136196081446,
      "language_loss": 0.74316698,
      "learning_rate": 9.625775209499254e-09,
      "loss": 0.76507974,
      "num_input_tokens_seen": 174090565,
      "step": 8064,
      "time_per_iteration": 3.216752290725708
    },
    {
      "auxiliary_loss_clip": 0.01115396,
      "auxiliary_loss_mlp": 0.010284,
      "balance_loss_clip": 1.04031372,
      "balance_loss_mlp": 1.02107418,
      "epoch": 0.9697589130042686,
      "flos": 15121876360320.0,
      "grad_norm": 1.9153722257479209,
      "language_loss": 0.74084312,
      "learning_rate": 9.549592478370172e-09,
      "loss": 0.76228106,
      "num_input_tokens_seen": 174108745,
      "step": 8065,
      "time_per_iteration": 3.3043999671936035
    },
    {
      "auxiliary_loss_clip": 0.01151803,
      "auxiliary_loss_mlp": 0.010282,
      "balance_loss_clip": 1.0459131,
      "balance_loss_mlp": 1.02150369,
      "epoch": 0.9698791558949077,
      "flos": 18879532824960.0,
      "grad_norm": 1.7098136012528384,
      "language_loss": 0.79346538,
      "learning_rate": 9.473711696985632e-09,
      "loss": 0.81526542,
      "num_input_tokens_seen": 174128075,
      "step": 8066,
      "time_per_iteration": 2.486731767654419
    },
    {
      "auxiliary_loss_clip": 0.01129948,
      "auxiliary_loss_mlp": 0.01021145,
      "balance_loss_clip": 1.04362917,
      "balance_loss_mlp": 1.01378345,
      "epoch": 0.9699993987855468,
      "flos": 17931455297280.0,
      "grad_norm": 2.0947633348214207,
      "language_loss": 0.75658727,
      "learning_rate": 9.398132876856201e-09,
      "loss": 0.77809817,
      "num_input_tokens_seen": 174147040,
      "step": 8067,
      "time_per_iteration": 3.1684863567352295
    },
    {
      "auxiliary_loss_clip": 0.01015668,
      "auxiliary_loss_mlp": 0.01002015,
      "balance_loss_clip": 1.00845218,
      "balance_loss_mlp": 1.00082898,
      "epoch": 0.9701196416761859,
      "flos": 67182186297600.0,
      "grad_norm": 0.7969104749980803,
      "language_loss": 0.60806471,
      "learning_rate": 9.322856029447379e-09,
      "loss": 0.62824154,
      "num_input_tokens_seen": 174208225,
      "step": 8068,
      "time_per_iteration": 3.0715153217315674
    },
    {
      "auxiliary_loss_clip": 0.01159466,
      "auxiliary_loss_mlp": 0.01024808,
      "balance_loss_clip": 1.04563808,
      "balance_loss_mlp": 1.01827264,
      "epoch": 0.970239884566825,
      "flos": 24277804012800.0,
      "grad_norm": 1.9462868313382278,
      "language_loss": 0.80149496,
      "learning_rate": 9.247881166178695e-09,
      "loss": 0.82333767,
      "num_input_tokens_seen": 174226935,
      "step": 8069,
      "time_per_iteration": 2.4667961597442627
    },
    {
      "auxiliary_loss_clip": 0.01129881,
      "auxiliary_loss_mlp": 0.01027024,
      "balance_loss_clip": 1.0436312,
      "balance_loss_mlp": 1.02007711,
      "epoch": 0.970360127457464,
      "flos": 25301689194240.0,
      "grad_norm": 2.2430495345324006,
      "language_loss": 0.76865351,
      "learning_rate": 9.173208298423274e-09,
      "loss": 0.79022253,
      "num_input_tokens_seen": 174248140,
      "step": 8070,
      "time_per_iteration": 2.6132006645202637
    },
    {
      "auxiliary_loss_clip": 0.01099928,
      "auxiliary_loss_mlp": 0.00760612,
      "balance_loss_clip": 1.03994918,
      "balance_loss_mlp": 1.00030088,
      "epoch": 0.9704803703481032,
      "flos": 29572473398400.0,
      "grad_norm": 1.5262850194703033,
      "language_loss": 0.76314145,
      "learning_rate": 9.09883743750961e-09,
      "loss": 0.78174686,
      "num_input_tokens_seen": 174271030,
      "step": 8071,
      "time_per_iteration": 2.670663356781006
    },
    {
      "auxiliary_loss_clip": 0.01132525,
      "auxiliary_loss_mlp": 0.01021422,
      "balance_loss_clip": 1.04459739,
      "balance_loss_mlp": 1.01496065,
      "epoch": 0.9706006132387422,
      "flos": 17380046638080.0,
      "grad_norm": 1.5953721746917666,
      "language_loss": 0.84062743,
      "learning_rate": 9.024768594719124e-09,
      "loss": 0.86216688,
      "num_input_tokens_seen": 174289410,
      "step": 8072,
      "time_per_iteration": 2.5339019298553467
    },
    {
      "auxiliary_loss_clip": 0.01118864,
      "auxiliary_loss_mlp": 0.0102202,
      "balance_loss_clip": 1.04227567,
      "balance_loss_mlp": 1.01494491,
      "epoch": 0.9707208561293813,
      "flos": 18186421011840.0,
      "grad_norm": 2.0774861144161267,
      "language_loss": 0.72566676,
      "learning_rate": 8.95100178128816e-09,
      "loss": 0.74707556,
      "num_input_tokens_seen": 174308550,
      "step": 8073,
      "time_per_iteration": 2.5342602729797363
    },
    {
      "auxiliary_loss_clip": 0.01134596,
      "auxiliary_loss_mlp": 0.01026896,
      "balance_loss_clip": 1.04360175,
      "balance_loss_mlp": 1.01952589,
      "epoch": 0.9708410990200205,
      "flos": 31248388212480.0,
      "grad_norm": 1.8243659228545839,
      "language_loss": 0.70059413,
      "learning_rate": 8.877537008407321e-09,
      "loss": 0.72220904,
      "num_input_tokens_seen": 174328600,
      "step": 8074,
      "time_per_iteration": 2.6209964752197266
    },
    {
      "auxiliary_loss_clip": 0.01139122,
      "auxiliary_loss_mlp": 0.01024687,
      "balance_loss_clip": 1.04533923,
      "balance_loss_mlp": 1.01743925,
      "epoch": 0.9709613419106595,
      "flos": 30554450386560.0,
      "grad_norm": 1.6304366229978284,
      "language_loss": 0.68915123,
      "learning_rate": 8.804374287221028e-09,
      "loss": 0.71078932,
      "num_input_tokens_seen": 174349835,
      "step": 8075,
      "time_per_iteration": 2.594649076461792
    },
    {
      "auxiliary_loss_clip": 0.01110122,
      "auxiliary_loss_mlp": 0.01025043,
      "balance_loss_clip": 1.03663349,
      "balance_loss_mlp": 1.01835251,
      "epoch": 0.9710815848012986,
      "flos": 23730166281600.0,
      "grad_norm": 1.5925995348221036,
      "language_loss": 0.8452059,
      "learning_rate": 8.731513628827958e-09,
      "loss": 0.86655748,
      "num_input_tokens_seen": 174369200,
      "step": 8076,
      "time_per_iteration": 2.6688039302825928
    },
    {
      "auxiliary_loss_clip": 0.01150363,
      "auxiliary_loss_mlp": 0.01023915,
      "balance_loss_clip": 1.04595697,
      "balance_loss_mlp": 1.0175494,
      "epoch": 0.9712018276919377,
      "flos": 23761875012480.0,
      "grad_norm": 1.7963410357847347,
      "language_loss": 0.8237555,
      "learning_rate": 8.658955044280825e-09,
      "loss": 0.84549826,
      "num_input_tokens_seen": 174388125,
      "step": 8077,
      "time_per_iteration": 2.5142860412597656
    },
    {
      "auxiliary_loss_clip": 0.01143944,
      "auxiliary_loss_mlp": 0.0102763,
      "balance_loss_clip": 1.04519153,
      "balance_loss_mlp": 1.02082014,
      "epoch": 0.9713220705825768,
      "flos": 23330983461120.0,
      "grad_norm": 1.9055696764295664,
      "language_loss": 0.77714193,
      "learning_rate": 8.586698544587268e-09,
      "loss": 0.79885769,
      "num_input_tokens_seen": 174409735,
      "step": 8078,
      "time_per_iteration": 3.207247257232666
    },
    {
      "auxiliary_loss_clip": 0.01127779,
      "auxiliary_loss_mlp": 0.01030417,
      "balance_loss_clip": 1.04272985,
      "balance_loss_mlp": 1.02348483,
      "epoch": 0.9714423134732159,
      "flos": 22200946611840.0,
      "grad_norm": 2.2947661143438864,
      "language_loss": 0.73990333,
      "learning_rate": 8.514744140707853e-09,
      "loss": 0.76148522,
      "num_input_tokens_seen": 174428875,
      "step": 8079,
      "time_per_iteration": 2.5396230220794678
    },
    {
      "auxiliary_loss_clip": 0.01163397,
      "auxiliary_loss_mlp": 0.01024535,
      "balance_loss_clip": 1.04877496,
      "balance_loss_mlp": 1.01799941,
      "epoch": 0.971562556363855,
      "flos": 20229917656320.0,
      "grad_norm": 1.5626393340455627,
      "language_loss": 0.76435322,
      "learning_rate": 8.443091843558515e-09,
      "loss": 0.78623247,
      "num_input_tokens_seen": 174447960,
      "step": 8080,
      "time_per_iteration": 2.453338384628296
    },
    {
      "auxiliary_loss_clip": 0.01130958,
      "auxiliary_loss_mlp": 0.01025416,
      "balance_loss_clip": 1.04528356,
      "balance_loss_mlp": 1.017977,
      "epoch": 0.9716827992544941,
      "flos": 24970197553920.0,
      "grad_norm": 2.1393933601517148,
      "language_loss": 0.6463204,
      "learning_rate": 8.37174166400878e-09,
      "loss": 0.66788417,
      "num_input_tokens_seen": 174463535,
      "step": 8081,
      "time_per_iteration": 2.5425164699554443
    },
    {
      "auxiliary_loss_clip": 0.01165516,
      "auxiliary_loss_mlp": 0.01031544,
      "balance_loss_clip": 1.04973578,
      "balance_loss_mlp": 1.02468634,
      "epoch": 0.9718030421451331,
      "flos": 24681476033280.0,
      "grad_norm": 2.211691472579568,
      "language_loss": 0.85322189,
      "learning_rate": 8.300693612881992e-09,
      "loss": 0.87519252,
      "num_input_tokens_seen": 174483600,
      "step": 8082,
      "time_per_iteration": 2.4732353687286377
    },
    {
      "auxiliary_loss_clip": 0.01149634,
      "auxiliary_loss_mlp": 0.00760804,
      "balance_loss_clip": 1.04652202,
      "balance_loss_mlp": 1.00037432,
      "epoch": 0.9719232850357723,
      "flos": 22090700793600.0,
      "grad_norm": 1.8367841699028,
      "language_loss": 0.81086242,
      "learning_rate": 8.22994770095664e-09,
      "loss": 0.82996678,
      "num_input_tokens_seen": 174502175,
      "step": 8083,
      "time_per_iteration": 2.492948293685913
    },
    {
      "auxiliary_loss_clip": 0.01133968,
      "auxiliary_loss_mlp": 0.0103103,
      "balance_loss_clip": 1.0483017,
      "balance_loss_mlp": 1.02385688,
      "epoch": 0.9720435279264114,
      "flos": 23656908493440.0,
      "grad_norm": 2.0567772561964044,
      "language_loss": 0.75549889,
      "learning_rate": 8.159503938964585e-09,
      "loss": 0.77714896,
      "num_input_tokens_seen": 174519495,
      "step": 8084,
      "time_per_iteration": 2.514460802078247
    },
    {
      "auxiliary_loss_clip": 0.01111672,
      "auxiliary_loss_mlp": 0.01028721,
      "balance_loss_clip": 1.04154742,
      "balance_loss_mlp": 1.02246261,
      "epoch": 0.9721637708170504,
      "flos": 28365910623360.0,
      "grad_norm": 2.0234816602474672,
      "language_loss": 0.7009306,
      "learning_rate": 8.089362337592164e-09,
      "loss": 0.7223345,
      "num_input_tokens_seen": 174543120,
      "step": 8085,
      "time_per_iteration": 2.6185779571533203
    },
    {
      "auxiliary_loss_clip": 0.01134102,
      "auxiliary_loss_mlp": 0.01026225,
      "balance_loss_clip": 1.0464915,
      "balance_loss_mlp": 1.01963305,
      "epoch": 0.9722840137076896,
      "flos": 29130807767040.0,
      "grad_norm": 1.5559959883015504,
      "language_loss": 0.72174954,
      "learning_rate": 8.019522907479536e-09,
      "loss": 0.74335277,
      "num_input_tokens_seen": 174563480,
      "step": 8086,
      "time_per_iteration": 2.610379457473755
    },
    {
      "auxiliary_loss_clip": 0.01151386,
      "auxiliary_loss_mlp": 0.01027153,
      "balance_loss_clip": 1.04642642,
      "balance_loss_mlp": 1.02008998,
      "epoch": 0.9724042565983286,
      "flos": 19243954258560.0,
      "grad_norm": 2.1070561135427113,
      "language_loss": 0.77498078,
      "learning_rate": 7.949985659221558e-09,
      "loss": 0.79676622,
      "num_input_tokens_seen": 174580745,
      "step": 8087,
      "time_per_iteration": 2.505854368209839
    },
    {
      "auxiliary_loss_clip": 0.01138107,
      "auxiliary_loss_mlp": 0.01030383,
      "balance_loss_clip": 1.04434991,
      "balance_loss_mlp": 1.02367687,
      "epoch": 0.9725244994889677,
      "flos": 23039676161280.0,
      "grad_norm": 1.9035924405317586,
      "language_loss": 0.78760314,
      "learning_rate": 7.880750603366904e-09,
      "loss": 0.80928802,
      "num_input_tokens_seen": 174599615,
      "step": 8088,
      "time_per_iteration": 2.5834314823150635
    },
    {
      "auxiliary_loss_clip": 0.0113118,
      "auxiliary_loss_mlp": 0.0102961,
      "balance_loss_clip": 1.04316616,
      "balance_loss_mlp": 1.02222204,
      "epoch": 0.9726447423796069,
      "flos": 23367468700800.0,
      "grad_norm": 1.7882575666380598,
      "language_loss": 0.79615492,
      "learning_rate": 7.811817750418282e-09,
      "loss": 0.81776285,
      "num_input_tokens_seen": 174618375,
      "step": 8089,
      "time_per_iteration": 2.5912866592407227
    },
    {
      "auxiliary_loss_clip": 0.01117942,
      "auxiliary_loss_mlp": 0.01027688,
      "balance_loss_clip": 1.04500079,
      "balance_loss_mlp": 1.02029145,
      "epoch": 0.9727649852702459,
      "flos": 26541648639360.0,
      "grad_norm": 1.5263123485955303,
      "language_loss": 0.80120122,
      "learning_rate": 7.743187110833105e-09,
      "loss": 0.82265753,
      "num_input_tokens_seen": 174641135,
      "step": 8090,
      "time_per_iteration": 3.4511404037475586
    },
    {
      "auxiliary_loss_clip": 0.01136567,
      "auxiliary_loss_mlp": 0.0102261,
      "balance_loss_clip": 1.04227948,
      "balance_loss_mlp": 1.01619065,
      "epoch": 0.972885228160885,
      "flos": 20522338277760.0,
      "grad_norm": 1.4116855766068876,
      "language_loss": 0.80627573,
      "learning_rate": 7.674858695022602e-09,
      "loss": 0.82786751,
      "num_input_tokens_seen": 174659490,
      "step": 8091,
      "time_per_iteration": 3.327000856399536
    },
    {
      "auxiliary_loss_clip": 0.01166252,
      "auxiliary_loss_mlp": 0.01022249,
      "balance_loss_clip": 1.04886746,
      "balance_loss_mlp": 1.01463127,
      "epoch": 0.9730054710515241,
      "flos": 17566064196480.0,
      "grad_norm": 2.3331948240445417,
      "language_loss": 0.76291311,
      "learning_rate": 7.606832513351591e-09,
      "loss": 0.78479815,
      "num_input_tokens_seen": 174677440,
      "step": 8092,
      "time_per_iteration": 2.4912893772125244
    },
    {
      "auxiliary_loss_clip": 0.01054518,
      "auxiliary_loss_mlp": 0.00751194,
      "balance_loss_clip": 1.00776875,
      "balance_loss_mlp": 1.00017309,
      "epoch": 0.9731257139421632,
      "flos": 68972010117120.0,
      "grad_norm": 0.8192911684423997,
      "language_loss": 0.63923639,
      "learning_rate": 7.539108576140264e-09,
      "loss": 0.65729356,
      "num_input_tokens_seen": 174741550,
      "step": 8093,
      "time_per_iteration": 3.8809845447540283
    },
    {
      "auxiliary_loss_clip": 0.01105716,
      "auxiliary_loss_mlp": 0.01026162,
      "balance_loss_clip": 1.04143751,
      "balance_loss_mlp": 1.01942074,
      "epoch": 0.9732459568328022,
      "flos": 18478841633280.0,
      "grad_norm": 2.5597798674010104,
      "language_loss": 0.70629871,
      "learning_rate": 7.471686893661732e-09,
      "loss": 0.72761744,
      "num_input_tokens_seen": 174759845,
      "step": 8094,
      "time_per_iteration": 2.575995445251465
    },
    {
      "auxiliary_loss_clip": 0.01134082,
      "auxiliary_loss_mlp": 0.01025947,
      "balance_loss_clip": 1.0468061,
      "balance_loss_mlp": 1.01899385,
      "epoch": 0.9733661997234414,
      "flos": 20883886623360.0,
      "grad_norm": 1.8070783566164774,
      "language_loss": 0.64235735,
      "learning_rate": 7.4045674761442636e-09,
      "loss": 0.66395772,
      "num_input_tokens_seen": 174777175,
      "step": 8095,
      "time_per_iteration": 2.5133304595947266
    },
    {
      "auxiliary_loss_clip": 0.01161325,
      "auxiliary_loss_mlp": 0.00760721,
      "balance_loss_clip": 1.04728925,
      "balance_loss_mlp": 1.00031745,
      "epoch": 0.9734864426140805,
      "flos": 23766795175680.0,
      "grad_norm": 1.6667388396975555,
      "language_loss": 0.74245483,
      "learning_rate": 7.337750333769488e-09,
      "loss": 0.76167524,
      "num_input_tokens_seen": 174796980,
      "step": 8096,
      "time_per_iteration": 2.4868946075439453
    },
    {
      "auxiliary_loss_clip": 0.01139406,
      "auxiliary_loss_mlp": 0.0102772,
      "balance_loss_clip": 1.04232049,
      "balance_loss_mlp": 1.0198853,
      "epoch": 0.9736066855047195,
      "flos": 35042422176000.0,
      "grad_norm": 1.742743131935378,
      "language_loss": 0.72891945,
      "learning_rate": 7.2712354766737425e-09,
      "loss": 0.75059074,
      "num_input_tokens_seen": 174817310,
      "step": 8097,
      "time_per_iteration": 2.6347479820251465
    },
    {
      "auxiliary_loss_clip": 0.01112232,
      "auxiliary_loss_mlp": 0.01026971,
      "balance_loss_clip": 1.04446495,
      "balance_loss_mlp": 1.01991701,
      "epoch": 0.9737269283953586,
      "flos": 20410620001920.0,
      "grad_norm": 1.5450109162765466,
      "language_loss": 0.80678189,
      "learning_rate": 7.2050229149469565e-09,
      "loss": 0.82817394,
      "num_input_tokens_seen": 174837320,
      "step": 8098,
      "time_per_iteration": 2.5532073974609375
    },
    {
      "auxiliary_loss_clip": 0.01123507,
      "auxiliary_loss_mlp": 0.01022838,
      "balance_loss_clip": 1.04042423,
      "balance_loss_mlp": 1.01614165,
      "epoch": 0.9738471712859977,
      "flos": 28911680847360.0,
      "grad_norm": 1.7137882644764117,
      "language_loss": 0.63454342,
      "learning_rate": 7.139112658633984e-09,
      "loss": 0.65600687,
      "num_input_tokens_seen": 174857470,
      "step": 8099,
      "time_per_iteration": 2.6178789138793945
    },
    {
      "auxiliary_loss_clip": 0.01121181,
      "auxiliary_loss_mlp": 0.0102124,
      "balance_loss_clip": 1.04534793,
      "balance_loss_mlp": 1.01407301,
      "epoch": 0.9739674141766368,
      "flos": 27782326356480.0,
      "grad_norm": 2.0393717650867744,
      "language_loss": 0.70405692,
      "learning_rate": 7.073504717733048e-09,
      "loss": 0.72548115,
      "num_input_tokens_seen": 174877035,
      "step": 8100,
      "time_per_iteration": 2.6085379123687744
    },
    {
      "auxiliary_loss_clip": 0.01008449,
      "auxiliary_loss_mlp": 0.01002919,
      "balance_loss_clip": 1.00922012,
      "balance_loss_mlp": 1.00194192,
      "epoch": 0.9740876570672758,
      "flos": 68863057188480.0,
      "grad_norm": 0.7318306031136627,
      "language_loss": 0.57223362,
      "learning_rate": 7.008199102196855e-09,
      "loss": 0.59234726,
      "num_input_tokens_seen": 174938460,
      "step": 8101,
      "time_per_iteration": 3.120814085006714
    },
    {
      "auxiliary_loss_clip": 0.01031341,
      "auxiliary_loss_mlp": 0.01001373,
      "balance_loss_clip": 1.01064944,
      "balance_loss_mlp": 1.00043094,
      "epoch": 0.974207899957915,
      "flos": 58236622646400.0,
      "grad_norm": 0.7937686289077596,
      "language_loss": 0.58983928,
      "learning_rate": 6.9431958219321464e-09,
      "loss": 0.61016637,
      "num_input_tokens_seen": 174994625,
      "step": 8102,
      "time_per_iteration": 3.091003179550171
    },
    {
      "auxiliary_loss_clip": 0.01134289,
      "auxiliary_loss_mlp": 0.01024649,
      "balance_loss_clip": 1.04358077,
      "balance_loss_mlp": 1.01740408,
      "epoch": 0.9743281428485541,
      "flos": 22600057605120.0,
      "grad_norm": 1.4682443897966269,
      "language_loss": 0.77720892,
      "learning_rate": 6.878494886800146e-09,
      "loss": 0.79879832,
      "num_input_tokens_seen": 175015400,
      "step": 8103,
      "time_per_iteration": 2.533651828765869
    },
    {
      "auxiliary_loss_clip": 0.01134014,
      "auxiliary_loss_mlp": 0.01025236,
      "balance_loss_clip": 1.04719496,
      "balance_loss_mlp": 1.01838171,
      "epoch": 0.9744483857391931,
      "flos": 20008815488640.0,
      "grad_norm": 1.8921676450977447,
      "language_loss": 0.76410925,
      "learning_rate": 6.814096306615669e-09,
      "loss": 0.78570175,
      "num_input_tokens_seen": 175033540,
      "step": 8104,
      "time_per_iteration": 3.308861017227173
    },
    {
      "auxiliary_loss_clip": 0.01139884,
      "auxiliary_loss_mlp": 0.01026371,
      "balance_loss_clip": 1.04345751,
      "balance_loss_mlp": 1.01904917,
      "epoch": 0.9745686286298323,
      "flos": 17675268520320.0,
      "grad_norm": 2.349906533584389,
      "language_loss": 0.64958918,
      "learning_rate": 6.750000091148011e-09,
      "loss": 0.67125171,
      "num_input_tokens_seen": 175050835,
      "step": 8105,
      "time_per_iteration": 2.494753837585449
    },
    {
      "auxiliary_loss_clip": 0.01164801,
      "auxiliary_loss_mlp": 0.01026875,
      "balance_loss_clip": 1.04883742,
      "balance_loss_mlp": 1.02001119,
      "epoch": 0.9746888715204713,
      "flos": 29460252332160.0,
      "grad_norm": 1.84913374105514,
      "language_loss": 0.72452414,
      "learning_rate": 6.686206250120729e-09,
      "loss": 0.74644095,
      "num_input_tokens_seen": 175072330,
      "step": 8106,
      "time_per_iteration": 2.5196568965911865
    },
    {
      "auxiliary_loss_clip": 0.01125218,
      "auxiliary_loss_mlp": 0.01027048,
      "balance_loss_clip": 1.04104519,
      "balance_loss_mlp": 1.01998186,
      "epoch": 0.9748091144111104,
      "flos": 18479308510080.0,
      "grad_norm": 1.851565209008299,
      "language_loss": 0.74677575,
      "learning_rate": 6.622714793210749e-09,
      "loss": 0.76829839,
      "num_input_tokens_seen": 175091250,
      "step": 8107,
      "time_per_iteration": 2.54463529586792
    },
    {
      "auxiliary_loss_clip": 0.01163905,
      "auxiliary_loss_mlp": 0.01022901,
      "balance_loss_clip": 1.04737616,
      "balance_loss_mlp": 1.01628184,
      "epoch": 0.9749293573017496,
      "flos": 20665154753280.0,
      "grad_norm": 1.7923424113228867,
      "language_loss": 0.78644866,
      "learning_rate": 6.559525730050364e-09,
      "loss": 0.80831671,
      "num_input_tokens_seen": 175111350,
      "step": 8108,
      "time_per_iteration": 2.4503021240234375
    },
    {
      "auxiliary_loss_clip": 0.0112522,
      "auxiliary_loss_mlp": 0.01027256,
      "balance_loss_clip": 1.04553723,
      "balance_loss_mlp": 1.02072012,
      "epoch": 0.9750496001923886,
      "flos": 18478590238080.0,
      "grad_norm": 3.4666589370651226,
      "language_loss": 0.75882512,
      "learning_rate": 6.496639070224574e-09,
      "loss": 0.78034985,
      "num_input_tokens_seen": 175129835,
      "step": 8109,
      "time_per_iteration": 2.5428624153137207
    },
    {
      "auxiliary_loss_clip": 0.01153931,
      "auxiliary_loss_mlp": 0.01026795,
      "balance_loss_clip": 1.0479846,
      "balance_loss_mlp": 1.02005374,
      "epoch": 0.9751698430830277,
      "flos": 19572967860480.0,
      "grad_norm": 2.1001098801539695,
      "language_loss": 0.83430654,
      "learning_rate": 6.4340548232739714e-09,
      "loss": 0.85611379,
      "num_input_tokens_seen": 175146035,
      "step": 8110,
      "time_per_iteration": 2.458155632019043
    },
    {
      "auxiliary_loss_clip": 0.01126026,
      "auxiliary_loss_mlp": 0.01025705,
      "balance_loss_clip": 1.04226553,
      "balance_loss_mlp": 1.01921654,
      "epoch": 0.9752900859736668,
      "flos": 23550325862400.0,
      "grad_norm": 2.0629284809186066,
      "language_loss": 0.79172021,
      "learning_rate": 6.371772998692071e-09,
      "loss": 0.81323755,
      "num_input_tokens_seen": 175165290,
      "step": 8111,
      "time_per_iteration": 2.6033778190612793
    },
    {
      "auxiliary_loss_clip": 0.01124401,
      "auxiliary_loss_mlp": 0.0102811,
      "balance_loss_clip": 1.04242039,
      "balance_loss_mlp": 1.02113879,
      "epoch": 0.9754103288643059,
      "flos": 20303211358080.0,
      "grad_norm": 3.313251058324904,
      "language_loss": 0.64667892,
      "learning_rate": 6.309793605927094e-09,
      "loss": 0.66820401,
      "num_input_tokens_seen": 175183610,
      "step": 8112,
      "time_per_iteration": 2.6252126693725586
    },
    {
      "auxiliary_loss_clip": 0.01137657,
      "auxiliary_loss_mlp": 0.01019222,
      "balance_loss_clip": 1.04499686,
      "balance_loss_mlp": 1.01214671,
      "epoch": 0.975530571754945,
      "flos": 19350680544000.0,
      "grad_norm": 1.7670501539824324,
      "language_loss": 0.80257463,
      "learning_rate": 6.248116654381297e-09,
      "loss": 0.82414341,
      "num_input_tokens_seen": 175202080,
      "step": 8113,
      "time_per_iteration": 2.517439842224121
    },
    {
      "auxiliary_loss_clip": 0.01135949,
      "auxiliary_loss_mlp": 0.01019507,
      "balance_loss_clip": 1.04132771,
      "balance_loss_mlp": 1.01336145,
      "epoch": 0.9756508146455841,
      "flos": 23583399310080.0,
      "grad_norm": 1.619109069468923,
      "language_loss": 0.72603869,
      "learning_rate": 6.186742153410751e-09,
      "loss": 0.74759322,
      "num_input_tokens_seen": 175221575,
      "step": 8114,
      "time_per_iteration": 2.550299644470215
    },
    {
      "auxiliary_loss_clip": 0.01135344,
      "auxiliary_loss_mlp": 0.01032051,
      "balance_loss_clip": 1.04614377,
      "balance_loss_mlp": 1.0249964,
      "epoch": 0.9757710575362232,
      "flos": 22966921163520.0,
      "grad_norm": 2.0678730766693656,
      "language_loss": 0.87404859,
      "learning_rate": 6.125670112326453e-09,
      "loss": 0.89572257,
      "num_input_tokens_seen": 175240835,
      "step": 8115,
      "time_per_iteration": 2.547908306121826
    },
    {
      "auxiliary_loss_clip": 0.01147005,
      "auxiliary_loss_mlp": 0.01027857,
      "balance_loss_clip": 1.04207921,
      "balance_loss_mlp": 1.02074361,
      "epoch": 0.9758913004268622,
      "flos": 27966009530880.0,
      "grad_norm": 1.4563347907475137,
      "language_loss": 0.69758636,
      "learning_rate": 6.064900540392548e-09,
      "loss": 0.71933496,
      "num_input_tokens_seen": 175262930,
      "step": 8116,
      "time_per_iteration": 3.3527634143829346
    },
    {
      "auxiliary_loss_clip": 0.01130711,
      "auxiliary_loss_mlp": 0.01020712,
      "balance_loss_clip": 1.04597473,
      "balance_loss_mlp": 1.01478422,
      "epoch": 0.9760115433175014,
      "flos": 22200156512640.0,
      "grad_norm": 3.675106257629933,
      "language_loss": 0.78747511,
      "learning_rate": 6.0044334468278835e-09,
      "loss": 0.80898935,
      "num_input_tokens_seen": 175282275,
      "step": 8117,
      "time_per_iteration": 3.3055875301361084
    },
    {
      "auxiliary_loss_clip": 0.01105521,
      "auxiliary_loss_mlp": 0.01029235,
      "balance_loss_clip": 1.03974879,
      "balance_loss_mlp": 1.02204931,
      "epoch": 0.9761317862081405,
      "flos": 26250736389120.0,
      "grad_norm": 2.1004759457793636,
      "language_loss": 0.71790421,
      "learning_rate": 5.944268840805345e-09,
      "loss": 0.73925173,
      "num_input_tokens_seen": 175303020,
      "step": 8118,
      "time_per_iteration": 3.3869924545288086
    },
    {
      "auxiliary_loss_clip": 0.01115597,
      "auxiliary_loss_mlp": 0.01025611,
      "balance_loss_clip": 1.04155397,
      "balance_loss_mlp": 1.01903677,
      "epoch": 0.9762520290987795,
      "flos": 26575440359040.0,
      "grad_norm": 2.074114051420245,
      "language_loss": 0.64118737,
      "learning_rate": 5.88440673145163e-09,
      "loss": 0.66259944,
      "num_input_tokens_seen": 175324070,
      "step": 8119,
      "time_per_iteration": 2.5887725353240967
    },
    {
      "auxiliary_loss_clip": 0.01148902,
      "auxiliary_loss_mlp": 0.01029965,
      "balance_loss_clip": 1.04847777,
      "balance_loss_mlp": 1.02311385,
      "epoch": 0.9763722719894187,
      "flos": 18005036307840.0,
      "grad_norm": 2.4995641137639524,
      "language_loss": 0.8222388,
      "learning_rate": 5.824847127848142e-09,
      "loss": 0.84402746,
      "num_input_tokens_seen": 175342595,
      "step": 8120,
      "time_per_iteration": 2.4793779850006104
    },
    {
      "auxiliary_loss_clip": 0.01109721,
      "auxiliary_loss_mlp": 0.0102602,
      "balance_loss_clip": 1.04192913,
      "balance_loss_mlp": 1.01911211,
      "epoch": 0.9764925148800577,
      "flos": 22455660931200.0,
      "grad_norm": 1.8912771846638132,
      "language_loss": 0.79150844,
      "learning_rate": 5.765590039029433e-09,
      "loss": 0.81286585,
      "num_input_tokens_seen": 175361915,
      "step": 8121,
      "time_per_iteration": 2.6196846961975098
    },
    {
      "auxiliary_loss_clip": 0.01163517,
      "auxiliary_loss_mlp": 0.01027366,
      "balance_loss_clip": 1.04880714,
      "balance_loss_mlp": 1.02056551,
      "epoch": 0.9766127577706968,
      "flos": 36757084786560.0,
      "grad_norm": 1.830093689581446,
      "language_loss": 0.70972621,
      "learning_rate": 5.706635473985422e-09,
      "loss": 0.73163497,
      "num_input_tokens_seen": 175385785,
      "step": 8122,
      "time_per_iteration": 2.598020076751709
    },
    {
      "auxiliary_loss_clip": 0.01149677,
      "auxiliary_loss_mlp": 0.01027794,
      "balance_loss_clip": 1.04675031,
      "balance_loss_mlp": 1.02096367,
      "epoch": 0.976733000661336,
      "flos": 22309971367680.0,
      "grad_norm": 1.7708314579665103,
      "language_loss": 0.84732628,
      "learning_rate": 5.6479834416591764e-09,
      "loss": 0.86910093,
      "num_input_tokens_seen": 175405145,
      "step": 8123,
      "time_per_iteration": 2.5231235027313232
    },
    {
      "auxiliary_loss_clip": 0.01148366,
      "auxiliary_loss_mlp": 0.00761227,
      "balance_loss_clip": 1.04634452,
      "balance_loss_mlp": 1.00029039,
      "epoch": 0.976853243551975,
      "flos": 25810938264960.0,
      "grad_norm": 1.7649009470239023,
      "language_loss": 0.68388832,
      "learning_rate": 5.589633950947803e-09,
      "loss": 0.70298421,
      "num_input_tokens_seen": 175422645,
      "step": 8124,
      "time_per_iteration": 2.5520458221435547
    },
    {
      "auxiliary_loss_clip": 0.0113376,
      "auxiliary_loss_mlp": 0.01027976,
      "balance_loss_clip": 1.04494739,
      "balance_loss_mlp": 1.02035904,
      "epoch": 0.9769734864426141,
      "flos": 21397445326080.0,
      "grad_norm": 1.9622206196131053,
      "language_loss": 0.69693482,
      "learning_rate": 5.5315870107035535e-09,
      "loss": 0.71855217,
      "num_input_tokens_seen": 175440695,
      "step": 8125,
      "time_per_iteration": 2.5286264419555664
    },
    {
      "auxiliary_loss_clip": 0.01128625,
      "auxiliary_loss_mlp": 0.01024466,
      "balance_loss_clip": 1.04361653,
      "balance_loss_mlp": 1.01743853,
      "epoch": 0.9770937293332532,
      "flos": 13990977584640.0,
      "grad_norm": 1.8439173250890952,
      "language_loss": 0.78845292,
      "learning_rate": 5.473842629731607e-09,
      "loss": 0.80998385,
      "num_input_tokens_seen": 175459195,
      "step": 8126,
      "time_per_iteration": 2.52412486076355
    },
    {
      "auxiliary_loss_clip": 0.0114117,
      "auxiliary_loss_mlp": 0.00761316,
      "balance_loss_clip": 1.0429852,
      "balance_loss_mlp": 1.00033402,
      "epoch": 0.9772139722238923,
      "flos": 17931994001280.0,
      "grad_norm": 2.648472577256281,
      "language_loss": 0.778602,
      "learning_rate": 5.416400816792066e-09,
      "loss": 0.79762685,
      "num_input_tokens_seen": 175476710,
      "step": 8127,
      "time_per_iteration": 2.506303548812866
    },
    {
      "auxiliary_loss_clip": 0.01162325,
      "auxiliary_loss_mlp": 0.01026523,
      "balance_loss_clip": 1.04698205,
      "balance_loss_mlp": 1.01963866,
      "epoch": 0.9773342151145313,
      "flos": 20446171488000.0,
      "grad_norm": 2.3620208117501935,
      "language_loss": 0.78060627,
      "learning_rate": 5.359261580598407e-09,
      "loss": 0.80249476,
      "num_input_tokens_seen": 175492550,
      "step": 8128,
      "time_per_iteration": 2.464015007019043
    },
    {
      "auxiliary_loss_clip": 0.01149982,
      "auxiliary_loss_mlp": 0.01023565,
      "balance_loss_clip": 1.04575109,
      "balance_loss_mlp": 1.01599562,
      "epoch": 0.9774544580051704,
      "flos": 11837306949120.0,
      "grad_norm": 2.232291467392377,
      "language_loss": 0.78115386,
      "learning_rate": 5.302424929819027e-09,
      "loss": 0.80288935,
      "num_input_tokens_seen": 175506560,
      "step": 8129,
      "time_per_iteration": 2.4585084915161133
    },
    {
      "auxiliary_loss_clip": 0.01151011,
      "auxiliary_loss_mlp": 0.01023771,
      "balance_loss_clip": 1.0435369,
      "balance_loss_mlp": 1.01673174,
      "epoch": 0.9775747008958096,
      "flos": 13479932833920.0,
      "grad_norm": 2.4873863576204953,
      "language_loss": 0.73043025,
      "learning_rate": 5.24589087307592e-09,
      "loss": 0.75217807,
      "num_input_tokens_seen": 175524180,
      "step": 8130,
      "time_per_iteration": 2.4824674129486084
    },
    {
      "auxiliary_loss_clip": 0.0116469,
      "auxiliary_loss_mlp": 0.01025224,
      "balance_loss_clip": 1.047261,
      "balance_loss_mlp": 1.0185008,
      "epoch": 0.9776949437864486,
      "flos": 59532314042880.0,
      "grad_norm": 1.379042151878069,
      "language_loss": 0.64668667,
      "learning_rate": 5.189659418944891e-09,
      "loss": 0.66858578,
      "num_input_tokens_seen": 175554355,
      "step": 8131,
      "time_per_iteration": 3.559797525405884
    },
    {
      "auxiliary_loss_clip": 0.01163623,
      "auxiliary_loss_mlp": 0.010256,
      "balance_loss_clip": 1.04829586,
      "balance_loss_mlp": 1.01903141,
      "epoch": 0.9778151866770877,
      "flos": 21178605715200.0,
      "grad_norm": 3.337282486063652,
      "language_loss": 0.78229892,
      "learning_rate": 5.133730575956674e-09,
      "loss": 0.80419111,
      "num_input_tokens_seen": 175574025,
      "step": 8132,
      "time_per_iteration": 2.456928253173828
    },
    {
      "auxiliary_loss_clip": 0.01135117,
      "auxiliary_loss_mlp": 0.01033027,
      "balance_loss_clip": 1.04385424,
      "balance_loss_mlp": 1.0259161,
      "epoch": 0.9779354295677268,
      "flos": 20886795624960.0,
      "grad_norm": 1.8958627347827732,
      "language_loss": 0.71999693,
      "learning_rate": 5.0781043525953696e-09,
      "loss": 0.74167836,
      "num_input_tokens_seen": 175592090,
      "step": 8133,
      "time_per_iteration": 2.522958993911743
    },
    {
      "auxiliary_loss_clip": 0.01133132,
      "auxiliary_loss_mlp": 0.01026661,
      "balance_loss_clip": 1.04692996,
      "balance_loss_mlp": 1.01973498,
      "epoch": 0.9780556724583659,
      "flos": 23440618748160.0,
      "grad_norm": 1.5636947504241716,
      "language_loss": 0.73686719,
      "learning_rate": 5.0227807572995605e-09,
      "loss": 0.75846505,
      "num_input_tokens_seen": 175614065,
      "step": 8134,
      "time_per_iteration": 2.546506643295288
    },
    {
      "auxiliary_loss_clip": 0.01138432,
      "auxiliary_loss_mlp": 0.01024041,
      "balance_loss_clip": 1.04434395,
      "balance_loss_mlp": 1.0173744,
      "epoch": 0.9781759153490049,
      "flos": 20923244951040.0,
      "grad_norm": 2.0613216496413864,
      "language_loss": 0.67396057,
      "learning_rate": 4.967759798461646e-09,
      "loss": 0.69558525,
      "num_input_tokens_seen": 175632410,
      "step": 8135,
      "time_per_iteration": 2.5524251461029053
    },
    {
      "auxiliary_loss_clip": 0.01162052,
      "auxiliary_loss_mlp": 0.01026303,
      "balance_loss_clip": 1.04799819,
      "balance_loss_mlp": 1.01998174,
      "epoch": 0.9782961582396441,
      "flos": 28293191539200.0,
      "grad_norm": 1.9920821027096027,
      "language_loss": 0.74743414,
      "learning_rate": 4.913041484428282e-09,
      "loss": 0.76931763,
      "num_input_tokens_seen": 175652885,
      "step": 8136,
      "time_per_iteration": 2.5039172172546387
    },
    {
      "auxiliary_loss_clip": 0.01150666,
      "auxiliary_loss_mlp": 0.01026306,
      "balance_loss_clip": 1.04572475,
      "balance_loss_mlp": 1.01974607,
      "epoch": 0.9784164011302832,
      "flos": 25552955808000.0,
      "grad_norm": 1.7335580604052399,
      "language_loss": 0.74004799,
      "learning_rate": 4.858625823500384e-09,
      "loss": 0.76181769,
      "num_input_tokens_seen": 175670585,
      "step": 8137,
      "time_per_iteration": 2.530212879180908
    },
    {
      "auxiliary_loss_clip": 0.01152927,
      "auxiliary_loss_mlp": 0.01026264,
      "balance_loss_clip": 1.04605591,
      "balance_loss_mlp": 1.0191983,
      "epoch": 0.9785366440209222,
      "flos": 29965945956480.0,
      "grad_norm": 1.783457638464804,
      "language_loss": 0.73589694,
      "learning_rate": 4.80451282393246e-09,
      "loss": 0.75768888,
      "num_input_tokens_seen": 175690570,
      "step": 8138,
      "time_per_iteration": 2.565626621246338
    },
    {
      "auxiliary_loss_clip": 0.01132408,
      "auxiliary_loss_mlp": 0.01026721,
      "balance_loss_clip": 1.04477429,
      "balance_loss_mlp": 1.02008152,
      "epoch": 0.9786568869115614,
      "flos": 32343591847680.0,
      "grad_norm": 1.9356718164200746,
      "language_loss": 0.67370594,
      "learning_rate": 4.750702493933722e-09,
      "loss": 0.69529724,
      "num_input_tokens_seen": 175710455,
      "step": 8139,
      "time_per_iteration": 2.6023685932159424
    },
    {
      "auxiliary_loss_clip": 0.01131475,
      "auxiliary_loss_mlp": 0.00760663,
      "balance_loss_clip": 1.04379165,
      "balance_loss_mlp": 1.00031805,
      "epoch": 0.9787771298022004,
      "flos": 23331414424320.0,
      "grad_norm": 1.8531392822683552,
      "language_loss": 0.85275543,
      "learning_rate": 4.697194841666974e-09,
      "loss": 0.87167686,
      "num_input_tokens_seen": 175729380,
      "step": 8140,
      "time_per_iteration": 2.5328710079193115
    },
    {
      "auxiliary_loss_clip": 0.01148913,
      "auxiliary_loss_mlp": 0.01025211,
      "balance_loss_clip": 1.04408312,
      "balance_loss_mlp": 1.0179038,
      "epoch": 0.9788973726928395,
      "flos": 21468548298240.0,
      "grad_norm": 2.0229016328631744,
      "language_loss": 0.81767964,
      "learning_rate": 4.6439898752492764e-09,
      "loss": 0.83942091,
      "num_input_tokens_seen": 175749520,
      "step": 8141,
      "time_per_iteration": 2.4916458129882812
    },
    {
      "auxiliary_loss_clip": 0.01045897,
      "auxiliary_loss_mlp": 0.00751165,
      "balance_loss_clip": 1.00840497,
      "balance_loss_mlp": 1.00023317,
      "epoch": 0.9790176155834787,
      "flos": 68897459439360.0,
      "grad_norm": 0.7512680982481615,
      "language_loss": 0.63699257,
      "learning_rate": 4.591087602751731e-09,
      "loss": 0.65496325,
      "num_input_tokens_seen": 175811380,
      "step": 8142,
      "time_per_iteration": 3.9472527503967285
    },
    {
      "auxiliary_loss_clip": 0.01147399,
      "auxiliary_loss_mlp": 0.01022887,
      "balance_loss_clip": 1.04554772,
      "balance_loss_mlp": 1.01613081,
      "epoch": 0.9791378584741177,
      "flos": 21430877909760.0,
      "grad_norm": 1.611340909500763,
      "language_loss": 0.71945065,
      "learning_rate": 4.538488032199916e-09,
      "loss": 0.74115348,
      "num_input_tokens_seen": 175829480,
      "step": 8143,
      "time_per_iteration": 3.290969133377075
    },
    {
      "auxiliary_loss_clip": 0.01150175,
      "auxiliary_loss_mlp": 0.01022784,
      "balance_loss_clip": 1.04336429,
      "balance_loss_mlp": 1.01554501,
      "epoch": 0.9792581013647568,
      "flos": 20153032594560.0,
      "grad_norm": 1.9266721278007346,
      "language_loss": 0.68979269,
      "learning_rate": 4.486191171572784e-09,
      "loss": 0.71152228,
      "num_input_tokens_seen": 175846750,
      "step": 8144,
      "time_per_iteration": 3.239699125289917
    },
    {
      "auxiliary_loss_clip": 0.01150144,
      "auxiliary_loss_mlp": 0.01025408,
      "balance_loss_clip": 1.0460577,
      "balance_loss_mlp": 1.01871991,
      "epoch": 0.9793783442553959,
      "flos": 23728191033600.0,
      "grad_norm": 1.827865833262024,
      "language_loss": 0.77738929,
      "learning_rate": 4.434197028803766e-09,
      "loss": 0.79914486,
      "num_input_tokens_seen": 175865975,
      "step": 8145,
      "time_per_iteration": 2.5160465240478516
    },
    {
      "auxiliary_loss_clip": 0.01124774,
      "auxiliary_loss_mlp": 0.01028424,
      "balance_loss_clip": 1.0405333,
      "balance_loss_mlp": 1.02154589,
      "epoch": 0.979498587146035,
      "flos": 23038742407680.0,
      "grad_norm": 1.9806929308350252,
      "language_loss": 0.82265782,
      "learning_rate": 4.3825056117805514e-09,
      "loss": 0.84418976,
      "num_input_tokens_seen": 175881860,
      "step": 8146,
      "time_per_iteration": 2.5685462951660156
    },
    {
      "auxiliary_loss_clip": 0.01163813,
      "auxiliary_loss_mlp": 0.0102437,
      "balance_loss_clip": 1.04701364,
      "balance_loss_mlp": 1.01741147,
      "epoch": 0.979618830036674,
      "flos": 14318841951360.0,
      "grad_norm": 2.0551798304332283,
      "language_loss": 0.79296279,
      "learning_rate": 4.331116928344425e-09,
      "loss": 0.81484455,
      "num_input_tokens_seen": 175898175,
      "step": 8147,
      "time_per_iteration": 2.4252238273620605
    },
    {
      "auxiliary_loss_clip": 0.0114019,
      "auxiliary_loss_mlp": 0.0076096,
      "balance_loss_clip": 1.04375589,
      "balance_loss_mlp": 1.00028753,
      "epoch": 0.9797390729273132,
      "flos": 16727514215040.0,
      "grad_norm": 2.0273246086046766,
      "language_loss": 0.62441576,
      "learning_rate": 4.28003098629115e-09,
      "loss": 0.64342725,
      "num_input_tokens_seen": 175914310,
      "step": 8148,
      "time_per_iteration": 2.5265822410583496
    },
    {
      "auxiliary_loss_clip": 0.01115449,
      "auxiliary_loss_mlp": 0.01023977,
      "balance_loss_clip": 1.03632116,
      "balance_loss_mlp": 1.01725399,
      "epoch": 0.9798593158179523,
      "flos": 24532661986560.0,
      "grad_norm": 1.6846564874758947,
      "language_loss": 0.78381091,
      "learning_rate": 4.229247793370305e-09,
      "loss": 0.80520517,
      "num_input_tokens_seen": 175933435,
      "step": 8149,
      "time_per_iteration": 2.5859546661376953
    },
    {
      "auxiliary_loss_clip": 0.01164656,
      "auxiliary_loss_mlp": 0.01028653,
      "balance_loss_clip": 1.04818368,
      "balance_loss_mlp": 1.02172685,
      "epoch": 0.9799795587085913,
      "flos": 27308808339840.0,
      "grad_norm": 1.5806613636903595,
      "language_loss": 0.70384502,
      "learning_rate": 4.178767357285951e-09,
      "loss": 0.7257781,
      "num_input_tokens_seen": 175955065,
      "step": 8150,
      "time_per_iteration": 2.5207607746124268
    },
    {
      "auxiliary_loss_clip": 0.01150257,
      "auxiliary_loss_mlp": 0.00760564,
      "balance_loss_clip": 1.0465585,
      "balance_loss_mlp": 1.00031447,
      "epoch": 0.9800998015992305,
      "flos": 26286575184000.0,
      "grad_norm": 2.4914055156214086,
      "language_loss": 0.7162239,
      "learning_rate": 4.128589685695516e-09,
      "loss": 0.73533213,
      "num_input_tokens_seen": 175975490,
      "step": 8151,
      "time_per_iteration": 2.547163724899292
    },
    {
      "auxiliary_loss_clip": 0.01162251,
      "auxiliary_loss_mlp": 0.01033398,
      "balance_loss_clip": 1.04668212,
      "balance_loss_mlp": 1.0265491,
      "epoch": 0.9802200444898695,
      "flos": 16723635546240.0,
      "grad_norm": 1.9612872546750713,
      "language_loss": 0.84468496,
      "learning_rate": 4.078714786211135e-09,
      "loss": 0.86664146,
      "num_input_tokens_seen": 175991340,
      "step": 8152,
      "time_per_iteration": 2.425194501876831
    },
    {
      "auxiliary_loss_clip": 0.01147926,
      "auxiliary_loss_mlp": 0.01025418,
      "balance_loss_clip": 1.04599524,
      "balance_loss_mlp": 1.01916528,
      "epoch": 0.9803402873805086,
      "flos": 24900459298560.0,
      "grad_norm": 1.6359488173187708,
      "language_loss": 0.76825488,
      "learning_rate": 4.029142666398977e-09,
      "loss": 0.78998828,
      "num_input_tokens_seen": 176011505,
      "step": 8153,
      "time_per_iteration": 2.5276401042938232
    },
    {
      "auxiliary_loss_clip": 0.01160425,
      "auxiliary_loss_mlp": 0.01029032,
      "balance_loss_clip": 1.04720986,
      "balance_loss_mlp": 1.02241588,
      "epoch": 0.9804605302711478,
      "flos": 22564937082240.0,
      "grad_norm": 1.807123162345723,
      "language_loss": 0.79807222,
      "learning_rate": 3.979873333778805e-09,
      "loss": 0.81996679,
      "num_input_tokens_seen": 176029680,
      "step": 8154,
      "time_per_iteration": 2.4637038707733154
    },
    {
      "auxiliary_loss_clip": 0.01140608,
      "auxiliary_loss_mlp": 0.01024811,
      "balance_loss_clip": 1.0459919,
      "balance_loss_mlp": 1.01807821,
      "epoch": 0.9805807731617868,
      "flos": 38905368382080.0,
      "grad_norm": 1.944079679902687,
      "language_loss": 0.73945308,
      "learning_rate": 3.930906795824862e-09,
      "loss": 0.76110721,
      "num_input_tokens_seen": 176050355,
      "step": 8155,
      "time_per_iteration": 2.6672439575195312
    },
    {
      "auxiliary_loss_clip": 0.01145554,
      "auxiliary_loss_mlp": 0.01024052,
      "balance_loss_clip": 1.04515743,
      "balance_loss_mlp": 1.01749516,
      "epoch": 0.9807010160524259,
      "flos": 17821999578240.0,
      "grad_norm": 2.243285063647544,
      "language_loss": 0.76883078,
      "learning_rate": 3.882243059965207e-09,
      "loss": 0.79052675,
      "num_input_tokens_seen": 176068070,
      "step": 8156,
      "time_per_iteration": 3.195885181427002
    },
    {
      "auxiliary_loss_clip": 0.01141307,
      "auxiliary_loss_mlp": 0.0102517,
      "balance_loss_clip": 1.04347301,
      "balance_loss_mlp": 1.01805639,
      "epoch": 0.980821258943065,
      "flos": 13552975140480.0,
      "grad_norm": 3.082380597786366,
      "language_loss": 0.65473318,
      "learning_rate": 3.833882133582156e-09,
      "loss": 0.67639798,
      "num_input_tokens_seen": 176083730,
      "step": 8157,
      "time_per_iteration": 2.4624996185302734
    },
    {
      "auxiliary_loss_clip": 0.01150785,
      "auxiliary_loss_mlp": 0.01026981,
      "balance_loss_clip": 1.04545939,
      "balance_loss_mlp": 1.02066338,
      "epoch": 0.9809415018337041,
      "flos": 21689794120320.0,
      "grad_norm": 1.5425903302566295,
      "language_loss": 0.77942395,
      "learning_rate": 3.785824024012285e-09,
      "loss": 0.80120158,
      "num_input_tokens_seen": 176102730,
      "step": 8158,
      "time_per_iteration": 2.4978280067443848
    },
    {
      "auxiliary_loss_clip": 0.0112631,
      "auxiliary_loss_mlp": 0.01030107,
      "balance_loss_clip": 1.04366827,
      "balance_loss_mlp": 1.02387261,
      "epoch": 0.9810617447243432,
      "flos": 23294857357440.0,
      "grad_norm": 1.523929864464651,
      "language_loss": 0.78236234,
      "learning_rate": 3.738068738545541e-09,
      "loss": 0.80392647,
      "num_input_tokens_seen": 176121815,
      "step": 8159,
      "time_per_iteration": 2.530754566192627
    },
    {
      "auxiliary_loss_clip": 0.01152703,
      "auxiliary_loss_mlp": 0.01030112,
      "balance_loss_clip": 1.04649401,
      "balance_loss_mlp": 1.02295971,
      "epoch": 0.9811819876149822,
      "flos": 18332038748160.0,
      "grad_norm": 2.1719848361027903,
      "language_loss": 0.78346843,
      "learning_rate": 3.6906162844265733e-09,
      "loss": 0.8052966,
      "num_input_tokens_seen": 176138900,
      "step": 8160,
      "time_per_iteration": 2.4768030643463135
    },
    {
      "auxiliary_loss_clip": 0.0113101,
      "auxiliary_loss_mlp": 0.01031797,
      "balance_loss_clip": 1.04454792,
      "balance_loss_mlp": 1.02485943,
      "epoch": 0.9813022305056214,
      "flos": 22601961025920.0,
      "grad_norm": 1.936235984993442,
      "language_loss": 0.70592302,
      "learning_rate": 3.643466668853845e-09,
      "loss": 0.7275511,
      "num_input_tokens_seen": 176156925,
      "step": 8161,
      "time_per_iteration": 2.540848970413208
    },
    {
      "auxiliary_loss_clip": 0.01137902,
      "auxiliary_loss_mlp": 0.01026633,
      "balance_loss_clip": 1.04392147,
      "balance_loss_mlp": 1.02026391,
      "epoch": 0.9814224733962604,
      "flos": 25413335642880.0,
      "grad_norm": 1.90135152675509,
      "language_loss": 0.75154173,
      "learning_rate": 3.59661989898008e-09,
      "loss": 0.7731871,
      "num_input_tokens_seen": 176177980,
      "step": 8162,
      "time_per_iteration": 2.59548282623291
    },
    {
      "auxiliary_loss_clip": 0.01116425,
      "auxiliary_loss_mlp": 0.01024571,
      "balance_loss_clip": 1.0449959,
      "balance_loss_mlp": 1.0179193,
      "epoch": 0.9815427162868995,
      "flos": 25007185584000.0,
      "grad_norm": 1.6147359137053718,
      "language_loss": 0.76838565,
      "learning_rate": 3.5500759819115934e-09,
      "loss": 0.78979558,
      "num_input_tokens_seen": 176198345,
      "step": 8163,
      "time_per_iteration": 2.5875895023345947
    },
    {
      "auxiliary_loss_clip": 0.01166815,
      "auxiliary_loss_mlp": 0.01030114,
      "balance_loss_clip": 1.0508287,
      "balance_loss_mlp": 1.02325988,
      "epoch": 0.9816629591775387,
      "flos": 20662604887680.0,
      "grad_norm": 2.1436356060995982,
      "language_loss": 0.80874419,
      "learning_rate": 3.5038349247094034e-09,
      "loss": 0.83071351,
      "num_input_tokens_seen": 176215605,
      "step": 8164,
      "time_per_iteration": 2.45546293258667
    },
    {
      "auxiliary_loss_clip": 0.01135142,
      "auxiliary_loss_mlp": 0.01024724,
      "balance_loss_clip": 1.0424993,
      "balance_loss_mlp": 1.01777995,
      "epoch": 0.9817832020681777,
      "flos": 17712220636800.0,
      "grad_norm": 2.0860430882892826,
      "language_loss": 0.77082467,
      "learning_rate": 3.4578967343878994e-09,
      "loss": 0.79242337,
      "num_input_tokens_seen": 176231810,
      "step": 8165,
      "time_per_iteration": 2.5021111965179443
    },
    {
      "auxiliary_loss_clip": 0.01134597,
      "auxiliary_loss_mlp": 0.01023959,
      "balance_loss_clip": 1.04554451,
      "balance_loss_mlp": 1.01714611,
      "epoch": 0.9819034449588168,
      "flos": 22530032040960.0,
      "grad_norm": 1.7193013271618676,
      "language_loss": 0.80814213,
      "learning_rate": 3.4122614179161733e-09,
      "loss": 0.82972765,
      "num_input_tokens_seen": 176251770,
      "step": 8166,
      "time_per_iteration": 2.5480239391326904
    },
    {
      "auxiliary_loss_clip": 0.01107681,
      "auxiliary_loss_mlp": 0.01027632,
      "balance_loss_clip": 1.03865242,
      "balance_loss_mlp": 1.02103114,
      "epoch": 0.9820236878494559,
      "flos": 20011221699840.0,
      "grad_norm": 1.6451849046050369,
      "language_loss": 0.77908462,
      "learning_rate": 3.36692898221691e-09,
      "loss": 0.80043781,
      "num_input_tokens_seen": 176270135,
      "step": 8167,
      "time_per_iteration": 2.5637834072113037
    },
    {
      "auxiliary_loss_clip": 0.01147385,
      "auxiliary_loss_mlp": 0.01025606,
      "balance_loss_clip": 1.0446595,
      "balance_loss_mlp": 1.01954091,
      "epoch": 0.982143930740095,
      "flos": 18807316531200.0,
      "grad_norm": 1.6167771871881964,
      "language_loss": 0.73665965,
      "learning_rate": 3.3218994341668305e-09,
      "loss": 0.75838953,
      "num_input_tokens_seen": 176289065,
      "step": 8168,
      "time_per_iteration": 2.4940452575683594
    },
    {
      "auxiliary_loss_clip": 0.01163367,
      "auxiliary_loss_mlp": 0.01029162,
      "balance_loss_clip": 1.04997993,
      "balance_loss_mlp": 1.0226531,
      "epoch": 0.982264173630734,
      "flos": 26578026138240.0,
      "grad_norm": 1.5734498918607238,
      "language_loss": 0.75553048,
      "learning_rate": 3.2771727805971373e-09,
      "loss": 0.77745575,
      "num_input_tokens_seen": 176310450,
      "step": 8169,
      "time_per_iteration": 3.3301572799682617
    },
    {
      "auxiliary_loss_clip": 0.0110206,
      "auxiliary_loss_mlp": 0.01027451,
      "balance_loss_clip": 1.03947198,
      "balance_loss_mlp": 1.02011633,
      "epoch": 0.9823844165213732,
      "flos": 22014462176640.0,
      "grad_norm": 1.77313643247294,
      "language_loss": 0.77467692,
      "learning_rate": 3.232749028292847e-09,
      "loss": 0.79597199,
      "num_input_tokens_seen": 176327415,
      "step": 8170,
      "time_per_iteration": 3.370128631591797
    },
    {
      "auxiliary_loss_clip": 0.01164443,
      "auxiliary_loss_mlp": 0.0102961,
      "balance_loss_clip": 1.04705179,
      "balance_loss_mlp": 1.02237439,
      "epoch": 0.9825046594120123,
      "flos": 21908166854400.0,
      "grad_norm": 1.6360489136248668,
      "language_loss": 0.88105559,
      "learning_rate": 3.188628183992792e-09,
      "loss": 0.90299618,
      "num_input_tokens_seen": 176347680,
      "step": 8171,
      "time_per_iteration": 2.472970724105835
    },
    {
      "auxiliary_loss_clip": 0.01045749,
      "auxiliary_loss_mlp": 0.01001841,
      "balance_loss_clip": 1.00737262,
      "balance_loss_mlp": 1.00078571,
      "epoch": 0.9826249023026513,
      "flos": 59494610718720.0,
      "grad_norm": 0.7403031592516569,
      "language_loss": 0.62527519,
      "learning_rate": 3.1448102543902844e-09,
      "loss": 0.64575112,
      "num_input_tokens_seen": 176411595,
      "step": 8172,
      "time_per_iteration": 3.0478413105010986
    },
    {
      "auxiliary_loss_clip": 0.01128375,
      "auxiliary_loss_mlp": 0.01027192,
      "balance_loss_clip": 1.04328918,
      "balance_loss_mlp": 1.0203644,
      "epoch": 0.9827451451932905,
      "flos": 16071031296000.0,
      "grad_norm": 1.9463393444583774,
      "language_loss": 0.67570108,
      "learning_rate": 3.1012952461324515e-09,
      "loss": 0.69725674,
      "num_input_tokens_seen": 176430570,
      "step": 8173,
      "time_per_iteration": 2.5091633796691895
    },
    {
      "auxiliary_loss_clip": 0.01149032,
      "auxiliary_loss_mlp": 0.010253,
      "balance_loss_clip": 1.0477891,
      "balance_loss_mlp": 1.01845717,
      "epoch": 0.9828653880839295,
      "flos": 20262775622400.0,
      "grad_norm": 2.14252282874822,
      "language_loss": 0.73904163,
      "learning_rate": 3.0580831658204575e-09,
      "loss": 0.76078492,
      "num_input_tokens_seen": 176448150,
      "step": 8174,
      "time_per_iteration": 2.4705028533935547
    },
    {
      "auxiliary_loss_clip": 0.01149618,
      "auxiliary_loss_mlp": 0.01026352,
      "balance_loss_clip": 1.0478164,
      "balance_loss_mlp": 1.01959848,
      "epoch": 0.9829856309745686,
      "flos": 21616141282560.0,
      "grad_norm": 1.4709225684114746,
      "language_loss": 0.77836394,
      "learning_rate": 3.015174020009281e-09,
      "loss": 0.80012363,
      "num_input_tokens_seen": 176467475,
      "step": 8175,
      "time_per_iteration": 2.5138156414031982
    },
    {
      "auxiliary_loss_clip": 0.01121332,
      "auxiliary_loss_mlp": 0.01022186,
      "balance_loss_clip": 1.03983045,
      "balance_loss_mlp": 1.01562333,
      "epoch": 0.9831058738652078,
      "flos": 23764209396480.0,
      "grad_norm": 2.0264487779550513,
      "language_loss": 0.75084436,
      "learning_rate": 2.9725678152086043e-09,
      "loss": 0.7722795,
      "num_input_tokens_seen": 176486045,
      "step": 8176,
      "time_per_iteration": 2.571632146835327
    },
    {
      "auxiliary_loss_clip": 0.01122336,
      "auxiliary_loss_mlp": 0.01025216,
      "balance_loss_clip": 1.04176581,
      "balance_loss_mlp": 1.0181644,
      "epoch": 0.9832261167558468,
      "flos": 11320911072000.0,
      "grad_norm": 3.54638297370488,
      "language_loss": 0.82084394,
      "learning_rate": 2.930264557881257e-09,
      "loss": 0.84231943,
      "num_input_tokens_seen": 176501230,
      "step": 8177,
      "time_per_iteration": 2.482571601867676
    },
    {
      "auxiliary_loss_clip": 0.01054537,
      "auxiliary_loss_mlp": 0.01002671,
      "balance_loss_clip": 1.00779736,
      "balance_loss_mlp": 1.00159192,
      "epoch": 0.9833463596464859,
      "flos": 60000304343040.0,
      "grad_norm": 0.837948524885748,
      "language_loss": 0.582165,
      "learning_rate": 2.8882642544452163e-09,
      "loss": 0.60273707,
      "num_input_tokens_seen": 176565955,
      "step": 8178,
      "time_per_iteration": 3.089775562286377
    },
    {
      "auxiliary_loss_clip": 0.01125621,
      "auxiliary_loss_mlp": 0.0102081,
      "balance_loss_clip": 1.04209661,
      "balance_loss_mlp": 1.01400924,
      "epoch": 0.983466602537125,
      "flos": 13626699805440.0,
      "grad_norm": 2.147168718196381,
      "language_loss": 0.7452659,
      "learning_rate": 2.8465669112716083e-09,
      "loss": 0.76673025,
      "num_input_tokens_seen": 176583480,
      "step": 8179,
      "time_per_iteration": 2.486344814300537
    },
    {
      "auxiliary_loss_clip": 0.01150106,
      "auxiliary_loss_mlp": 0.00761714,
      "balance_loss_clip": 1.04458237,
      "balance_loss_mlp": 1.00031996,
      "epoch": 0.9835868454277641,
      "flos": 22926844563840.0,
      "grad_norm": 1.9981548481843108,
      "language_loss": 0.76324677,
      "learning_rate": 2.8051725346858177e-09,
      "loss": 0.78236496,
      "num_input_tokens_seen": 176603740,
      "step": 8180,
      "time_per_iteration": 2.508835554122925
    },
    {
      "auxiliary_loss_clip": 0.0116353,
      "auxiliary_loss_mlp": 0.01025067,
      "balance_loss_clip": 1.04562807,
      "balance_loss_mlp": 1.01797724,
      "epoch": 0.9837070883184031,
      "flos": 27673409341440.0,
      "grad_norm": 1.9940841843442412,
      "language_loss": 0.71152163,
      "learning_rate": 2.7640811309674883e-09,
      "loss": 0.7334075,
      "num_input_tokens_seen": 176623240,
      "step": 8181,
      "time_per_iteration": 2.501286506652832
    },
    {
      "auxiliary_loss_clip": 0.01113113,
      "auxiliary_loss_mlp": 0.01024681,
      "balance_loss_clip": 1.04401636,
      "balance_loss_mlp": 1.01770735,
      "epoch": 0.9838273312090423,
      "flos": 29241951425280.0,
      "grad_norm": 1.6707796148732172,
      "language_loss": 0.80710185,
      "learning_rate": 2.7232927063498557e-09,
      "loss": 0.82847977,
      "num_input_tokens_seen": 176643615,
      "step": 8182,
      "time_per_iteration": 3.3174688816070557
    },
    {
      "auxiliary_loss_clip": 0.01148228,
      "auxiliary_loss_mlp": 0.01025948,
      "balance_loss_clip": 1.04555023,
      "balance_loss_mlp": 1.01896858,
      "epoch": 0.9839475740996814,
      "flos": 40110207304320.0,
      "grad_norm": 1.8327314822473735,
      "language_loss": 0.69184673,
      "learning_rate": 2.682807267020859e-09,
      "loss": 0.71358848,
      "num_input_tokens_seen": 176666375,
      "step": 8183,
      "time_per_iteration": 2.6475791931152344
    },
    {
      "auxiliary_loss_clip": 0.01147294,
      "auxiliary_loss_mlp": 0.01029271,
      "balance_loss_clip": 1.04522562,
      "balance_loss_mlp": 1.02233005,
      "epoch": 0.9840678169903204,
      "flos": 24169389788160.0,
      "grad_norm": 1.5488926685734743,
      "language_loss": 0.62547517,
      "learning_rate": 2.642624819121808e-09,
      "loss": 0.64724076,
      "num_input_tokens_seen": 176686525,
      "step": 8184,
      "time_per_iteration": 2.532592296600342
    },
    {
      "auxiliary_loss_clip": 0.0113582,
      "auxiliary_loss_mlp": 0.01028416,
      "balance_loss_clip": 1.04625916,
      "balance_loss_mlp": 1.02200222,
      "epoch": 0.9841880598809596,
      "flos": 14684484447360.0,
      "grad_norm": 2.001601571641615,
      "language_loss": 0.61615413,
      "learning_rate": 2.6027453687487154e-09,
      "loss": 0.63779652,
      "num_input_tokens_seen": 176703615,
      "step": 8185,
      "time_per_iteration": 2.491265058517456
    },
    {
      "auxiliary_loss_clip": 0.01135011,
      "auxiliary_loss_mlp": 0.01028785,
      "balance_loss_clip": 1.04439139,
      "balance_loss_mlp": 1.02129292,
      "epoch": 0.9843083027715986,
      "flos": 22344768668160.0,
      "grad_norm": 2.2284485073033244,
      "language_loss": 0.53784055,
      "learning_rate": 2.5631689219509643e-09,
      "loss": 0.55947852,
      "num_input_tokens_seen": 176722295,
      "step": 8186,
      "time_per_iteration": 2.541133165359497
    },
    {
      "auxiliary_loss_clip": 0.01133447,
      "auxiliary_loss_mlp": 0.01024994,
      "balance_loss_clip": 1.04689288,
      "balance_loss_mlp": 1.01868439,
      "epoch": 0.9844285456622377,
      "flos": 21800111765760.0,
      "grad_norm": 1.6577787944144287,
      "language_loss": 0.83786309,
      "learning_rate": 2.523895484732197e-09,
      "loss": 0.85944748,
      "num_input_tokens_seen": 176741750,
      "step": 8187,
      "time_per_iteration": 2.5377063751220703
    },
    {
      "auxiliary_loss_clip": 0.01153862,
      "auxiliary_loss_mlp": 0.01025956,
      "balance_loss_clip": 1.04559278,
      "balance_loss_mlp": 1.01889884,
      "epoch": 0.9845487885528769,
      "flos": 18035380321920.0,
      "grad_norm": 1.9360269030646078,
      "language_loss": 0.74586439,
      "learning_rate": 2.4849250630505357e-09,
      "loss": 0.76766253,
      "num_input_tokens_seen": 176759995,
      "step": 8188,
      "time_per_iteration": 2.4962258338928223
    },
    {
      "auxiliary_loss_clip": 0.01069828,
      "auxiliary_loss_mlp": 0.01024351,
      "balance_loss_clip": 1.03855348,
      "balance_loss_mlp": 1.01779437,
      "epoch": 0.9846690314435159,
      "flos": 25228610974080.0,
      "grad_norm": 1.7174683639460764,
      "language_loss": 0.72983849,
      "learning_rate": 2.4462576628172528e-09,
      "loss": 0.75078028,
      "num_input_tokens_seen": 176778625,
      "step": 8189,
      "time_per_iteration": 2.6770622730255127
    },
    {
      "auxiliary_loss_clip": 0.01148284,
      "auxiliary_loss_mlp": 0.01028986,
      "balance_loss_clip": 1.04738927,
      "balance_loss_mlp": 1.02219677,
      "epoch": 0.984789274334155,
      "flos": 18552171248640.0,
      "grad_norm": 1.7361001528416078,
      "language_loss": 0.73796427,
      "learning_rate": 2.407893289898766e-09,
      "loss": 0.75973701,
      "num_input_tokens_seen": 176797655,
      "step": 8190,
      "time_per_iteration": 2.4910712242126465
    },
    {
      "auxiliary_loss_clip": 0.01115925,
      "auxiliary_loss_mlp": 0.01023943,
      "balance_loss_clip": 1.04220855,
      "balance_loss_mlp": 1.01725245,
      "epoch": 0.984909517224794,
      "flos": 27345437233920.0,
      "grad_norm": 1.9460504451299345,
      "language_loss": 0.83923459,
      "learning_rate": 2.3698319501144202e-09,
      "loss": 0.86063337,
      "num_input_tokens_seen": 176818640,
      "step": 8191,
      "time_per_iteration": 2.5962395668029785
    },
    {
      "auxiliary_loss_clip": 0.0115095,
      "auxiliary_loss_mlp": 0.01024646,
      "balance_loss_clip": 1.04669356,
      "balance_loss_mlp": 1.01764596,
      "epoch": 0.9850297601154332,
      "flos": 18734058743040.0,
      "grad_norm": 1.554356296684526,
      "language_loss": 0.73114651,
      "learning_rate": 2.3320736492382644e-09,
      "loss": 0.75290251,
      "num_input_tokens_seen": 176837475,
      "step": 8192,
      "time_per_iteration": 2.470937967300415
    },
    {
      "auxiliary_loss_clip": 0.01159692,
      "auxiliary_loss_mlp": 0.01025763,
      "balance_loss_clip": 1.04689169,
      "balance_loss_mlp": 1.01916146,
      "epoch": 0.9851500030060723,
      "flos": 22308247514880.0,
      "grad_norm": 2.2062386970828256,
      "language_loss": 0.67761075,
      "learning_rate": 2.29461839299816e-09,
      "loss": 0.69946539,
      "num_input_tokens_seen": 176857190,
      "step": 8193,
      "time_per_iteration": 2.4660072326660156
    },
    {
      "auxiliary_loss_clip": 0.01120551,
      "auxiliary_loss_mlp": 0.01022984,
      "balance_loss_clip": 1.04248714,
      "balance_loss_mlp": 1.01652241,
      "epoch": 0.9852702458967113,
      "flos": 26353691746560.0,
      "grad_norm": 1.522430066956628,
      "language_loss": 0.79833531,
      "learning_rate": 2.257466187076229e-09,
      "loss": 0.81977069,
      "num_input_tokens_seen": 176876395,
      "step": 8194,
      "time_per_iteration": 2.5858240127563477
    },
    {
      "auxiliary_loss_clip": 0.01153673,
      "auxiliary_loss_mlp": 0.00760894,
      "balance_loss_clip": 1.04569829,
      "balance_loss_mlp": 1.00025594,
      "epoch": 0.9853904887873505,
      "flos": 20883599314560.0,
      "grad_norm": 1.7394319295092493,
      "language_loss": 0.71054125,
      "learning_rate": 2.2206170371081854e-09,
      "loss": 0.72968698,
      "num_input_tokens_seen": 176894980,
      "step": 8195,
      "time_per_iteration": 3.260167121887207
    },
    {
      "auxiliary_loss_clip": 0.01135969,
      "auxiliary_loss_mlp": 0.01025883,
      "balance_loss_clip": 1.0438931,
      "balance_loss_mlp": 1.0192225,
      "epoch": 0.9855107316779895,
      "flos": 25263444188160.0,
      "grad_norm": 1.6220214230925114,
      "language_loss": 0.84774864,
      "learning_rate": 2.1840709486842247e-09,
      "loss": 0.86936718,
      "num_input_tokens_seen": 176914600,
      "step": 8196,
      "time_per_iteration": 3.356483221054077
    },
    {
      "auxiliary_loss_clip": 0.0112816,
      "auxiliary_loss_mlp": 0.01034031,
      "balance_loss_clip": 1.04369676,
      "balance_loss_mlp": 1.02709615,
      "epoch": 0.9856309745686286,
      "flos": 19062102677760.0,
      "grad_norm": 2.0303278437867434,
      "language_loss": 0.79370087,
      "learning_rate": 2.1478279273481335e-09,
      "loss": 0.81532276,
      "num_input_tokens_seen": 176933085,
      "step": 8197,
      "time_per_iteration": 2.4991955757141113
    },
    {
      "auxiliary_loss_clip": 0.01148086,
      "auxiliary_loss_mlp": 0.01027056,
      "balance_loss_clip": 1.04735041,
      "balance_loss_mlp": 1.02015352,
      "epoch": 0.9857512174592677,
      "flos": 34130758060800.0,
      "grad_norm": 2.0942337449043946,
      "language_loss": 0.79973018,
      "learning_rate": 2.1118879785981815e-09,
      "loss": 0.82148159,
      "num_input_tokens_seen": 176953225,
      "step": 8198,
      "time_per_iteration": 2.608306646347046
    },
    {
      "auxiliary_loss_clip": 0.01132618,
      "auxiliary_loss_mlp": 0.01022905,
      "balance_loss_clip": 1.04415452,
      "balance_loss_mlp": 1.01602328,
      "epoch": 0.9858714603499068,
      "flos": 25994693266560.0,
      "grad_norm": 2.232802340793844,
      "language_loss": 0.79398656,
      "learning_rate": 2.0762511078862288e-09,
      "loss": 0.8155418,
      "num_input_tokens_seen": 176973570,
      "step": 8199,
      "time_per_iteration": 2.5608739852905273
    },
    {
      "auxiliary_loss_clip": 0.0113902,
      "auxiliary_loss_mlp": 0.01023708,
      "balance_loss_clip": 1.04515886,
      "balance_loss_mlp": 1.01715171,
      "epoch": 0.9859917032405459,
      "flos": 23696230907520.0,
      "grad_norm": 2.962773814585931,
      "language_loss": 0.64849603,
      "learning_rate": 2.0409173206186183e-09,
      "loss": 0.67012328,
      "num_input_tokens_seen": 176992810,
      "step": 8200,
      "time_per_iteration": 2.568518877029419
    },
    {
      "auxiliary_loss_clip": 0.01118744,
      "auxiliary_loss_mlp": 0.0102324,
      "balance_loss_clip": 1.04428244,
      "balance_loss_mlp": 1.01658225,
      "epoch": 0.986111946131185,
      "flos": 19938287134080.0,
      "grad_norm": 1.8272063480607426,
      "language_loss": 0.87122774,
      "learning_rate": 2.0058866221550617e-09,
      "loss": 0.89264762,
      "num_input_tokens_seen": 177011050,
      "step": 8201,
      "time_per_iteration": 2.5547945499420166
    },
    {
      "auxiliary_loss_clip": 0.01163604,
      "auxiliary_loss_mlp": 0.01022761,
      "balance_loss_clip": 1.04645944,
      "balance_loss_mlp": 1.01557899,
      "epoch": 0.9862321890218241,
      "flos": 19828831415040.0,
      "grad_norm": 2.193537908167759,
      "language_loss": 0.74987245,
      "learning_rate": 1.971159017809976e-09,
      "loss": 0.77173615,
      "num_input_tokens_seen": 177029340,
      "step": 8202,
      "time_per_iteration": 2.5150418281555176
    },
    {
      "auxiliary_loss_clip": 0.01150488,
      "auxiliary_loss_mlp": 0.01031273,
      "balance_loss_clip": 1.04754865,
      "balance_loss_mlp": 1.02417433,
      "epoch": 0.9863524319124631,
      "flos": 21652051904640.0,
      "grad_norm": 2.793705109920593,
      "language_loss": 0.77965641,
      "learning_rate": 1.93673451285159e-09,
      "loss": 0.80147398,
      "num_input_tokens_seen": 177048390,
      "step": 8203,
      "time_per_iteration": 2.4971063137054443
    },
    {
      "auxiliary_loss_clip": 0.01035985,
      "auxiliary_loss_mlp": 0.01003809,
      "balance_loss_clip": 1.00736725,
      "balance_loss_mlp": 1.00273645,
      "epoch": 0.9864726748031023,
      "flos": 52769977920000.0,
      "grad_norm": 0.7498271653482068,
      "language_loss": 0.56573653,
      "learning_rate": 1.9026131125019495e-09,
      "loss": 0.58613443,
      "num_input_tokens_seen": 177105760,
      "step": 8204,
      "time_per_iteration": 3.0454232692718506
    },
    {
      "auxiliary_loss_clip": 0.01142332,
      "auxiliary_loss_mlp": 0.01025106,
      "balance_loss_clip": 1.04477835,
      "balance_loss_mlp": 1.0185194,
      "epoch": 0.9865929176937414,
      "flos": 23364631526400.0,
      "grad_norm": 1.7962566798958002,
      "language_loss": 0.8679539,
      "learning_rate": 1.8687948219371363e-09,
      "loss": 0.88962829,
      "num_input_tokens_seen": 177124985,
      "step": 8205,
      "time_per_iteration": 2.4930319786071777
    },
    {
      "auxiliary_loss_clip": 0.01164306,
      "auxiliary_loss_mlp": 0.01024172,
      "balance_loss_clip": 1.04511404,
      "balance_loss_mlp": 1.01668513,
      "epoch": 0.9867131605843804,
      "flos": 21616679986560.0,
      "grad_norm": 1.996033335828532,
      "language_loss": 0.88128829,
      "learning_rate": 1.835279646287491e-09,
      "loss": 0.90317303,
      "num_input_tokens_seen": 177142995,
      "step": 8206,
      "time_per_iteration": 2.4537179470062256
    },
    {
      "auxiliary_loss_clip": 0.01158139,
      "auxiliary_loss_mlp": 0.01036924,
      "balance_loss_clip": 1.04872131,
      "balance_loss_mlp": 1.02980685,
      "epoch": 0.9868334034750196,
      "flos": 22271403139200.0,
      "grad_norm": 3.206933660395623,
      "language_loss": 0.76368988,
      "learning_rate": 1.8020675906371685e-09,
      "loss": 0.7856406,
      "num_input_tokens_seen": 177162390,
      "step": 8207,
      "time_per_iteration": 2.5134470462799072
    },
    {
      "auxiliary_loss_clip": 0.0110676,
      "auxiliary_loss_mlp": 0.01030019,
      "balance_loss_clip": 1.0414598,
      "balance_loss_mlp": 1.02328944,
      "epoch": 0.9869536463656586,
      "flos": 25809573548160.0,
      "grad_norm": 3.983203842884186,
      "language_loss": 0.75255853,
      "learning_rate": 1.7691586600243612e-09,
      "loss": 0.77392626,
      "num_input_tokens_seen": 177181290,
      "step": 8208,
      "time_per_iteration": 2.622799873352051
    },
    {
      "auxiliary_loss_clip": 0.01131381,
      "auxiliary_loss_mlp": 0.01026236,
      "balance_loss_clip": 1.04603016,
      "balance_loss_mlp": 1.01971531,
      "epoch": 0.9870738892562977,
      "flos": 16398500613120.0,
      "grad_norm": 2.2645542638476392,
      "language_loss": 0.86302316,
      "learning_rate": 1.7365528594415202e-09,
      "loss": 0.88459933,
      "num_input_tokens_seen": 177195360,
      "step": 8209,
      "time_per_iteration": 3.2535054683685303
    },
    {
      "auxiliary_loss_clip": 0.01153677,
      "auxiliary_loss_mlp": 0.00761043,
      "balance_loss_clip": 1.04635656,
      "balance_loss_mlp": 1.00031209,
      "epoch": 0.9871941321469369,
      "flos": 35481358373760.0,
      "grad_norm": 1.5976337467065016,
      "language_loss": 0.6728065,
      "learning_rate": 1.7042501938346888e-09,
      "loss": 0.69195366,
      "num_input_tokens_seen": 177218090,
      "step": 8210,
      "time_per_iteration": 2.6417734622955322
    },
    {
      "auxiliary_loss_clip": 0.01122655,
      "auxiliary_loss_mlp": 0.01024303,
      "balance_loss_clip": 1.03955686,
      "balance_loss_mlp": 1.01805091,
      "epoch": 0.9873143750375759,
      "flos": 21434217874560.0,
      "grad_norm": 1.8517183646738022,
      "language_loss": 0.76141292,
      "learning_rate": 1.6722506681043913e-09,
      "loss": 0.78288245,
      "num_input_tokens_seen": 177237050,
      "step": 8211,
      "time_per_iteration": 2.521176338195801
    },
    {
      "auxiliary_loss_clip": 0.01138876,
      "auxiliary_loss_mlp": 0.01031927,
      "balance_loss_clip": 1.04464686,
      "balance_loss_mlp": 1.02525151,
      "epoch": 0.987434617928215,
      "flos": 16326499800960.0,
      "grad_norm": 2.1497479606657097,
      "language_loss": 0.69620323,
      "learning_rate": 1.640554287104745e-09,
      "loss": 0.7179113,
      "num_input_tokens_seen": 177255325,
      "step": 8212,
      "time_per_iteration": 2.4906036853790283
    },
    {
      "auxiliary_loss_clip": 0.01122875,
      "auxiliary_loss_mlp": 0.01022217,
      "balance_loss_clip": 1.03986454,
      "balance_loss_mlp": 1.01468611,
      "epoch": 0.9875548608188541,
      "flos": 17851984456320.0,
      "grad_norm": 2.013528553741755,
      "language_loss": 0.79918098,
      "learning_rate": 1.609161055644348e-09,
      "loss": 0.82063198,
      "num_input_tokens_seen": 177271250,
      "step": 8213,
      "time_per_iteration": 2.536616086959839
    },
    {
      "auxiliary_loss_clip": 0.01154903,
      "auxiliary_loss_mlp": 0.01025499,
      "balance_loss_clip": 1.04550719,
      "balance_loss_mlp": 1.01810479,
      "epoch": 0.9876751037094932,
      "flos": 26132876887680.0,
      "grad_norm": 3.450638883567828,
      "language_loss": 0.68048888,
      "learning_rate": 1.5780709784849467e-09,
      "loss": 0.70229292,
      "num_input_tokens_seen": 177288270,
      "step": 8214,
      "time_per_iteration": 2.5331637859344482
    },
    {
      "auxiliary_loss_clip": 0.01098144,
      "auxiliary_loss_mlp": 0.01023337,
      "balance_loss_clip": 1.042045,
      "balance_loss_mlp": 1.01654243,
      "epoch": 0.9877953466001322,
      "flos": 15991344973440.0,
      "grad_norm": 1.8151629525076283,
      "language_loss": 0.82468772,
      "learning_rate": 1.5472840603436565e-09,
      "loss": 0.8459025,
      "num_input_tokens_seen": 177305500,
      "step": 8215,
      "time_per_iteration": 2.5798938274383545
    },
    {
      "auxiliary_loss_clip": 0.01135109,
      "auxiliary_loss_mlp": 0.01021263,
      "balance_loss_clip": 1.04520011,
      "balance_loss_mlp": 1.0145483,
      "epoch": 0.9879155894907714,
      "flos": 18806777827200.0,
      "grad_norm": 2.0134818481240244,
      "language_loss": 0.78353298,
      "learning_rate": 1.5168003058900757e-09,
      "loss": 0.80509675,
      "num_input_tokens_seen": 177323500,
      "step": 8216,
      "time_per_iteration": 2.519059896469116
    },
    {
      "auxiliary_loss_clip": 0.01120079,
      "auxiliary_loss_mlp": 0.01025222,
      "balance_loss_clip": 1.04205334,
      "balance_loss_mlp": 1.01897287,
      "epoch": 0.9880358323814105,
      "flos": 22382044007040.0,
      "grad_norm": 1.9036590323907412,
      "language_loss": 0.92143029,
      "learning_rate": 1.4866197197491715e-09,
      "loss": 0.94288325,
      "num_input_tokens_seen": 177342860,
      "step": 8217,
      "time_per_iteration": 2.5625405311584473
    },
    {
      "auxiliary_loss_clip": 0.01151728,
      "auxiliary_loss_mlp": 0.00761152,
      "balance_loss_clip": 1.04529512,
      "balance_loss_mlp": 1.00034046,
      "epoch": 0.9881560752720495,
      "flos": 15668831733120.0,
      "grad_norm": 3.1567063740449544,
      "language_loss": 0.7880404,
      "learning_rate": 1.4567423064988371e-09,
      "loss": 0.8071692,
      "num_input_tokens_seen": 177360210,
      "step": 8218,
      "time_per_iteration": 2.5072479248046875
    },
    {
      "auxiliary_loss_clip": 0.01164529,
      "auxiliary_loss_mlp": 0.01024509,
      "balance_loss_clip": 1.04771662,
      "balance_loss_mlp": 1.01802969,
      "epoch": 0.9882763181626887,
      "flos": 21500113374720.0,
      "grad_norm": 1.975171414687983,
      "language_loss": 0.78038013,
      "learning_rate": 1.4271680706718913e-09,
      "loss": 0.80227053,
      "num_input_tokens_seen": 177377885,
      "step": 8219,
      "time_per_iteration": 2.4703071117401123
    },
    {
      "auxiliary_loss_clip": 0.01150343,
      "auxiliary_loss_mlp": 0.0103172,
      "balance_loss_clip": 1.04665148,
      "balance_loss_mlp": 1.02440667,
      "epoch": 0.9883965610533277,
      "flos": 28034598551040.0,
      "grad_norm": 1.5987444560432211,
      "language_loss": 0.8229087,
      "learning_rate": 1.3978970167543013e-09,
      "loss": 0.8447293,
      "num_input_tokens_seen": 177398065,
      "step": 8220,
      "time_per_iteration": 2.561683177947998
    },
    {
      "auxiliary_loss_clip": 0.01127812,
      "auxiliary_loss_mlp": 0.01026548,
      "balance_loss_clip": 1.04373085,
      "balance_loss_mlp": 1.01929426,
      "epoch": 0.9885168039439668,
      "flos": 14098601710080.0,
      "grad_norm": 2.364618533429402,
      "language_loss": 0.77380878,
      "learning_rate": 1.3689291491867372e-09,
      "loss": 0.7953524,
      "num_input_tokens_seen": 177416380,
      "step": 8221,
      "time_per_iteration": 3.2372937202453613
    },
    {
      "auxiliary_loss_clip": 0.01164601,
      "auxiliary_loss_mlp": 0.01027007,
      "balance_loss_clip": 1.04731178,
      "balance_loss_mlp": 1.01975906,
      "epoch": 0.988637046834606,
      "flos": 26432013352320.0,
      "grad_norm": 1.921853861705802,
      "language_loss": 0.73580325,
      "learning_rate": 1.3402644723636836e-09,
      "loss": 0.75771928,
      "num_input_tokens_seen": 177438410,
      "step": 8222,
      "time_per_iteration": 3.2701165676116943
    },
    {
      "auxiliary_loss_clip": 0.01131952,
      "auxiliary_loss_mlp": 0.01026661,
      "balance_loss_clip": 1.04692733,
      "balance_loss_mlp": 1.01980627,
      "epoch": 0.988757289725245,
      "flos": 25229113764480.0,
      "grad_norm": 2.605076301146215,
      "language_loss": 0.83521318,
      "learning_rate": 1.311902990633218e-09,
      "loss": 0.85679924,
      "num_input_tokens_seen": 177457375,
      "step": 8223,
      "time_per_iteration": 2.559272050857544
    },
    {
      "auxiliary_loss_clip": 0.01123821,
      "auxiliary_loss_mlp": 0.0102579,
      "balance_loss_clip": 1.03776658,
      "balance_loss_mlp": 1.01902509,
      "epoch": 0.9888775326158841,
      "flos": 26359042872960.0,
      "grad_norm": 1.626584511073724,
      "language_loss": 0.71252638,
      "learning_rate": 1.2838447082978987e-09,
      "loss": 0.7340225,
      "num_input_tokens_seen": 177478530,
      "step": 8224,
      "time_per_iteration": 2.5671772956848145
    },
    {
      "auxiliary_loss_clip": 0.01145468,
      "auxiliary_loss_mlp": 0.01027006,
      "balance_loss_clip": 1.0429107,
      "balance_loss_mlp": 1.01980293,
      "epoch": 0.9889977755065231,
      "flos": 24316120846080.0,
      "grad_norm": 2.554812434572028,
      "language_loss": 0.83076161,
      "learning_rate": 1.2560896296143208e-09,
      "loss": 0.85248631,
      "num_input_tokens_seen": 177496995,
      "step": 8225,
      "time_per_iteration": 2.4873874187469482
    },
    {
      "auxiliary_loss_clip": 0.01161744,
      "auxiliary_loss_mlp": 0.01031345,
      "balance_loss_clip": 1.04676151,
      "balance_loss_mlp": 1.02448404,
      "epoch": 0.9891180183971623,
      "flos": 18951066760320.0,
      "grad_norm": 2.3495612003462685,
      "language_loss": 0.82457685,
      "learning_rate": 1.2286377587926722e-09,
      "loss": 0.84650779,
      "num_input_tokens_seen": 177513785,
      "step": 8226,
      "time_per_iteration": 2.45072078704834
    },
    {
      "auxiliary_loss_clip": 0.01161806,
      "auxiliary_loss_mlp": 0.01027999,
      "balance_loss_clip": 1.045506,
      "balance_loss_mlp": 1.02109683,
      "epoch": 0.9892382612878013,
      "flos": 26176580760960.0,
      "grad_norm": 1.9790991968782994,
      "language_loss": 0.74568784,
      "learning_rate": 1.2014890999973992e-09,
      "loss": 0.76758587,
      "num_input_tokens_seen": 177530705,
      "step": 8227,
      "time_per_iteration": 2.470072031021118
    },
    {
      "auxiliary_loss_clip": 0.011592,
      "auxiliary_loss_mlp": 0.01022905,
      "balance_loss_clip": 1.04446363,
      "balance_loss_mlp": 1.01611948,
      "epoch": 0.9893585041784404,
      "flos": 25449605400960.0,
      "grad_norm": 1.4224390814015884,
      "language_loss": 0.78233516,
      "learning_rate": 1.1746436573472073e-09,
      "loss": 0.80415618,
      "num_input_tokens_seen": 177552440,
      "step": 8228,
      "time_per_iteration": 2.5024070739746094
    },
    {
      "auxiliary_loss_clip": 0.01144111,
      "auxiliary_loss_mlp": 0.01028103,
      "balance_loss_clip": 1.04526353,
      "balance_loss_mlp": 1.02089095,
      "epoch": 0.9894787470690796,
      "flos": 20189302352640.0,
      "grad_norm": 1.9558620046210509,
      "language_loss": 0.69219267,
      "learning_rate": 1.1481014349141726e-09,
      "loss": 0.71391475,
      "num_input_tokens_seen": 177569660,
      "step": 8229,
      "time_per_iteration": 2.483761787414551
    },
    {
      "auxiliary_loss_clip": 0.01136575,
      "auxiliary_loss_mlp": 0.01030235,
      "balance_loss_clip": 1.04510188,
      "balance_loss_mlp": 1.02309096,
      "epoch": 0.9895989899597186,
      "flos": 24644308435200.0,
      "grad_norm": 1.9585461652833278,
      "language_loss": 0.84365648,
      "learning_rate": 1.121862436724852e-09,
      "loss": 0.86532462,
      "num_input_tokens_seen": 177588500,
      "step": 8230,
      "time_per_iteration": 2.557873487472534
    },
    {
      "auxiliary_loss_clip": 0.0115073,
      "auxiliary_loss_mlp": 0.01026706,
      "balance_loss_clip": 1.04829359,
      "balance_loss_mlp": 1.01961899,
      "epoch": 0.9897192328503577,
      "flos": 21799034357760.0,
      "grad_norm": 1.7210756587880818,
      "language_loss": 0.70398289,
      "learning_rate": 1.0959266667598388e-09,
      "loss": 0.7257573,
      "num_input_tokens_seen": 177607315,
      "step": 8231,
      "time_per_iteration": 2.4850831031799316
    },
    {
      "auxiliary_loss_clip": 0.01124728,
      "auxiliary_loss_mlp": 0.01027234,
      "balance_loss_clip": 1.04511976,
      "balance_loss_mlp": 1.01941407,
      "epoch": 0.9898394757409968,
      "flos": 21325229032320.0,
      "grad_norm": 1.8996835612537466,
      "language_loss": 0.74465835,
      "learning_rate": 1.0702941289533196e-09,
      "loss": 0.76617795,
      "num_input_tokens_seen": 177625990,
      "step": 8232,
      "time_per_iteration": 2.561312675476074
    },
    {
      "auxiliary_loss_clip": 0.01120903,
      "auxiliary_loss_mlp": 0.01021785,
      "balance_loss_clip": 1.04407215,
      "balance_loss_mlp": 1.01529741,
      "epoch": 0.9899597186316359,
      "flos": 18545024442240.0,
      "grad_norm": 2.013369576142551,
      "language_loss": 0.88509655,
      "learning_rate": 1.0449648271939615e-09,
      "loss": 0.90652347,
      "num_input_tokens_seen": 177642335,
      "step": 8233,
      "time_per_iteration": 2.519787311553955
    },
    {
      "auxiliary_loss_clip": 0.01110563,
      "auxiliary_loss_mlp": 0.00760718,
      "balance_loss_clip": 1.04354286,
      "balance_loss_mlp": 1.00030005,
      "epoch": 0.990079961522275,
      "flos": 23766723348480.0,
      "grad_norm": 1.5738199474029242,
      "language_loss": 0.72332215,
      "learning_rate": 1.0199387653240243e-09,
      "loss": 0.74203503,
      "num_input_tokens_seen": 177662025,
      "step": 8234,
      "time_per_iteration": 2.620490789413452
    },
    {
      "auxiliary_loss_clip": 0.01129792,
      "auxiliary_loss_mlp": 0.01028316,
      "balance_loss_clip": 1.04358482,
      "balance_loss_mlp": 1.02179849,
      "epoch": 0.9902002044129141,
      "flos": 16399182971520.0,
      "grad_norm": 1.571913397748716,
      "language_loss": 0.70571208,
      "learning_rate": 9.952159471400267e-10,
      "loss": 0.72729319,
      "num_input_tokens_seen": 177679065,
      "step": 8235,
      "time_per_iteration": 3.262352466583252
    },
    {
      "auxiliary_loss_clip": 0.01146125,
      "auxiliary_loss_mlp": 0.00760843,
      "balance_loss_clip": 1.04368079,
      "balance_loss_mlp": 1.00033343,
      "epoch": 0.9903204473035532,
      "flos": 22559657783040.0,
      "grad_norm": 1.8115604247725943,
      "language_loss": 0.84358883,
      "learning_rate": 9.707963763923022e-10,
      "loss": 0.86265856,
      "num_input_tokens_seen": 177698115,
      "step": 8236,
      "time_per_iteration": 2.5048279762268066
    },
    {
      "auxiliary_loss_clip": 0.01132087,
      "auxiliary_loss_mlp": 0.01029499,
      "balance_loss_clip": 1.04136646,
      "balance_loss_mlp": 1.02265596,
      "epoch": 0.9904406901941922,
      "flos": 16144001775360.0,
      "grad_norm": 1.712476354820829,
      "language_loss": 0.79204631,
      "learning_rate": 9.466800567854427e-10,
      "loss": 0.81366211,
      "num_input_tokens_seen": 177716715,
      "step": 8237,
      "time_per_iteration": 2.512277603149414
    },
    {
      "auxiliary_loss_clip": 0.01116969,
      "auxiliary_loss_mlp": 0.0103291,
      "balance_loss_clip": 1.03966057,
      "balance_loss_mlp": 1.02526832,
      "epoch": 0.9905609330848314,
      "flos": 26651499408000.0,
      "grad_norm": 1.9871302730263303,
      "language_loss": 0.68110609,
      "learning_rate": 9.228669919778553e-10,
      "loss": 0.70260483,
      "num_input_tokens_seen": 177735640,
      "step": 8238,
      "time_per_iteration": 2.598832607269287
    },
    {
      "auxiliary_loss_clip": 0.01126816,
      "auxiliary_loss_mlp": 0.01029805,
      "balance_loss_clip": 1.04293346,
      "balance_loss_mlp": 1.02220201,
      "epoch": 0.9906811759754705,
      "flos": 23111820627840.0,
      "grad_norm": 2.5266020650754917,
      "language_loss": 0.79402167,
      "learning_rate": 8.993571855817617e-10,
      "loss": 0.81558788,
      "num_input_tokens_seen": 177754470,
      "step": 8239,
      "time_per_iteration": 2.5466606616973877
    },
    {
      "auxiliary_loss_clip": 0.01146196,
      "auxiliary_loss_mlp": 0.01029063,
      "balance_loss_clip": 1.0442344,
      "balance_loss_mlp": 1.02205384,
      "epoch": 0.9908014188661095,
      "flos": 22090593052800.0,
      "grad_norm": 1.726437701185197,
      "language_loss": 0.74972498,
      "learning_rate": 8.761506411638642e-10,
      "loss": 0.77147758,
      "num_input_tokens_seen": 177773935,
      "step": 8240,
      "time_per_iteration": 2.504885196685791
    },
    {
      "auxiliary_loss_clip": 0.01132727,
      "auxiliary_loss_mlp": 0.0102744,
      "balance_loss_clip": 1.04496789,
      "balance_loss_mlp": 1.02053499,
      "epoch": 0.9909216617567487,
      "flos": 19242948677760.0,
      "grad_norm": 1.6442008899685288,
      "language_loss": 0.73800987,
      "learning_rate": 8.53247362244236e-10,
      "loss": 0.75961155,
      "num_input_tokens_seen": 177792745,
      "step": 8241,
      "time_per_iteration": 2.620187997817993
    },
    {
      "auxiliary_loss_clip": 0.01135702,
      "auxiliary_loss_mlp": 0.01025289,
      "balance_loss_clip": 1.04525709,
      "balance_loss_mlp": 1.01841688,
      "epoch": 0.9910419046473877,
      "flos": 23621213352960.0,
      "grad_norm": 1.624548295150006,
      "language_loss": 0.68210828,
      "learning_rate": 8.306473522976532e-10,
      "loss": 0.70371819,
      "num_input_tokens_seen": 177812150,
      "step": 8242,
      "time_per_iteration": 2.6017086505889893
    },
    {
      "auxiliary_loss_clip": 0.01163104,
      "auxiliary_loss_mlp": 0.01023171,
      "balance_loss_clip": 1.04809892,
      "balance_loss_mlp": 1.01636446,
      "epoch": 0.9911621475380268,
      "flos": 22711380831360.0,
      "grad_norm": 1.841881164736898,
      "language_loss": 0.71777678,
      "learning_rate": 8.083506147522623e-10,
      "loss": 0.73963952,
      "num_input_tokens_seen": 177831545,
      "step": 8243,
      "time_per_iteration": 2.4943394660949707
    },
    {
      "auxiliary_loss_clip": 0.01144494,
      "auxiliary_loss_mlp": 0.0103207,
      "balance_loss_clip": 1.04450583,
      "balance_loss_mlp": 1.02518296,
      "epoch": 0.991282390428666,
      "flos": 13516956777600.0,
      "grad_norm": 2.0285459582719585,
      "language_loss": 0.85283607,
      "learning_rate": 7.863571529906909e-10,
      "loss": 0.87460172,
      "num_input_tokens_seen": 177847130,
      "step": 8244,
      "time_per_iteration": 2.461913585662842
    },
    {
      "auxiliary_loss_clip": 0.01045539,
      "auxiliary_loss_mlp": 0.01002473,
      "balance_loss_clip": 1.00771713,
      "balance_loss_mlp": 1.00138807,
      "epoch": 0.991402633319305,
      "flos": 61830492071040.0,
      "grad_norm": 0.723432033885823,
      "language_loss": 0.5969578,
      "learning_rate": 7.646669703489372e-10,
      "loss": 0.61743796,
      "num_input_tokens_seen": 177911440,
      "step": 8245,
      "time_per_iteration": 3.191058874130249
    },
    {
      "auxiliary_loss_clip": 0.0105614,
      "auxiliary_loss_mlp": 0.01025123,
      "balance_loss_clip": 1.03320813,
      "balance_loss_mlp": 1.01825953,
      "epoch": 0.9915228762099441,
      "flos": 18770148933120.0,
      "grad_norm": 2.000644186691286,
      "language_loss": 0.57299036,
      "learning_rate": 7.432800701177023e-10,
      "loss": 0.59380305,
      "num_input_tokens_seen": 177929440,
      "step": 8246,
      "time_per_iteration": 2.9199795722961426
    },
    {
      "auxiliary_loss_clip": 0.01038188,
      "auxiliary_loss_mlp": 0.01003859,
      "balance_loss_clip": 1.01003313,
      "balance_loss_mlp": 1.00272608,
      "epoch": 0.9916431191005832,
      "flos": 65936660244480.0,
      "grad_norm": 0.7970068630094421,
      "language_loss": 0.57820463,
      "learning_rate": 7.221964555415017e-10,
      "loss": 0.59862506,
      "num_input_tokens_seen": 177989100,
      "step": 8247,
      "time_per_iteration": 3.3220455646514893
    },
    {
      "auxiliary_loss_clip": 0.01131756,
      "auxiliary_loss_mlp": 0.01025492,
      "balance_loss_clip": 1.04293537,
      "balance_loss_mlp": 1.01891792,
      "epoch": 0.9917633619912223,
      "flos": 16581573256320.0,
      "grad_norm": 1.7686047441929207,
      "language_loss": 0.74613559,
      "learning_rate": 7.01416129818222e-10,
      "loss": 0.76770806,
      "num_input_tokens_seen": 178006720,
      "step": 8248,
      "time_per_iteration": 5.224648475646973
    },
    {
      "auxiliary_loss_clip": 0.01129829,
      "auxiliary_loss_mlp": 0.01033557,
      "balance_loss_clip": 1.0453217,
      "balance_loss_mlp": 1.02676797,
      "epoch": 0.9918836048818613,
      "flos": 25411108999680.0,
      "grad_norm": 1.86511210999876,
      "language_loss": 0.58109784,
      "learning_rate": 6.809390961006745e-10,
      "loss": 0.60273176,
      "num_input_tokens_seen": 178026850,
      "step": 8249,
      "time_per_iteration": 2.628622531890869
    },
    {
      "auxiliary_loss_clip": 0.01133489,
      "auxiliary_loss_mlp": 0.01027643,
      "balance_loss_clip": 1.04473758,
      "balance_loss_mlp": 1.0209372,
      "epoch": 0.9920038477725005,
      "flos": 25046867134080.0,
      "grad_norm": 1.7874624000010249,
      "language_loss": 0.6868434,
      "learning_rate": 6.607653574948191e-10,
      "loss": 0.70845461,
      "num_input_tokens_seen": 178047630,
      "step": 8250,
      "time_per_iteration": 2.5689046382904053
    },
    {
      "auxiliary_loss_clip": 0.01140737,
      "auxiliary_loss_mlp": 0.0102462,
      "balance_loss_clip": 1.0421226,
      "balance_loss_mlp": 1.0180968,
      "epoch": 0.9921240906631396,
      "flos": 21829773421440.0,
      "grad_norm": 1.9909335729117181,
      "language_loss": 0.81322026,
      "learning_rate": 6.408949170613187e-10,
      "loss": 0.83487386,
      "num_input_tokens_seen": 178066895,
      "step": 8251,
      "time_per_iteration": 2.5335917472839355
    },
    {
      "auxiliary_loss_clip": 0.01133822,
      "auxiliary_loss_mlp": 0.01025561,
      "balance_loss_clip": 1.04313517,
      "balance_loss_mlp": 1.0179975,
      "epoch": 0.9922443335537786,
      "flos": 24864225454080.0,
      "grad_norm": 1.5339671870432552,
      "language_loss": 0.81686127,
      "learning_rate": 6.213277778144288e-10,
      "loss": 0.83845508,
      "num_input_tokens_seen": 178088540,
      "step": 8252,
      "time_per_iteration": 2.5822854042053223
    },
    {
      "auxiliary_loss_clip": 0.01093083,
      "auxiliary_loss_mlp": 0.01026512,
      "balance_loss_clip": 1.03903866,
      "balance_loss_mlp": 1.01973176,
      "epoch": 0.9923645764444178,
      "flos": 21613088626560.0,
      "grad_norm": 2.012503473027396,
      "language_loss": 0.66732574,
      "learning_rate": 6.020639427224416e-10,
      "loss": 0.68852174,
      "num_input_tokens_seen": 178106185,
      "step": 8253,
      "time_per_iteration": 2.647977113723755
    },
    {
      "auxiliary_loss_clip": 0.01138293,
      "auxiliary_loss_mlp": 0.01033463,
      "balance_loss_clip": 1.0463984,
      "balance_loss_mlp": 1.026281,
      "epoch": 0.9924848193350568,
      "flos": 25001798544000.0,
      "grad_norm": 1.9179504151595543,
      "language_loss": 0.72529209,
      "learning_rate": 5.831034147076864e-10,
      "loss": 0.74700975,
      "num_input_tokens_seen": 178123435,
      "step": 8254,
      "time_per_iteration": 2.56266188621521
    },
    {
      "auxiliary_loss_clip": 0.01041809,
      "auxiliary_loss_mlp": 0.01001653,
      "balance_loss_clip": 1.00751042,
      "balance_loss_mlp": 1.00063956,
      "epoch": 0.9926050622256959,
      "flos": 68912543151360.0,
      "grad_norm": 0.6861660800769509,
      "language_loss": 0.5571776,
      "learning_rate": 5.644461966463065e-10,
      "loss": 0.57761222,
      "num_input_tokens_seen": 178191045,
      "step": 8255,
      "time_per_iteration": 3.1965014934539795
    },
    {
      "auxiliary_loss_clip": 0.01134776,
      "auxiliary_loss_mlp": 0.01023885,
      "balance_loss_clip": 1.04640961,
      "balance_loss_mlp": 1.0176084,
      "epoch": 0.9927253051163349,
      "flos": 20923675914240.0,
      "grad_norm": 1.74973038744095,
      "language_loss": 0.75481474,
      "learning_rate": 5.460922913687049e-10,
      "loss": 0.77640134,
      "num_input_tokens_seen": 178210135,
      "step": 8256,
      "time_per_iteration": 2.561108112335205
    },
    {
      "auxiliary_loss_clip": 0.01104357,
      "auxiliary_loss_mlp": 0.00761282,
      "balance_loss_clip": 1.03893089,
      "balance_loss_mlp": 1.00031126,
      "epoch": 0.9928455480069741,
      "flos": 22308211601280.0,
      "grad_norm": 1.9479586282224022,
      "language_loss": 0.75148511,
      "learning_rate": 5.280417016593208e-10,
      "loss": 0.77014148,
      "num_input_tokens_seen": 178229925,
      "step": 8257,
      "time_per_iteration": 2.6225130558013916
    },
    {
      "auxiliary_loss_clip": 0.01148667,
      "auxiliary_loss_mlp": 0.00760374,
      "balance_loss_clip": 1.04801011,
      "balance_loss_mlp": 1.00031579,
      "epoch": 0.9929657908976132,
      "flos": 17383889393280.0,
      "grad_norm": 1.6202572454842186,
      "language_loss": 0.74849093,
      "learning_rate": 5.102944302559642e-10,
      "loss": 0.76758134,
      "num_input_tokens_seen": 178247420,
      "step": 8258,
      "time_per_iteration": 2.4791719913482666
    },
    {
      "auxiliary_loss_clip": 0.01098333,
      "auxiliary_loss_mlp": 0.01027366,
      "balance_loss_clip": 1.03991568,
      "balance_loss_mlp": 1.02008581,
      "epoch": 0.9930860337882522,
      "flos": 22674680110080.0,
      "grad_norm": 2.002865524421131,
      "language_loss": 0.79429346,
      "learning_rate": 4.9285047985137e-10,
      "loss": 0.81555045,
      "num_input_tokens_seen": 178266840,
      "step": 8259,
      "time_per_iteration": 2.6615512371063232
    },
    {
      "auxiliary_loss_clip": 0.0115328,
      "auxiliary_loss_mlp": 0.01025202,
      "balance_loss_clip": 1.04713154,
      "balance_loss_mlp": 1.01811182,
      "epoch": 0.9932062766788914,
      "flos": 28147789284480.0,
      "grad_norm": 1.7540982261873745,
      "language_loss": 0.74542493,
      "learning_rate": 4.757098530916436e-10,
      "loss": 0.76720977,
      "num_input_tokens_seen": 178287285,
      "step": 8260,
      "time_per_iteration": 2.5757315158843994
    },
    {
      "auxiliary_loss_clip": 0.01151571,
      "auxiliary_loss_mlp": 0.01034117,
      "balance_loss_clip": 1.04770994,
      "balance_loss_mlp": 1.02638054,
      "epoch": 0.9933265195695304,
      "flos": 20156659868160.0,
      "grad_norm": 3.153098093849159,
      "language_loss": 0.76993167,
      "learning_rate": 4.5887255257670563e-10,
      "loss": 0.79178852,
      "num_input_tokens_seen": 178304325,
      "step": 8261,
      "time_per_iteration": 3.2127346992492676
    },
    {
      "auxiliary_loss_clip": 0.01161722,
      "auxiliary_loss_mlp": 0.01027057,
      "balance_loss_clip": 1.04608214,
      "balance_loss_mlp": 1.0199374,
      "epoch": 0.9934467624601695,
      "flos": 21362037494400.0,
      "grad_norm": 2.02111997817168,
      "language_loss": 0.76946223,
      "learning_rate": 4.4233858086117906e-10,
      "loss": 0.79135007,
      "num_input_tokens_seen": 178322850,
      "step": 8262,
      "time_per_iteration": 2.4835519790649414
    },
    {
      "auxiliary_loss_clip": 0.01106765,
      "auxiliary_loss_mlp": 0.01024837,
      "balance_loss_clip": 1.04657459,
      "balance_loss_mlp": 1.01760149,
      "epoch": 0.9935670053508087,
      "flos": 19756040503680.0,
      "grad_norm": 9.714070513744952,
      "language_loss": 0.67707002,
      "learning_rate": 4.261079404528356e-10,
      "loss": 0.69838601,
      "num_input_tokens_seen": 178342330,
      "step": 8263,
      "time_per_iteration": 2.58988356590271
    },
    {
      "auxiliary_loss_clip": 0.0114498,
      "auxiliary_loss_mlp": 0.01028123,
      "balance_loss_clip": 1.04365969,
      "balance_loss_mlp": 1.0211997,
      "epoch": 0.9936872482414477,
      "flos": 21978838863360.0,
      "grad_norm": 1.7509387414411932,
      "language_loss": 0.69003582,
      "learning_rate": 4.1018063381437205e-10,
      "loss": 0.7117669,
      "num_input_tokens_seen": 178362715,
      "step": 8264,
      "time_per_iteration": 2.521165132522583
    },
    {
      "auxiliary_loss_clip": 0.01038792,
      "auxiliary_loss_mlp": 0.01001495,
      "balance_loss_clip": 1.0084796,
      "balance_loss_mlp": 1.00050592,
      "epoch": 0.9938074911320868,
      "flos": 69810667839360.0,
      "grad_norm": 0.8672093628831413,
      "language_loss": 0.6121372,
      "learning_rate": 3.9455666336141167e-10,
      "loss": 0.63254005,
      "num_input_tokens_seen": 178426495,
      "step": 8265,
      "time_per_iteration": 3.1320581436157227
    },
    {
      "auxiliary_loss_clip": 0.01163977,
      "auxiliary_loss_mlp": 0.01024371,
      "balance_loss_clip": 1.04938245,
      "balance_loss_mlp": 1.0171324,
      "epoch": 0.9939277340227259,
      "flos": 15084170058240.0,
      "grad_norm": 4.226307397096864,
      "language_loss": 0.82864684,
      "learning_rate": 3.7923603146450267e-10,
      "loss": 0.85053033,
      "num_input_tokens_seen": 178442555,
      "step": 8266,
      "time_per_iteration": 2.437268018722534
    },
    {
      "auxiliary_loss_clip": 0.01120861,
      "auxiliary_loss_mlp": 0.01027898,
      "balance_loss_clip": 1.04052806,
      "balance_loss_mlp": 1.02114749,
      "epoch": 0.994047976913365,
      "flos": 17712364291200.0,
      "grad_norm": 2.0514783637448093,
      "language_loss": 0.80019104,
      "learning_rate": 3.642187404473418e-10,
      "loss": 0.82167864,
      "num_input_tokens_seen": 178460715,
      "step": 8267,
      "time_per_iteration": 2.5241494178771973
    },
    {
      "auxiliary_loss_clip": 0.01148497,
      "auxiliary_loss_mlp": 0.01022546,
      "balance_loss_clip": 1.044119,
      "balance_loss_mlp": 1.01616549,
      "epoch": 0.994168219804004,
      "flos": 19171558396800.0,
      "grad_norm": 2.0150538113329417,
      "language_loss": 0.858181,
      "learning_rate": 3.495047925885508e-10,
      "loss": 0.8798914,
      "num_input_tokens_seen": 178479050,
      "step": 8268,
      "time_per_iteration": 2.484640598297119
    },
    {
      "auxiliary_loss_clip": 0.011323,
      "auxiliary_loss_mlp": 0.01029348,
      "balance_loss_clip": 1.04295444,
      "balance_loss_mlp": 1.02204657,
      "epoch": 0.9942884626946432,
      "flos": 17851589406720.0,
      "grad_norm": 2.024992408144145,
      "language_loss": 0.82485622,
      "learning_rate": 3.350941901199e-10,
      "loss": 0.84647274,
      "num_input_tokens_seen": 178495970,
      "step": 8269,
      "time_per_iteration": 2.505986452102661
    },
    {
      "auxiliary_loss_clip": 0.01137386,
      "auxiliary_loss_mlp": 0.01028062,
      "balance_loss_clip": 1.04319561,
      "balance_loss_mlp": 1.02091837,
      "epoch": 0.9944087055852823,
      "flos": 18796578364800.0,
      "grad_norm": 2.370854847107553,
      "language_loss": 0.83335185,
      "learning_rate": 3.2098693522764066e-10,
      "loss": 0.85500634,
      "num_input_tokens_seen": 178509170,
      "step": 8270,
      "time_per_iteration": 2.5096073150634766
    },
    {
      "auxiliary_loss_clip": 0.01140958,
      "auxiliary_loss_mlp": 0.00760637,
      "balance_loss_clip": 1.04443812,
      "balance_loss_mlp": 1.00028133,
      "epoch": 0.9945289484759213,
      "flos": 20996969616000.0,
      "grad_norm": 1.9514665098082589,
      "language_loss": 0.8108741,
      "learning_rate": 3.071830300516165e-10,
      "loss": 0.82989013,
      "num_input_tokens_seen": 178527000,
      "step": 8271,
      "time_per_iteration": 2.5300827026367188
    },
    {
      "auxiliary_loss_clip": 0.01155083,
      "auxiliary_loss_mlp": 0.01032204,
      "balance_loss_clip": 1.04580247,
      "balance_loss_mlp": 1.02440202,
      "epoch": 0.9946491913665605,
      "flos": 14756952136320.0,
      "grad_norm": 2.536991268510617,
      "language_loss": 0.70575607,
      "learning_rate": 2.9368247668615234e-10,
      "loss": 0.72762895,
      "num_input_tokens_seen": 178545590,
      "step": 8272,
      "time_per_iteration": 2.4701550006866455
    },
    {
      "auxiliary_loss_clip": 0.01169532,
      "auxiliary_loss_mlp": 0.01030814,
      "balance_loss_clip": 1.05077982,
      "balance_loss_mlp": 1.02278531,
      "epoch": 0.9947694342571995,
      "flos": 12669931186560.0,
      "grad_norm": 2.305085448935493,
      "language_loss": 0.61534834,
      "learning_rate": 2.804852771789434e-10,
      "loss": 0.63735169,
      "num_input_tokens_seen": 178558890,
      "step": 8273,
      "time_per_iteration": 3.163057327270508
    },
    {
      "auxiliary_loss_clip": 0.01160099,
      "auxiliary_loss_mlp": 0.0102705,
      "balance_loss_clip": 1.04595995,
      "balance_loss_mlp": 1.02031183,
      "epoch": 0.9948896771478386,
      "flos": 18843442634880.0,
      "grad_norm": 1.7121309036992203,
      "language_loss": 0.55949914,
      "learning_rate": 2.675914335321661e-10,
      "loss": 0.58137071,
      "num_input_tokens_seen": 178577645,
      "step": 8274,
      "time_per_iteration": 3.2814059257507324
    },
    {
      "auxiliary_loss_clip": 0.01154009,
      "auxiliary_loss_mlp": 0.01029494,
      "balance_loss_clip": 1.04646564,
      "balance_loss_mlp": 1.02187109,
      "epoch": 0.9950099200384778,
      "flos": 24900207903360.0,
      "grad_norm": 2.332244309402494,
      "language_loss": 0.78862119,
      "learning_rate": 2.550009477018111e-10,
      "loss": 0.81045628,
      "num_input_tokens_seen": 178596415,
      "step": 8275,
      "time_per_iteration": 2.530961513519287
    },
    {
      "auxiliary_loss_clip": 0.01135944,
      "auxiliary_loss_mlp": 0.00761084,
      "balance_loss_clip": 1.04614568,
      "balance_loss_mlp": 1.00030708,
      "epoch": 0.9951301629291168,
      "flos": 23733613987200.0,
      "grad_norm": 1.8633928903625439,
      "language_loss": 0.63044238,
      "learning_rate": 2.4271382159790634e-10,
      "loss": 0.64941263,
      "num_input_tokens_seen": 178613845,
      "step": 8276,
      "time_per_iteration": 2.556291341781616
    },
    {
      "auxiliary_loss_clip": 0.0110119,
      "auxiliary_loss_mlp": 0.01030644,
      "balance_loss_clip": 1.0423528,
      "balance_loss_mlp": 1.02355993,
      "epoch": 0.9952504058197559,
      "flos": 22236893147520.0,
      "grad_norm": 3.100525542981254,
      "language_loss": 0.85798132,
      "learning_rate": 2.3073005708429406e-10,
      "loss": 0.87929964,
      "num_input_tokens_seen": 178633490,
      "step": 8277,
      "time_per_iteration": 2.711155891418457
    },
    {
      "auxiliary_loss_clip": 0.0111937,
      "auxiliary_loss_mlp": 0.01030169,
      "balance_loss_clip": 1.04615808,
      "balance_loss_mlp": 1.02336514,
      "epoch": 0.995370648710395,
      "flos": 21211032718080.0,
      "grad_norm": 1.757792554636068,
      "language_loss": 0.72062683,
      "learning_rate": 2.190496559788535e-10,
      "loss": 0.74212229,
      "num_input_tokens_seen": 178651775,
      "step": 8278,
      "time_per_iteration": 2.758758306503296
    },
    {
      "auxiliary_loss_clip": 0.01132923,
      "auxiliary_loss_mlp": 0.01028474,
      "balance_loss_clip": 1.04462183,
      "balance_loss_mlp": 1.0214076,
      "epoch": 0.9954908916010341,
      "flos": 14866731077760.0,
      "grad_norm": 2.175126843485569,
      "language_loss": 0.76892829,
      "learning_rate": 2.0767262005372265e-10,
      "loss": 0.79054219,
      "num_input_tokens_seen": 178669290,
      "step": 8279,
      "time_per_iteration": 2.534245014190674
    },
    {
      "auxiliary_loss_clip": 0.01123476,
      "auxiliary_loss_mlp": 0.01025329,
      "balance_loss_clip": 1.04259968,
      "balance_loss_mlp": 1.01863277,
      "epoch": 0.9956111344916732,
      "flos": 19208259118080.0,
      "grad_norm": 2.7156187235500395,
      "language_loss": 0.75649679,
      "learning_rate": 1.965989510346322e-10,
      "loss": 0.77798486,
      "num_input_tokens_seen": 178688410,
      "step": 8280,
      "time_per_iteration": 2.5697991847991943
    },
    {
      "auxiliary_loss_clip": 0.01102474,
      "auxiliary_loss_mlp": 0.01024854,
      "balance_loss_clip": 1.04103434,
      "balance_loss_mlp": 1.01701009,
      "epoch": 0.9957313773823123,
      "flos": 20047060494720.0,
      "grad_norm": 2.1810898914497607,
      "language_loss": 0.71021485,
      "learning_rate": 1.8582865060134955e-10,
      "loss": 0.73148811,
      "num_input_tokens_seen": 178706600,
      "step": 8281,
      "time_per_iteration": 2.594902753829956
    },
    {
      "auxiliary_loss_clip": 0.01054195,
      "auxiliary_loss_mlp": 0.01002704,
      "balance_loss_clip": 1.00752318,
      "balance_loss_mlp": 1.00170314,
      "epoch": 0.9958516202729514,
      "flos": 57483253768320.0,
      "grad_norm": 0.7765989122426452,
      "language_loss": 0.55758941,
      "learning_rate": 1.7536172038790098e-10,
      "loss": 0.57815844,
      "num_input_tokens_seen": 178766910,
      "step": 8282,
      "time_per_iteration": 3.1289587020874023
    },
    {
      "auxiliary_loss_clip": 0.0113599,
      "auxiliary_loss_mlp": 0.01029686,
      "balance_loss_clip": 1.04614258,
      "balance_loss_mlp": 1.02268541,
      "epoch": 0.9959718631635904,
      "flos": 27782900974080.0,
      "grad_norm": 2.0919306032052023,
      "language_loss": 0.69284832,
      "learning_rate": 1.651981619819054e-10,
      "loss": 0.71450502,
      "num_input_tokens_seen": 178784060,
      "step": 8283,
      "time_per_iteration": 2.5752103328704834
    },
    {
      "auxiliary_loss_clip": 0.01109819,
      "auxiliary_loss_mlp": 0.01025017,
      "balance_loss_clip": 1.04335594,
      "balance_loss_mlp": 1.01850259,
      "epoch": 0.9960921060542296,
      "flos": 24024095274240.0,
      "grad_norm": 2.5778428664027273,
      "language_loss": 0.7088697,
      "learning_rate": 1.5533797692546257e-10,
      "loss": 0.73021805,
      "num_input_tokens_seen": 178802795,
      "step": 8284,
      "time_per_iteration": 2.6088147163391113
    },
    {
      "auxiliary_loss_clip": 0.01145449,
      "auxiliary_loss_mlp": 0.01024066,
      "balance_loss_clip": 1.04370773,
      "balance_loss_mlp": 1.01657939,
      "epoch": 0.9962123489448687,
      "flos": 18697393935360.0,
      "grad_norm": 2.075797015525156,
      "language_loss": 0.84501386,
      "learning_rate": 1.4578116671404296e-10,
      "loss": 0.86670899,
      "num_input_tokens_seen": 178821075,
      "step": 8285,
      "time_per_iteration": 2.48188853263855
    },
    {
      "auxiliary_loss_clip": 0.01144405,
      "auxiliary_loss_mlp": 0.01029843,
      "balance_loss_clip": 1.04705262,
      "balance_loss_mlp": 1.02253258,
      "epoch": 0.9963325918355077,
      "flos": 20010754823040.0,
      "grad_norm": 1.8877487381632618,
      "language_loss": 0.71374524,
      "learning_rate": 1.3652773279759777e-10,
      "loss": 0.73548776,
      "num_input_tokens_seen": 178837725,
      "step": 8286,
      "time_per_iteration": 2.488166332244873
    },
    {
      "auxiliary_loss_clip": 0.01149019,
      "auxiliary_loss_mlp": 0.01028515,
      "balance_loss_clip": 1.04534185,
      "balance_loss_mlp": 1.02081084,
      "epoch": 0.9964528347261468,
      "flos": 33108488991360.0,
      "grad_norm": 1.7315404951726763,
      "language_loss": 0.6282227,
      "learning_rate": 1.2757767657989305e-10,
      "loss": 0.64999801,
      "num_input_tokens_seen": 178861515,
      "step": 8287,
      "time_per_iteration": 3.3093338012695312
    },
    {
      "auxiliary_loss_clip": 0.0114846,
      "auxiliary_loss_mlp": 0.01021393,
      "balance_loss_clip": 1.04671752,
      "balance_loss_mlp": 1.01463962,
      "epoch": 0.9965730776167859,
      "flos": 23109342589440.0,
      "grad_norm": 2.1406674432266914,
      "language_loss": 0.87206125,
      "learning_rate": 1.1893099941850948e-10,
      "loss": 0.89375979,
      "num_input_tokens_seen": 178880410,
      "step": 8288,
      "time_per_iteration": 2.5149528980255127
    },
    {
      "auxiliary_loss_clip": 0.01136835,
      "auxiliary_loss_mlp": 0.01026158,
      "balance_loss_clip": 1.04242706,
      "balance_loss_mlp": 1.01902056,
      "epoch": 0.996693320507425,
      "flos": 22965843755520.0,
      "grad_norm": 2.098934167889525,
      "language_loss": 0.77253771,
      "learning_rate": 1.105877026252866e-10,
      "loss": 0.79416764,
      "num_input_tokens_seen": 178898740,
      "step": 8289,
      "time_per_iteration": 2.54060697555542
    },
    {
      "auxiliary_loss_clip": 0.01162864,
      "auxiliary_loss_mlp": 0.01026906,
      "balance_loss_clip": 1.04515636,
      "balance_loss_mlp": 1.01981258,
      "epoch": 0.996813563398064,
      "flos": 13222740476160.0,
      "grad_norm": 1.867658553150117,
      "language_loss": 0.72035944,
      "learning_rate": 1.0254778746565663e-10,
      "loss": 0.74225712,
      "num_input_tokens_seen": 178914015,
      "step": 8290,
      "time_per_iteration": 2.4171926975250244
    },
    {
      "auxiliary_loss_clip": 0.01120771,
      "auxiliary_loss_mlp": 0.0102802,
      "balance_loss_clip": 1.04482841,
      "balance_loss_mlp": 1.02209842,
      "epoch": 0.9969338062887032,
      "flos": 14647855553280.0,
      "grad_norm": 1.9489265742607471,
      "language_loss": 0.73119605,
      "learning_rate": 9.481125515953259e-11,
      "loss": 0.75268394,
      "num_input_tokens_seen": 178932075,
      "step": 8291,
      "time_per_iteration": 2.538266897201538
    },
    {
      "auxiliary_loss_clip": 0.01107751,
      "auxiliary_loss_mlp": 0.0102401,
      "balance_loss_clip": 1.03910947,
      "balance_loss_mlp": 1.01651525,
      "epoch": 0.9970540491793423,
      "flos": 25735741142400.0,
      "grad_norm": 1.8025186974646996,
      "language_loss": 0.79609084,
      "learning_rate": 8.737810688064228e-11,
      "loss": 0.81740844,
      "num_input_tokens_seen": 178951910,
      "step": 8292,
      "time_per_iteration": 2.6331288814544678
    },
    {
      "auxiliary_loss_clip": 0.01113174,
      "auxiliary_loss_mlp": 0.01026012,
      "balance_loss_clip": 1.04163408,
      "balance_loss_mlp": 1.01823914,
      "epoch": 0.9971742920699813,
      "flos": 21470236237440.0,
      "grad_norm": 1.9234265745609482,
      "language_loss": 0.78945726,
      "learning_rate": 8.024834375608414e-11,
      "loss": 0.81084907,
      "num_input_tokens_seen": 178970500,
      "step": 8293,
      "time_per_iteration": 2.5736336708068848
    },
    {
      "auxiliary_loss_clip": 0.01054546,
      "auxiliary_loss_mlp": 0.01002371,
      "balance_loss_clip": 1.00784206,
      "balance_loss_mlp": 1.00128639,
      "epoch": 0.9972945349606205,
      "flos": 72211223629440.0,
      "grad_norm": 0.8171875339577178,
      "language_loss": 0.62877083,
      "learning_rate": 7.342196686788149e-11,
      "loss": 0.64934003,
      "num_input_tokens_seen": 179023665,
      "step": 8294,
      "time_per_iteration": 2.942809581756592
    },
    {
      "auxiliary_loss_clip": 0.01130575,
      "auxiliary_loss_mlp": 0.01027029,
      "balance_loss_clip": 1.04650044,
      "balance_loss_mlp": 1.02004004,
      "epoch": 0.9974147778512595,
      "flos": 19678293515520.0,
      "grad_norm": 2.07106461735359,
      "language_loss": 0.68687475,
      "learning_rate": 6.689897725142834e-11,
      "loss": 0.70845073,
      "num_input_tokens_seen": 179043140,
      "step": 8295,
      "time_per_iteration": 2.5541605949401855
    },
    {
      "auxiliary_loss_clip": 0.01136147,
      "auxiliary_loss_mlp": 0.01024277,
      "balance_loss_clip": 1.04398108,
      "balance_loss_mlp": 1.01738095,
      "epoch": 0.9975350207418986,
      "flos": 15960821391360.0,
      "grad_norm": 2.086324605457267,
      "language_loss": 0.88386971,
      "learning_rate": 6.067937589615545e-11,
      "loss": 0.90547401,
      "num_input_tokens_seen": 179061215,
      "step": 8296,
      "time_per_iteration": 2.5192599296569824
    },
    {
      "auxiliary_loss_clip": 0.01038721,
      "auxiliary_loss_mlp": 0.0100272,
      "balance_loss_clip": 1.00969887,
      "balance_loss_mlp": 1.00160491,
      "epoch": 0.9976552636325378,
      "flos": 59961879768960.0,
      "grad_norm": 0.7488544704656532,
      "language_loss": 0.57735109,
      "learning_rate": 5.476316374575241e-11,
      "loss": 0.59776545,
      "num_input_tokens_seen": 179124700,
      "step": 8297,
      "time_per_iteration": 3.1339709758758545
    },
    {
      "auxiliary_loss_clip": 0.0116467,
      "auxiliary_loss_mlp": 0.01035364,
      "balance_loss_clip": 1.04674149,
      "balance_loss_mlp": 1.02737725,
      "epoch": 0.9977755065231768,
      "flos": 22487872452480.0,
      "grad_norm": 1.853087203551493,
      "language_loss": 0.7226609,
      "learning_rate": 4.9150341697723476e-11,
      "loss": 0.74466121,
      "num_input_tokens_seen": 179144590,
      "step": 8298,
      "time_per_iteration": 2.4925825595855713
    },
    {
      "auxiliary_loss_clip": 0.01133317,
      "auxiliary_loss_mlp": 0.01029486,
      "balance_loss_clip": 1.04566193,
      "balance_loss_mlp": 1.0222497,
      "epoch": 0.9978957494138159,
      "flos": 26030280666240.0,
      "grad_norm": 1.4798195646132308,
      "language_loss": 0.66430199,
      "learning_rate": 4.384091060338768e-11,
      "loss": 0.68593001,
      "num_input_tokens_seen": 179165060,
      "step": 8299,
      "time_per_iteration": 3.442920446395874
    },
    {
      "auxiliary_loss_clip": 0.01148794,
      "auxiliary_loss_mlp": 0.01026982,
      "balance_loss_clip": 1.04579365,
      "balance_loss_mlp": 1.02058363,
      "epoch": 0.998015992304455,
      "flos": 22637835734400.0,
      "grad_norm": 2.0628464849608847,
      "language_loss": 0.73467684,
      "learning_rate": 3.883487126810081e-11,
      "loss": 0.75643456,
      "num_input_tokens_seen": 179184320,
      "step": 8300,
      "time_per_iteration": 3.2611000537872314
    },
    {
      "auxiliary_loss_clip": 0.01138198,
      "auxiliary_loss_mlp": 0.01022379,
      "balance_loss_clip": 1.04244947,
      "balance_loss_mlp": 1.01577473,
      "epoch": 0.9981362351950941,
      "flos": 18223444955520.0,
      "grad_norm": 1.8141399599732482,
      "language_loss": 0.79249209,
      "learning_rate": 3.41322244516995e-11,
      "loss": 0.81409782,
      "num_input_tokens_seen": 179202265,
      "step": 8301,
      "time_per_iteration": 3.275376558303833
    },
    {
      "auxiliary_loss_clip": 0.01094413,
      "auxiliary_loss_mlp": 0.01021061,
      "balance_loss_clip": 1.04030144,
      "balance_loss_mlp": 1.01451349,
      "epoch": 0.9982564780857331,
      "flos": 33474095573760.0,
      "grad_norm": 1.7001666641182565,
      "language_loss": 0.62839693,
      "learning_rate": 2.9732970866946925e-11,
      "loss": 0.64955175,
      "num_input_tokens_seen": 179222145,
      "step": 8302,
      "time_per_iteration": 2.6849722862243652
    },
    {
      "auxiliary_loss_clip": 0.0110661,
      "auxiliary_loss_mlp": 0.01029374,
      "balance_loss_clip": 1.03821993,
      "balance_loss_mlp": 1.02132726,
      "epoch": 0.9983767209763723,
      "flos": 15523465392000.0,
      "grad_norm": 2.214282309943189,
      "language_loss": 0.78438669,
      "learning_rate": 2.563711118175327e-11,
      "loss": 0.80574656,
      "num_input_tokens_seen": 179239030,
      "step": 8303,
      "time_per_iteration": 2.5452687740325928
    },
    {
      "auxiliary_loss_clip": 0.01115443,
      "auxiliary_loss_mlp": 0.01024877,
      "balance_loss_clip": 1.04265857,
      "balance_loss_mlp": 1.01854646,
      "epoch": 0.9984969638670114,
      "flos": 19974377324160.0,
      "grad_norm": 1.7804971820713567,
      "language_loss": 0.8384949,
      "learning_rate": 2.184464601717728e-11,
      "loss": 0.85989809,
      "num_input_tokens_seen": 179257345,
      "step": 8304,
      "time_per_iteration": 2.5694355964660645
    },
    {
      "auxiliary_loss_clip": 0.01154059,
      "auxiliary_loss_mlp": 0.01023873,
      "balance_loss_clip": 1.04924834,
      "balance_loss_mlp": 1.01661325,
      "epoch": 0.9986172067576504,
      "flos": 20375750874240.0,
      "grad_norm": 2.1957517949029235,
      "language_loss": 0.76946533,
      "learning_rate": 1.8355575948758585e-11,
      "loss": 0.79124469,
      "num_input_tokens_seen": 179275330,
      "step": 8305,
      "time_per_iteration": 2.4894847869873047
    },
    {
      "auxiliary_loss_clip": 0.011322,
      "auxiliary_loss_mlp": 0.01025451,
      "balance_loss_clip": 1.04084325,
      "balance_loss_mlp": 1.01835752,
      "epoch": 0.9987374496482896,
      "flos": 23727903724800.0,
      "grad_norm": 2.5588275736766932,
      "language_loss": 0.73392224,
      "learning_rate": 1.5169901505407424e-11,
      "loss": 0.75549883,
      "num_input_tokens_seen": 179292395,
      "step": 8306,
      "time_per_iteration": 2.5369577407836914
    },
    {
      "auxiliary_loss_clip": 0.01132256,
      "auxiliary_loss_mlp": 0.01027353,
      "balance_loss_clip": 1.04385328,
      "balance_loss_mlp": 1.02048659,
      "epoch": 0.9988576925389286,
      "flos": 25044029959680.0,
      "grad_norm": 1.7283166706903617,
      "language_loss": 0.74044228,
      "learning_rate": 1.228762317073695e-11,
      "loss": 0.76203841,
      "num_input_tokens_seen": 179311225,
      "step": 8307,
      "time_per_iteration": 2.557877779006958
    },
    {
      "auxiliary_loss_clip": 0.01132076,
      "auxiliary_loss_mlp": 0.01022001,
      "balance_loss_clip": 1.04345918,
      "balance_loss_mlp": 1.01535249,
      "epoch": 0.9989779354295677,
      "flos": 31285627637760.0,
      "grad_norm": 1.9077190797309398,
      "language_loss": 0.78826797,
      "learning_rate": 9.70874138195299e-12,
      "loss": 0.80980873,
      "num_input_tokens_seen": 179333135,
      "step": 8308,
      "time_per_iteration": 2.6038830280303955
    },
    {
      "auxiliary_loss_clip": 0.01163246,
      "auxiliary_loss_mlp": 0.01030927,
      "balance_loss_clip": 1.04632676,
      "balance_loss_mlp": 1.02279723,
      "epoch": 0.9990981783202069,
      "flos": 19573398823680.0,
      "grad_norm": 1.4680620295247162,
      "language_loss": 0.74751639,
      "learning_rate": 7.433256530076093e-12,
      "loss": 0.76945812,
      "num_input_tokens_seen": 179353090,
      "step": 8309,
      "time_per_iteration": 2.449739456176758
    },
    {
      "auxiliary_loss_clip": 0.01110433,
      "auxiliary_loss_mlp": 0.01022359,
      "balance_loss_clip": 1.0399704,
      "balance_loss_mlp": 1.01621079,
      "epoch": 0.9992184212108459,
      "flos": 17199667514880.0,
      "grad_norm": 2.1119011294751187,
      "language_loss": 0.75963062,
      "learning_rate": 5.46116896038562e-12,
      "loss": 0.78095847,
      "num_input_tokens_seen": 179367500,
      "step": 8310,
      "time_per_iteration": 2.5589561462402344
    },
    {
      "auxiliary_loss_clip": 0.01129165,
      "auxiliary_loss_mlp": 0.01028657,
      "balance_loss_clip": 1.04262745,
      "balance_loss_mlp": 1.02138269,
      "epoch": 0.999338664101485,
      "flos": 46497853681920.0,
      "grad_norm": 2.219431837669009,
      "language_loss": 0.61769891,
      "learning_rate": 3.792478972197699e-12,
      "loss": 0.6392771,
      "num_input_tokens_seen": 179388085,
      "step": 8311,
      "time_per_iteration": 2.715505361557007
    },
    {
      "auxiliary_loss_clip": 0.01161415,
      "auxiliary_loss_mlp": 0.0102128,
      "balance_loss_clip": 1.04585695,
      "balance_loss_mlp": 1.01499796,
      "epoch": 0.9994589069921241,
      "flos": 15158253859200.0,
      "grad_norm": 2.128951955632126,
      "language_loss": 0.6996817,
      "learning_rate": 2.4271868181990895e-12,
      "loss": 0.72150862,
      "num_input_tokens_seen": 179405250,
      "step": 8312,
      "time_per_iteration": 2.4336795806884766
    },
    {
      "auxiliary_loss_clip": 0.01149744,
      "auxiliary_loss_mlp": 0.01022448,
      "balance_loss_clip": 1.0444603,
      "balance_loss_mlp": 1.01545048,
      "epoch": 0.9995791498827632,
      "flos": 12531460256640.0,
      "grad_norm": 2.170793311852865,
      "language_loss": 0.81140238,
      "learning_rate": 1.3652927060014973e-12,
      "loss": 0.83312428,
      "num_input_tokens_seen": 179420845,
      "step": 8313,
      "time_per_iteration": 3.2025845050811768
    },
    {
      "auxiliary_loss_clip": 0.01120791,
      "auxiliary_loss_mlp": 0.01026921,
      "balance_loss_clip": 1.04264176,
      "balance_loss_mlp": 1.01942301,
      "epoch": 0.9996993927734023,
      "flos": 19245175320960.0,
      "grad_norm": 1.8571373452598983,
      "language_loss": 0.63931102,
      "learning_rate": 6.067967965872612e-13,
      "loss": 0.66078806,
      "num_input_tokens_seen": 179440455,
      "step": 8314,
      "time_per_iteration": 2.5529634952545166
    },
    {
      "auxiliary_loss_clip": 0.01119658,
      "auxiliary_loss_mlp": 0.01036136,
      "balance_loss_clip": 1.04478836,
      "balance_loss_mlp": 1.02894199,
      "epoch": 0.9998196356640414,
      "flos": 62952804518400.0,
      "grad_norm": 1.5538371306867649,
      "language_loss": 0.7709533,
      "learning_rate": 1.5169920497548615e-13,
      "loss": 0.79251122,
      "num_input_tokens_seen": 179465075,
      "step": 8315,
      "time_per_iteration": 2.9255104064941406
    },
    {
      "auxiliary_loss_clip": 0.01102091,
      "auxiliary_loss_mlp": 0.01017494,
      "balance_loss_clip": 1.02693558,
      "balance_loss_mlp": 1.0130868,
      "epoch": 0.9999398785546805,
      "flos": 50922375073920.0,
      "grad_norm": 1.1533668498728562,
      "language_loss": 0.54982793,
      "learning_rate": 0.0,
      "loss": 0.57102382,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "time_per_iteration": 3.0551352500915527
    },
    {
      "epoch": 0.9999398785546805,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "total_flos": 6.996749092776837e+17,
      "train_loss": 0.7889158849090611,
      "train_runtime": 23734.1053,
      "train_samples_per_second": 14.016,
      "train_steps_per_second": 0.35
    }
  ],
  "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
}